Internet corpus information expansion method, related device and medium

Abstract

The invention provides an internet corpus information expansion method, a related device and a medium. The method comprises the steps that a target internet corpus is displayed, the target internet corpus has a trigger word, and the trigger word is displayed in a display mode different from other parts of the target internet corpus; in response to first triggering of the triggering word, a first information extension page is displayed, the first information extension page is provided with a first area and a second area, the first area comprises first interpretation information of the triggering word, and the second area comprises the triggering word and a plurality of to-be-triggered labels; and in response to a second trigger for a target tag in the plurality of to-be-triggered tags, displaying a second information extension page, the second information extension page being associated with the target tag. According to the embodiment of the invention, the associated information corresponding to the corpus information can be further displayed, more effective information can be quickly displayed, and the associated information display efficiency and the displayed information effectiveness are improved.

Description

Technical Field

[0001] The present disclosure relates to the field of Internet technologies, and in particular, to an Internet corpus information expansion method, related apparatus, and medium. Background Art

[0002] In the prior art, when an object browses corpus information on the Internet (such as viewing search results), it is often necessary to click into a corpus information to view the details of the corpus (such as viewing a search result). The detailed information of the corpus is generally presented on the landing page of a single piece of information. In this way, the content presented on the landing page will be limited to the detailed information of the event, and it is difficult for the object to understand the background and knowledge related to the corpus, resulting in low efficiency of associated information display and low effectiveness of information display. Summary of the Invention

[0003] Embodiments of the present disclosure provide an Internet corpus information expansion method, related apparatus, and medium, which can further display the associated information corresponding to the corpus information, quickly display more effective information, and improve the efficiency of associated information display and the effectiveness of displayed information.

[0004] According to one aspect of the present disclosure, there is provided an Internet corpus information expansion method, including:

[0005] Display a target Internet corpus, where the target Internet corpus has a trigger word, and the trigger word is displayed in a display manner different from other parts of the target Internet corpus;

[0006] In response to a first trigger on the trigger word, display a first information expansion page, where the first information expansion page has a first area and a second area, the first area contains first explanation information about the trigger word, and the second area contains the trigger word and a plurality of pending trigger tags;

[0007] In response to a second trigger on a target tag among the plurality of pending trigger tags, display a second information expansion page, where the second information expansion page is associated with the target tag.

[0008] According to one aspect of the present disclosure, there is provided an Internet corpus information expansion apparatus, including:

[0009] A first display unit for displaying a target Internet corpus, where the target Internet corpus has a trigger word, and the trigger word is displayed in a display manner different from other parts of the target Internet corpus;

[0010] A second display unit, configured to display a first information expansion page in response to a first trigger on the trigger word, where the first information expansion page has a first area and a second area, the first area contains first explanatory information about the trigger word, and the second area contains the trigger word and a plurality of to-be-triggered tags;

[0011] A third display unit, configured to display a second information expansion page in response to a second trigger on a target tag among the plurality of to-be-triggered tags, where the second information expansion page is associated with the target tag.

[0012] Optionally, the third display unit is specifically configured to:

[0013] If the display mode of the target tag in the second area is the same as that of the trigger word, in a third area of the second information expansion page, display the common associated information of the target tag and the trigger word;

[0014] In a fourth area of the second information expansion page, display the plurality of to-be-triggered tags in the second area.

[0015] Optionally, the third display unit is specifically configured to:

[0016] Determine a first guiding text according to the types of the target tag and the trigger word;

[0017] Input the target tag, the trigger word, and the first guiding text into a large language model to obtain first detailed information about the common associated information;

[0018] Input the first detailed information into a Hunyuan model to obtain a first summary information about the common associated information;

[0019] In the third area, display the first summary information and the first detailed information as the common associated information.

[0020] Optionally, the third display unit is specifically configured to:

[0021] If the types of both the target tag and the trigger word are people or items, display the common characteristics of the people or items as the common associated information;

[0022] If the types of both the target tag and the trigger word are events, display the relationship between the events as the common associated information;

[0023] If one of the target tag and the trigger word is of the type of people or items and the other is of the type of event, display the role of the people or the item in the event as the common associated information.

[0024] Optionally, the third display unit is specifically configured to:

[0025] If the display mode of the target label in the second area is different from that of the trigger word, display the second explanation information of the target label in the third area of the second information expansion page;

[0026] In the fourth area of the second information expansion page, display the updated multiple pending trigger labels.

[0027] Optionally, the third display unit is specifically configured to:

[0028] Input the target label plus the second guiding text into a large language model to obtain the second detailed information of the target label;

[0029] Input the second detailed information into the Hunyuan model to obtain the second summary information of the target label;

[0030] In the third area, display the second summary information and the second detailed information as the second explanation information.

[0031] Optionally, the third display unit is specifically configured to:

[0032] For each candidate label in the candidate label set, obtain the first distance between the candidate label and the target label in the knowledge graph;

[0033] For each candidate label, obtain the first popularity of the candidate label;

[0034] Based on the first distance and the first popularity, select the updated pending trigger labels from the candidate labels that have not been selected in the candidate label set.

[0035] Optionally, the third display unit is specifically configured to:

[0036] Based on the first distance, select the first number of pending trigger labels from the candidate labels that have not been selected in the candidate label set;

[0037] Based on the first popularity, select the remaining pending trigger labels from the candidate labels that have not been selected in the candidate label set.

[0038] Optionally, the first information expansion page has an update control for the multiple pending trigger labels; the Internet corpus information expansion device further includes:

[0039] A first update unit, configured to update the multiple pending trigger labels in response to a third trigger on the update control.

[0040] Optionally, the multiple to-be-triggered tags are divided into multiple groups, where the display modes of the to-be-triggered tags in each group are the same and the numbers are balanced;

[0041] The first update unit is specifically configured to:

[0042] Update the to-be-triggered tags in each group so that the number of the updated to-be-triggered tags in each group is the same as the number of the to-be-triggered tags before the update.

[0043] Optionally, the first update unit is specifically configured to:

[0044] For the to-be-triggered tags in each group, determine the number of to-be-updated tags in the group based on the number of to-be-triggered tags in the group and a second ratio;

[0045] For each candidate tag in the candidate tag set, obtain a first distance between the candidate tag and the target tag in the knowledge graph;

[0046] For each candidate tag, obtain a first popularity of the candidate tag;

[0047] Based on the first distance and the first popularity, select the number of candidate tags equal to the number of to-be-updated tags from the unselected candidate tags in the candidate tag set for updating the to-be-triggered tags in the group.

[0048] Optionally, the Internet corpus information expansion device further includes:

[0049] A second update unit, configured to update the multiple to-be-triggered tags in response to a sliding operation on the second area.

[0050] Optionally, the sliding operation is a sliding operation along the screen width direction of the interface for displaying the second area;

[0051] The second update unit is specifically configured to:

[0052] Update a second number of the to-be-triggered tags among the multiple to-be-triggered tags, where the second number is associated with a ratio of a sliding path length of the sliding operation to the screen width.

[0053] Optionally, the second update unit is specifically configured to:

[0054] Determine the second number based on the ratio of the sliding path length of the sliding operation to the screen width;

[0055] For each candidate tag in the candidate tag set, obtain a first distance between the candidate tag and the target tag in the knowledge graph;

[0056] For each of the candidate tags, obtain the first popularity of the candidate tag;

[0057] Based on the first distance and the first popularity, select the second number of candidate tags from the candidate tags that have not been selected in the candidate tag set, and replace the second number of to-be-triggered tags among the multiple to-be-triggered tags.

[0058] Optionally, the first information expansion page further includes a fifth area for displaying the target tag sequence historically triggered by the target object that performs the first trigger. Among the target tag sequences, the target tag most recently triggered by the target object is displayed in a first display manner different from the display manners of other target tags in the target tag training;

[0059] The Internet corpus information expansion device further includes:

[0060] A sliding operation unit for performing a sliding operation on the target tag sequence, so that in the fifth area, the target tags in the target tag sequence are sequentially displayed in the first display manner;

[0061] A fourth display unit for displaying the second information expansion page corresponding to the target tag displayed in the first display manner in the fifth area.

[0062] Optionally, the fourth display unit is specifically configured to:

[0063] Obtain a browsing history database that stores the second information expansion page corresponding to each target tag in the target tag sequence;

[0064] Obtain and display the second information expansion page corresponding to the target tag displayed in the first display manner in the fifth area from the browsing history database.

[0065] Optionally, the target Internet corpus is an Internet search result;

[0066] The first display unit is specifically configured to:

[0067] Obtain an Internet search request;

[0068] Display a plurality of Internet search results corresponding to the Internet search request, and the Internet search results are sorted according to the number of trigger words included.

[0069] Optionally, the first display unit is specifically configured to:

[0070] Obtain the target Internet corpus;

[0071] Tokenize the target Internet corpus and determine the matching words among the tokens that match the candidate tags in the candidate tag set;

[0072] Obtain the second popularity of the matching words;

[0073] Determine the trigger words among the matching words based on the second popularity;

[0074] Display the target Internet corpus and display the trigger words in a display manner different from other parts of the target Internet corpus.

[0075] Optionally, the first display unit is specifically configured to:

[0076] Determine the first occurrence count of the matching words being searched by an object from Internet search records;

[0077] Determine the second occurrence count of the matching words in the search results of the Internet search records;

[0078] Determine the second popularity of the matching words based on the first occurrence count and the second occurrence count.

[0079] Optionally, the second display unit is specifically configured to:

[0080] In response to a first trigger on the trigger word, input the trigger word plus a third guiding text into a large language model to obtain third detail information of the trigger word;

[0081] Input the third detail information into a Hunyuan model to obtain third summary information of the trigger word;

[0082] In the first area, display the third summary information and the third detail information as the first explanation information.

[0083] Optionally, the Internet corpus information expansion device further includes:

[0084] A to-be-triggered tag selection unit, configured to select multiple to-be-triggered tags from the candidate tag set by the following method:

[0085] For each candidate tag in the candidate tag set, obtain the first distance between the candidate tag and the trigger word in the knowledge graph;

[0086] For each candidate tag, obtain the first popularity of the candidate tag;

[0087] Select the to-be-triggered tags from the candidate tag set based on the first distance and the first popularity.

[0088] Optionally, the to-be-triggered tag selection unit is specifically configured to:

[0089] Based on the first distance, select a first number of the to-be-triggered tags from the candidate tag set, where the first number is obtained based on the number of the to-be-triggered tags and a first ratio;

[0090] Based on the first popularity, select the remaining to-be-triggered tags among the multiple to-be-triggered tags from the unselected candidate tags in the candidate tag set.

[0091] Optionally, the to-be-triggered tag selection unit is specifically configured to:

[0092] Based on the first distance, determine a first score of the to-be-triggered tag;

[0093] Based on the first popularity, determine a second score of the to-be-triggered tag;

[0094] Based on the first score and the second score, determine a total score of the to-be-triggered tag;

[0095] Based on the total score, select the to-be-triggered tag from the candidate tag set.

[0096] Optionally, the candidate tag set includes the candidate tags and the display modes corresponding to the candidate tags;

[0097] The Internet corpus information expansion device further includes:

[0098] A candidate tag set generation unit, configured to generate the candidate tag set in the following manner:

[0099] Obtain a first corpus sample library;

[0100] Segment the first corpus sample library, and use each of the segmented words as the candidate tag and add it to the candidate tag library;

[0101] Based on at least one of a knowledge graph, object behavior data of a corpus sample having the candidate tag in the first corpus sample library, and a third popularity of the corpus sample, determine the display mode of the candidate tag, and store it corresponding to the candidate tag in the candidate tag set;

[0102] Displaying the first information expansion page includes: displaying the to-be-triggered tag in the second area in the display mode corresponding to the candidate tag in the candidate tag set.

[0103] Optionally, the candidate tag set generation unit is specifically configured to:

[0104] For any two candidate tags in the candidate tag set, obtain the target corpus in the first corpus sample library that contains both candidate tags;

[0105] If the target corpus has been accessed by the target object that has triggered the first trigger, the third popularity of the target corpus is greater than the first threshold, and the first distance between the two candidate tags in the knowledge graph is less than the second threshold, then assign the same display method to the two candidate tags.

[0106] Optionally, the candidate tag set generation unit is specifically configured to:

[0107] For any two candidate tags in the candidate tag set, obtain the relationship between the two candidate tags in the knowledge graph;

[0108] Based on the type of the relationship, determine the display method of the two candidate tags.

[0109] Optionally, the candidate tag set generation unit is specifically configured to:

[0110] For any two candidate tags in the candidate tag set, if the two candidate tags are co-associated with the same event in the knowledge graph, the time proximity in the two corpora with the two candidate tags in the first corpus sample library meets the first condition, and the two candidate tags have the same type, then assign the same display method to the two candidate tags.

[0111] Optionally, the candidate tag set generation unit is specifically configured to:

[0112] Obtain the candidate tag number threshold for each display method;

[0113] If, based on at least one of the knowledge graph, the object behavior data of the corpus sample with the candidate tag in the first corpus sample library, and the third popularity of the corpus sample, after assigning the display method to the candidate tag, the number of candidate tags of the display method exceeds the candidate tag number threshold, cancel the display method assigned to the candidate tag.

[0114] Optionally, the first display unit is specifically configured to:

[0115] Determine the first distance between the to-be-triggered tag and the trigger word in the knowledge graph;

[0116] Based on the first distance, display multiple to-be-triggered tags such that the second distance between the to-be-triggered tag and the trigger word in the second region is associated with the first distance.

[0117] According to one aspect of the present disclosure, there is provided an electronic device, including a memory and a processor, where the memory stores a computer program, and when the processor executes the computer program, the above-mentioned method for expanding Internet corpus information is implemented.

[0118] According to one aspect of the present disclosure, there is provided a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the above-mentioned method for expanding Internet corpus information is implemented.

[0119] According to one aspect of the present disclosure, there is provided a computer program product including a computer program, where the computer program is read and executed by a processor of a computer device, so that the computer device executes the above-mentioned method for expanding Internet corpus information.

[0120] In an embodiment of the present disclosure, the target Internet corpus has a trigger word, and when the target Internet corpus is displayed, the trigger word is displayed in a display manner different from other parts of the target Internet corpus. In response to a first trigger on the trigger word, a first information expansion page is displayed. The first information expansion page has a first area and a second area, where the first area contains first explanation information about the trigger word, and the second area contains the trigger word and a plurality of tags to be triggered. Further, in response to a second trigger on a target tag among the plurality of tags to be triggered, a second information expansion page is displayed, and the second information expansion page is associated with the target tag. In this way, through the second trigger on the tag to be triggered, the second information expansion page can be used to further display the corresponding associated information of the corpus information, so as to quickly display more effective information and improve the display efficiency of the associated information and the effectiveness of the displayed information.

[0121] Other features and advantages of the present disclosure will be described in the following description, and some will be obvious from the description, or will be understood by implementing the present disclosure. The objectives and other advantages of the present disclosure can be achieved and obtained through the structures specifically pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0122] The drawings are used to provide a further understanding of the technical solutions of the present disclosure, and constitute a part of the description. They are used together with the embodiments of the present disclosure to explain the technical solutions of the present disclosure, and do not constitute a limitation to the technical solutions of the present disclosure.

[0123] Figure 1 is a block diagram of a system to which the method for expanding Internet corpus information according to an embodiment of the present disclosure is applied;

[0124] Figures 2A - 2E is a schematic diagram of an interface in the scenario where the embodiment of the present disclosure is applied to obtain expanded corpus information during the process of browsing web page information.

[0125] Figure 3 is a flowchart of an Internet corpus information expansion method according to an embodiment of the present disclosure;

[0126] Figure 3A is Figure 3 an example display interface diagram of step 310 in

[0127] Figure 3B is Figure 3 an example display interface diagram of step 320 in

[0128] Figure 3C is Figure 3 an example display interface diagram of step 330 in

[0129] Figure 4 is Figure 3 a flowchart showing the Internet search results in step 310;

[0130] Figures 5A - 5B is in accordance with Figure 4 an example interface diagram showing the Internet search results;

[0131] Figure 6 is Figure 3 another flowchart showing the Internet search results in step 310;

[0132] Figures 7A - 7C is a schematic diagram of a specific implementation showing the Internet search results;

[0133] Figure 8 is Figure 6 a flowchart for obtaining the second popularity of matching words in step 630;

[0134] Figure 9 is a flowchart for generating the first explanatory information in the first information expansion page according to an embodiment of the present disclosure;

[0135] Figure 10 is a specific implementation example diagram for generating the first explanatory information in the first information expansion page according to an embodiment of the present disclosure;

[0136] Figure 11 is a flowchart for selecting multiple tags to be triggered from a candidate tag set according to an embodiment of the present disclosure;

[0137] Figures 12A - 12B is a specific implementation example diagram for selecting multiple tags to be triggered from a candidate tag set according to an embodiment of the present disclosure;

[0138] Figure 13 is Figure 11Flowchart of step 1130 in selecting the to-be-triggered tag from the candidate tag set;

[0139] Figure 14 Yes Figure 11 Another flowchart of step 1130 in selecting the to-be-triggered tag from the candidate tag set;

[0140] Figure 15 Flowchart of generating the candidate tag set before selecting multiple to-be-triggered tags from the candidate tag set according to an embodiment of the present disclosure;

[0141] Figures 16A - 16B According to Figure 15 A specific implementation example diagram of generating the candidate tag set before selecting multiple to-be-triggered tags from the candidate tag set;

[0142] Figure 17 Yes Figure 15 Flowchart of step 1530 in correspondingly storing the candidate tag in the candidate tag set;

[0143] Figure 18 Yes Figure 15 Another flowchart of step 1530 in correspondingly storing the candidate tag in the candidate tag set;

[0144] Figure 19 Yes Figure 15 Another flowchart of step 1530 in correspondingly storing the candidate tag in the candidate tag set;

[0145] Figure 20 Yes Figure 3 A flowchart of step 320 in displaying the first information expansion page;

[0146] Figures 21A to 21B A specific implementation example diagram of displaying the first information expansion page;

[0147] Figures 22A to 22B A specific implementation example diagram of displaying the first information expansion page;

[0148] Figure 23 Flowchart of updating the to-be-triggered tag in the first information expansion page according to an embodiment of the present disclosure;

[0149] Figures 24A to 24C A specific implementation example diagram of updating the to-be-triggered tag in the first information expansion page according to an embodiment of the present disclosure;

[0150] Figure 25 Another flowchart of updating the to-be-triggered tag in the first information expansion page according to an embodiment of the present disclosure;

[0151] Figure 26 Yes Figure 3Flowchart of step 330 in showing the second information expansion page;

[0152] Figures 27A to 27C It is in accordance with Figure 26 Specific implementation example diagram of showing the second information expansion page;

[0153] Figure 28 It is Figure 26 Flowchart of step 2610 in showing the common association information of the target label and the trigger word;

[0154] Figure 29 It is in accordance with Figure 28 Specific implementation example diagram of showing the common association information of the target label and the trigger word;

[0155] Figure 30 It is Figure 26 Another flowchart of step 2610 in showing the common association information of the target label and the trigger word;

[0156] Figure 31 It is Figure 3 Another flowchart of step 330 in showing the second information expansion page;

[0157] Figures 31A to 31B It is in accordance with Figure 31 Implementation example diagram of showing the second information expansion page;

[0158] Figure 32 It is Figure 31 Flowchart of step 3110 in showing the second interpretation information of the target label;

[0159] Figure 33 It is Figure 31 Flowchart of step 3120 in showing the updated multiple tags to be triggered;

[0160] Figure 34 It is Figure 33 Flowchart of step 3330 in selecting the updated tags to be triggered from the candidate tag set;

[0161] Figure 35 It is Figure 3 Flowchart of step 330 in showing the second information expansion page through the target label sequence;

[0162] Figures 36A - 36C It is Figure 3 Specific implementation example diagram of step 330 in showing the second information expansion page through the target label sequence;

[0163] Figure 37 It is Figure 35 Flowchart of obtaining and showing the second information expansion page from the browsing history database in step 3520;

[0164] Figure 38 Schematic block diagram of an Internet corpus information expansion device according to an embodiment of the present disclosure;

[0165] Figure 39 Execute according to an embodiment of the present disclosure Figure 3 Terminal structure diagram of the Internet corpus information expansion method shown;

[0166] Figure 40 Execute according to an embodiment of the present disclosure Figure 3 Server structure diagram of the Internet corpus information expansion method shown. Detailed implementation manners

[0167] In order to make the objectives, technical solutions and advantages of the present disclosure clearer, the present disclosure will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present disclosure, and are not used to limit the present disclosure.

[0168] Before further elaborating on the embodiments of the present disclosure, the nouns and terms involved in the embodiments of the present disclosure are described. The nouns and terms involved in the embodiments of the present disclosure are applicable to the following explanations:

[0169] Landing page: Also known as the landing page or lead page. In the Internet field, the lead page is the web page that is displayed to the user after a web page viewer clicks on the page display information or searches using a search engine. Generally, this page will display extended content related to the clicked page display information or search result link.

[0170] Associated information: Refers to the people, things, objects, time, location, etc. involved in a certain piece of corpus information, and the information obtained based on the commonality, intersection or extension with this content.

[0171] Trigger word: Mark the words in the Internet corpus that contain extensible knowledge to form trigger words. When a trigger word is triggered, the associated information extended from the trigger word can be provided to the object browsing the web page. The display method of the trigger word on the web page is different from that of other parts of the Internet corpus. For example, the trigger word can be highlighted by bold, underline, color, marking, etc.

[0172] Historical record navigation: Based on the trigger words clicked in the historical record, connect one or more aspects of the associated information in chronological order to form a relatively complete record system and present it to the object browsing the web page as navigation information on the page.

[0173] System architecture and scenario description applied in the embodiments of the present disclosure

[0174] Figure 1It is a system architecture diagram to which the Internet corpus information expansion method according to an embodiment of the present disclosure is applied. It includes a server 110, the Internet 120, an object terminal 130, etc.

[0175] The server 110 refers to a computer system that can provide Internet corpus information expansion services to the object terminal 130. Compared with the object terminal 130, the server 110 has higher requirements in terms of stability, security, performance, etc. The server 110 can be a high-performance computer in a network platform, a cluster of multiple high-performance computers, a part (such as a virtual machine) allocated from a high-performance computer, a combination of parts (such as virtual machines) allocated from multiple high-performance computers, etc. The server 110 can also communicate with the Internet 120 in a wired or wireless manner to exchange data.

[0176] The object terminal 130 is a device that provides corpus information expansion services to an object and provides Internet corpus information expansion services to the object by receiving the triggering operations of the object, such as the first trigger on a trigger word and the second trigger on a target label among multiple pending trigger labels. It includes various forms such as a desktop computer, a laptop computer, a PDA (Personal Digital Assistant), a mobile phone, a vehicle-mounted terminal, a home theater terminal, a dedicated terminal, etc. In addition, it can be a single device or a set composed of multiple devices. For example, multiple devices are connected through a local area network and share a display device to work together to form a terminal. The object terminal 130 can also communicate with the Internet 120 in a wired or wireless manner to exchange data.

[0177] Embodiments of the present disclosure can be applied in various scenarios, such as Figures 2A - 2E the scenario of obtaining expanded corpus information during the process of browsing web page information as shown, etc.

[0178] Such as Figure 2A As shown, when an object browsing a web page is in the process of browsing web page information, the object terminal can set various types of information pages such as "Event Interpretation", "Latest Progress", and "Public Opinion Analysis" for the context information of the xx event. Among them, when selecting to browse the information page of "Latest Progress", three information columns can be displayed on the interface of the object terminal.

[0179] The title of the information column at the lower part of the interface is displayed as "xxxx agrees to purchase an aircraft after xxx pilots complete Class A flight training", and the overview is displayed as "xxxx company claims that it will complete the purchase of Class A aircraft in three days, and at that time, the first batch of xxx pilots will complete flight training". Among them, "xxx pilots" and "Class A flight training" are both trigger words, so the corresponding "Get to know more" label can be displayed beside the information column in the middle of the interface.

[0180] The title of the information column in the middle of the interface is displayed as "Leakage of Cargo B during Transportation", and the summary is displayed as "When Cargo B passed through the section of xxxx during transportation, the storage warehouse leaked". Among them, "Cargo B" is the trigger word, so the corresponding "Gain Knowledge" label can be displayed beside the information column in the middle of the interface.

[0181] The title at the top of the interface is displayed as "The first batch of pilots start training to fly Class A aircraft", and the summary is displayed as "With multiple test flights, it means that Class A aircraft will be used as the main carrier for Cargo B in the future". Among them, "Class A aircraft" is the trigger word, so the corresponding "Gain Knowledge" label can be displayed beside the information column in the middle of the interface.

[0182] It should be understood that if there is no trigger word in the title of the information column, the "Gain Knowledge" label may not appear at the title of the information column. In addition, the trigger word is displayed in a different way from other parts of the target Internet corpus. Figure 2A The trigger word is displayed by bolding and underlining.

[0183] As Figure 2B shown, when the first trigger acts on Figure 2A the trigger word "Class A aircraft" in it, in response to the first trigger on "Class A aircraft", the first information expansion page is displayed. It should be noted that the first information expansion page has a first area and a second area.

[0184] The first area contains the first explanatory information of "Class A aircraft", and the first explanatory information is specifically displayed as "The fourth-generation transport aircraft, Class A aircraft is a transport aircraft developed by xxxx company; the advantages of Class A aircraft are high transport capacity, high mobility, short takeoff and landing ability, multi-purpose performance, advanced avionics system, good protection and reliability, and strong anti-leakage ability of the storage warehouse".

[0185] The second area contains "Class A aircraft" and multiple pending trigger labels. Among them, the words in the pending trigger labels can be words extracted from the associated information corresponding to the trigger word "Class A aircraft". For example, "Class A aircraft" and "F16 transport aircraft" both belong to the fourth-generation transport aircraft, which are words extracted from the same type of associated information corresponding to the trigger word "Class A aircraft"; "xxx route" is the route where the "Class A aircraft" had a transportation leakage, which is a word extracted from the associated information extended based on the trigger word "Class A aircraft"; "xxx group" is the group to which the "Class A aircraft" with transportation leakage belongs, which is a word extracted from the subordinate associated information based on the trigger word "Class A aircraft". It should be noted that the words in the pending trigger labels can be, but are not limited to, words extracted from the associated information corresponding to the trigger word "Class A aircraft", and the words in the pending trigger labels can also be words with relatively high current search popularity, so the types of words in the pending trigger labels are not limited to the above examples.

[0186] It should be noted that after the first information expansion page is displayed, the object browsing the web page can perform a second trigger on the target label "F16 transport aircraft" among multiple to-be-triggered labels.

[0187] It should be clear that the first information expansion page can also include an update control for updating multiple to-be-triggered labels within the second area of the first information expansion page.

[0188] As Figure 2C shown, after the object browsing the web page performs a second trigger on the target label "F16 transport aircraft" among multiple to-be-triggered labels, in response to the second trigger, the object terminal immediately further displays a second information expansion page. Among them, the second information expansion page is associated with the target label "F16 transport aircraft", and the second information expansion page can include a third area and a fourth area.

[0189] It should be pointed out that since "Type A aircraft" and "F16 transport aircraft" both belong to the fourth-generation transport aircraft, based on this associated information, the second explanatory information about "Type A aircraft" and "F16 transport aircraft" can be displayed in the third area of the second information expansion page, specifically shown as "Fourth-generation transport aircraft, both Type A aircraft and F16 transport aircraft are fourth-generation multi-purpose high-performance transport aircraft. They have advanced electronic systems, radars, cargo storage configurations, and flight performances. These two transport aircraft have many commonalities in flight performance and transport functions, but there are also certain differences, such as R & D backgrounds and actual application scenarios, etc.".

[0190] In addition, the fourth area of the second information expansion page can inherit the second area of the first information expansion page to display "Type A aircraft" and multiple to-be-triggered labels. It can also display "F16 transport aircraft" and multiple to-be-triggered labels. Among them, the words in the to-be-triggered labels can be words extracted from the associated information corresponding to the trigger word "Type A aircraft", or words extracted from the associated information corresponding to "F16 transport aircraft", or words extracted from the associated information that is related to both "Type A aircraft" and "F16 transport aircraft".

[0191] It should be clear that the second information expansion page can also include an update control for updating multiple to-be-triggered labels within the fourth area of the second information expansion page.

[0192] As Figure 2D shown, it should be noted that after the first information expansion page is displayed, the object browsing the web page can perform a second trigger on the target label "xx route transport efficiency" among multiple to-be-triggered labels.

[0193] As Figure 2EAs shown, after an object browsing a web page performs a second trigger on a target tag "xx route transportation efficiency" among multiple tags to be triggered, in response to the second trigger, the object terminal immediately further displays a second information expansion page. The second information expansion page is associated with the target tag "xx route transportation efficiency" and may include a third area and a fourth area.

[0194] It should be noted that since there is no direct association between "Type A aircraft" and "xx route transportation efficiency", "xx route transportation efficiency" is not direct associated information of "Type A aircraft". Therefore, based on the target tag "xx route transportation efficiency", the third area of the second information expansion page may display second explanatory information for the target tag "xx route transportation efficiency", specifically showing a schematic diagram of the recent monthly transportation efficiency of the xx route, and showing that "the transportation efficiency dropped to 0.0766 on June 23, and xxx Group is facing increasingly low transportation efficiency and decides to improve the transportation mode by purchasing Type A aircraft for operation".

[0195] In addition, the fourth area of the second information expansion page displays multiple tags to be triggered corresponding to "xx route transportation efficiency". Among them, the words in the tags to be triggered can be extracted from the associated information corresponding to the trigger word "xx route transportation efficiency". For example, "F22 transport aircraft" and "M142 transport aircraft" are also operated on the "xx route", so "F22 transport aircraft" and "M142 transport aircraft" can be used as the words in the tags to be triggered; the "xx route transportation efficiency" dropped to 0.0766 on June 23, resulting in "xx asset depreciation", so "xx asset depreciation" can be used as the word in the tag to be triggered. It should be understood that the types of tags to be triggered in the fourth area of the second information expansion page are diverse and may include, but are not limited to, the specific embodiments listed above.

[0196] It should be clear that the second information expansion page may also include an update control for updating multiple tags to be triggered in the fourth area of the second information expansion page.

[0197] General description of embodiments of the present disclosure

[0198] A vast amount of news reports, knowledge articles, and various other types of information are scattered throughout the Internet. If the public wants to obtain information about a certain event, they can obtain relevant information through various methods such as browsing the information columns pushed on the web page and searching for the descriptive words corresponding to this event to find relevant events. When an object browses the corpus information on the Internet (such as viewing search results), it often needs to click into a piece of corpus information to view the details of the corpus (such as viewing a search result). The detailed information of the corpus is generally presented on the landing page of a single piece of information. In this way, the content presented on the landing page will be limited to the detailed information of this event, and the presentation of content related to the antecedents, consequences, event context, etc. of this event will not be effective. In this case, the object browsing the web page will be difficult to understand the background and knowledge related to this corpus, the display efficiency of related information is low, and the effectiveness of information display is low. Therefore, how to further display the relevant information corresponding to the corpus information, quickly display more effective information, and improve the display efficiency of related information and the effectiveness of displayed information has become an urgent problem in the industry.

[0199] According to some embodiments of the present disclosure, an Internet corpus information expansion method is provided, which can be executed in an object terminal and is used to provide an Internet corpus information expansion service, capable of further displaying the relevant information corresponding to the corpus information, quickly displaying more effective information, and improving the display efficiency of related information and the effectiveness of displayed information.

[0200] Referring to Figure 3 , the Internet corpus information expansion method 300 of the embodiments of the present disclosure may include, but is not limited to, the following steps 310 to step 330.

[0201] Step 310, display a target Internet corpus, the target Internet corpus has a trigger word, and the trigger word is displayed in a display manner different from other parts of the target Internet corpus;

[0202] Step 320, in response to a first trigger on the trigger word, display a first information expansion page, the first information expansion page has a first area and a second area, the first area contains first explanation information about the trigger word, and the second area contains the trigger word and multiple to-be-triggered tags;

[0203] Step 330, in response to a second trigger on a target tag among the multiple to-be-triggered tags, display a second information expansion page, and the second information expansion page is associated with the target tag.

[0204] Through steps 310 to 330 shown in the embodiments of the present disclosure, the target Internet corpus has a trigger word, and when the target Internet corpus is displayed, the trigger word is displayed in a display manner different from other parts of the target Internet corpus. In response to the first trigger on the trigger word, a first information expansion page is displayed. The first information expansion page has a first area and a second area. The first area contains first explanation information about the trigger word, and the first explanation information is used to reveal the detailed information of the trigger word; the second area contains the trigger word and multiple to-be-triggered tags. Further, in response to the second trigger on the target tag among the multiple to-be-triggered tags, a second information expansion page is displayed, and the second information expansion page is associated with the target tag. In this way, through the second trigger on the to-be-triggered tag, the second information expansion page can be used to further display the associated information corresponding to the corpus information, so as to quickly display more effective information and improve the display efficiency of the associated information and the effectiveness of the displayed information.

[0205] The above steps 310 to 330 will be described below.

[0206] In step 310 of some embodiments, the target Internet corpus is displayed. The target Internet corpus has a trigger word, and the trigger word is displayed in a display manner different from other parts of the target Internet corpus. It should be noted that the target Internet corpus is the corpus that the object terminal is used to display on the terminal interface when browsing the web page on the Internet. It should be emphasized that the trigger word refers to the annotation word used to trigger the extendable associated information. The words containing extendable knowledge in the Internet corpus can be annotated to form the trigger word. When the trigger word is triggered, the associated information extended by the trigger word can be provided to the object browsing the web page. It should be pointed out that the display manner of the trigger word in the web page is different from that of other parts of the Internet corpus.

[0207] In some embodiments, if the target Internet corpus is displayed as "The first batch of pilots began to receive flight training for type A aircraft". If "type A aircraft" is the trigger word, then "type A aircraft" can be displayed in bold and underlined, and "The first batch of pilots began to receive" and "flight training" can be displayed in non-bold and non-underlined display manner. In this way, the trigger word can be displayed in a display manner different from other parts of the target Internet corpus. It should be understood that there are various feasible embodiments for the trigger word to be displayed in a display manner different from other parts of the target Internet corpus, and it is not limited to the above example.

[0208] Refer to Figure 3A , some more specific embodiments of displaying the target Internet corpus are shown. Figure 3A In the target Internet corpus is respectively displayed in three information columns. Specifically:

[0209] The target Internet corpus A shows that "the first batch of pilots began to receive flight training for type A aircraft; with multiple test flights, it means that type A aircraft will be used as the main carrier for cargo B in the future." Among them, "type A aircraft" is the trigger word A1, so the corresponding "Zhang Zhishi" label can be displayed beside the target Internet corpus A.

[0210] The target Internet corpus B shows that "cargo B leaked during transportation; when cargo B passed through the xxx flight segment during transportation, the storage warehouse leaked." Among them, "cargo B" is the trigger word B1, so the corresponding "Zhang Zhishi" label can be displayed beside the target Internet corpus B.

[0211] The target Internet corpus C shows that "xxxx agreed that xxx pilots would purchase an aircraft after completing type A flight training; xxxx company announced that it would complete the purchase of type A aircraft in three days, and at that time, the first batch of xxx pilots would complete flight training." Among them, "xxx pilots" is the trigger word C1 and "type A flight training" is the trigger word C2, so the corresponding "Zhang Zhishi" label can be displayed beside the Internet corpus C.

[0212] It should be understood that if there is no trigger word in the title of the information column, the "Zhang Zhishi" label may not appear in the title of the information column. In addition, the trigger word is displayed in a different way from other parts of the target Internet corpus. Figure 3A The trigger word is displayed by bolding and underlining.

[0213] In step 320 of some embodiments, in response to the first trigger of the trigger word, a first information expansion page is displayed. The first information expansion page has a first area and a second area. The first area contains the first explanatory information about the trigger word, and the second area contains the trigger word and multiple pending trigger labels. It should be noted that the first trigger refers to the corresponding action on the trigger word, such as a click operation of a keyboard and mouse device, a swiping operation of a touch screen, etc.

[0214] It should be clear that since the trigger word refers to a marked word used to trigger extendable associated information, the trigger word can display the first information expansion page through the first trigger. Among them, the first information expansion page has a first area and a second area. The first area contains the first explanatory information about the trigger word, and the second area contains the trigger word and multiple pending trigger labels. The first explanatory information is the explanatory information about the corresponding detailed content of the trigger word, and the specific meaning of the trigger word can be understood through the first explanatory information.

[0215] It should be understood that multiple tags to be triggered in the second region are generated based on the trigger word corresponding to the extendable associated information. Therefore, after responding to the first trigger, the tags to be triggered that can be further triggered can be displayed in the second region of the first information expansion page. After further triggering the tags to be triggered, the content of the associated information such as the cause and effect, event context, etc. of the trigger word can be presented on the interface.

[0216] It should be noted that the method for expanding Internet corpus information in the embodiments of the present disclosure for expanding the target Internet corpus includes both the expansion of the detailed information of the trigger word and the expansion of the associated information of the trigger word. The expansion of the detailed information of the trigger word is reflected in the first explanatory information of the trigger word in the first region; the expansion of the associated information of the trigger word is reflected in the multiple tags to be triggered that can be further triggered in the second region.

[0217] Refer to Figure 3B , which shows some more specific embodiments of displaying the first information expansion page. The first region includes the first explanatory information of the trigger word, and the second region includes the trigger word and multiple tags to be triggered. Specifically:

[0218] The first explanatory information of the trigger word "Type A aircraft" in the first region is displayed as: "Fourth-generation transport aircraft; Type A aircraft is a transport aircraft developed by xxxx company; Advantages of Type A aircraft: high transport capacity, high mobility, short takeoff and landing ability, multi-purpose performance, advanced avionics system, good protection and reliability, strong anti-leakage ability of the cargo hold".

[0219] The second region includes the trigger word "Type A aircraft" and multiple tags to be triggered. Among them, the tags to be triggered may include: "F16 transport aircraft", "xx route transport efficiency", "xxx depreciation", "xxxx bridge", "xxx route", "xxx opera house", "transport mode", "xxx transport leakage event", "xxx group".

[0220] In step 330 of some embodiments, in response to a second trigger on the target tag among the multiple tags to be triggered, a second information expansion page is displayed, and the second information expansion page is associated with the target tag. It should be noted that the second trigger is used to further trigger the tags to be triggered, and the tag to be triggered that is second-triggered is the target tag. Since the tags to be triggered are generated based on the extendable associated information corresponding to the trigger word, in response to the second trigger on the target tag among the multiple tags to be triggered, a second information expansion page can be displayed, and the second information expansion page is associated with the target tag. It should be pointed out that the second information expansion page is used to present the content of the associated information such as the cause and effect, event context, etc. of the trigger word. In this way, the corresponding associated information of the corpus information can be further displayed, and more effective information can be quickly displayed, improving the display efficiency of the associated information and the effectiveness of the displayed information.

[0221] Referring to Figure 3C , some more specific embodiments of displaying the second information expansion page are shown. After the target tag "xx route transportation efficiency" among multiple tags to be triggered is secondarily triggered, the second information expansion page can be displayed. The second information expansion page can include a third area and a fourth area. The third area contains second explanatory information about the target tag "xx route transportation efficiency". The second explanatory information specifically shows a schematic diagram of the transportation efficiency of the xx route in recent months, and shows that "the transportation efficiency dropped to 0.0766 on June 23, and xxx Group is facing increasingly low transportation efficiency and decides to improve the transportation mode by purchasing Class A aircraft for operation".

[0222] In addition, the fourth area of the second information expansion page shows multiple tags to be triggered corresponding to "xx route transportation efficiency". Among them, the words in the tags to be triggered can be words extracted from the associated information corresponding to the trigger word "xx route transportation efficiency". For example, "xx route" also operates "F22 transport aircraft" and "M142 transport aircraft" at the same time. Therefore, "F22 transport aircraft" and "M142 transport aircraft" can be used as words in the tags to be triggered; the "xx route transportation efficiency" dropped to 0.0766 on June 23, resulting in "xx asset depreciation". Therefore, "xx asset depreciation" can be used as a word in the tags to be triggered. It should be understood that the types of tags to be triggered in the fourth area of the second information expansion page are diverse and can include, but are not limited to, the specific embodiments listed above.

[0223] Detailed description of step 310

[0224] Referring to Figure 4 , according to some embodiments provided by the present disclosure, the target Internet corpus is Internet search results. Step 310 may include, but is not limited to, the following steps 410 to 420.

[0225] Step 410, obtaining an Internet search request;

[0226] Step 420, displaying multiple Internet search results corresponding to the Internet search request, and the Internet search results are sorted according to the number of trigger words included.

[0227] The above steps 410 to 420 will be described below.

[0228] It should be noted that if the object browsing the web wants to obtain information about a certain event, various ways such as searching for the descriptive words corresponding to this event can be used to find relevant information. Specifically, the target Internet corpus can be displayed through some embodiments shown in steps 410 to 420.

[0229] Step 410 of some embodiments is to obtain an Internet search request. It should be noted that an Internet search request refers to a request issued by an object browsing a web page for information search on the Internet, which is used to reflect the information that the object needs to search for.

[0230] Step 420 of some embodiments is to display multiple Internet search results corresponding to the Internet search request, and the Internet search results are sorted according to the number of trigger words they contain. It should be noted that the embodiments of the present disclosure can search by obtaining the information that the object inputs in the Internet search request, so as to obtain and display multiple Internet search results corresponding to the Internet search request. Among them, the Internet search results are sorted according to the number of trigger words they contain. It should be understood that since trigger words are used to trigger the annotation words of extensible associated information, if an Internet search result contains more trigger words, it means that this Internet search result can provide more extensible associated information for the Internet search request. Based on this, multiple Internet search results can be sorted according to the number of trigger words they contain.

[0231] In some more specific embodiments, controls such as a search bar and a search window can be used to collect the information that the object needs to search for, so as to obtain an Internet search request. Then, by calling a search engine or a large language model, the search can be further performed based on the Internet search request.

[0232] In the embodiments of the present disclosure shown in steps 410 to 420, an Internet search request is obtained, and multiple Internet search results corresponding to the Internet search request are displayed. The Internet search results are sorted according to the number of trigger words they contain. In this way, the Internet search results that provide more extensible associated information can be displayed, which is convenient for further displaying the associated information corresponding to the corpus information in subsequent steps.

[0233] Refer to Figure 5A , it is shown that "xxxx Group dispatches pilots for flight training" is filled in the search bar, aiming to issue an Internet search request to the object terminal according to this information that needs to be searched for.

[0234] Refer to Figure 5B , after obtaining the Internet search request based on "xxxx Group dispatches pilots for flight training", the object terminal displays multiple Internet search results corresponding to this Internet search request on the interface, and the Internet search results are sorted according to the number of trigger words they contain. Among them, the Internet search results containing the two trigger words "xxx pilot" and "Type A flight training" are ranked higher on the interface; the Internet search result containing the trigger word "B goods" is ranked in the middle of the interface; the Internet search results without trigger words are ranked lower on the interface.

[0235] It should be understood that there are various embodiments for obtaining an Internet search request, displaying a plurality of Internet search results corresponding to the Internet search request, and sorting the Internet search results according to the number of trigger words included, which may include, but are not limited to, the specific embodiments listed above.

[0236] Refer to Figure 6 , according to some embodiments provided by the present disclosure, 310 may include, but are not limited to, the following steps 610 to step 650.

[0237] Step 610, obtaining target Internet corpus;

[0238] Step 620, segmenting the target Internet corpus and determining matching words that match the candidate tags in the candidate tag set;

[0239] Step 630, obtaining the second popularity of the matching words;

[0240] Step 640, determining trigger words among the matching words based on the second popularity;

[0241] Step 650, displaying the target Internet corpus and displaying the trigger words in a display manner different from other parts of the target Internet corpus.

[0242] The following describes steps 610 to 650.

[0243] For steps 610 to 620 of some embodiments, the target Internet corpus is obtained, the target Internet corpus is segmented, and matching words that match the candidate tags in the candidate tag set are determined. It should be emphasized that the target Internet corpus is the corpus to be displayed on the terminal interface when the object terminal browses the web pages on the Internet. The candidate tag set refers to a set of candidate tags, and the candidate tags are obtained through pre-setting, and each candidate tag corresponds to a matching word for matching. Segmenting the target Internet corpus and determining the matching words that match the candidate tags in the candidate tag set aims to use the matching words to determine trigger words from the target Internet corpus.

[0244] It should be noted that word segmentation is the foundation of natural language processing. In some embodiments, since English sentences use spaces to separate words, word segmentation is not necessary in most cases, except for certain specific words (such as "how many" and "New York"). However, Chinese is different. It naturally lacks separators, requiring readers to segment words and punctuate sentences on their own. Therefore, when performing Chinese natural language processing, we need to first perform word segmentation. For Chinese word segmentation, current word segmentation methods are mainly divided into two categories: dictionary-based rule matching methods and statistical-based machine learning methods. First, dictionary-based word segmentation algorithms are essentially string matching. The string to be matched is matched against a sufficiently large dictionary based on a certain algorithm strategy. If a match is found, word segmentation can be performed. Depending on the different matching strategies, there are further categories such as forward maximum matching method, reverse maximum matching method, bidirectional matching word segmentation, and full segmentation path selection. Second, statistical-based word segmentation algorithms are essentially a sequence labeling problem. We label the words in the sentence according to their position in the word. The main annotations are: B (the first character of a word), E (the last character of a word), M (the middle character of a word, which may be multiple), and S (the word represented by a single character). For example, "Today's weather is really good" is annotated as "BESBESBEBME", and the corresponding word segmentation result is "Today / the / weather / is / really / good".

[0245] It should be understood that there are various methods for word segmentation, which can be word-by-word segmentation, segmentation according to Chinese vocabulary, or other segmentation methods. Therefore, word segmentation of the target text is not limited to the specific embodiments listed above.

[0246] In some embodiments, steps 630 to 650 obtain the second heat of the matching words, determine the trigger words in the matching words based on the second heat, display the target Internet corpus, and display the trigger words in a display method different from that of other parts of the target Internet corpus. It should be noted that in the target Internet corpus, there may be many words that can match the candidate tags. If a large number of trigger words are presented on a page, it will be inconvenient for the objects browsing the web page to select important words from them for triggering. Therefore, in the embodiments of the present disclosure, it is necessary to first obtain the second heat of the matching words, and then determine the trigger words based on the second heat, display the target Internet corpus, and display the trigger words in a display method different from that of other parts of the target Internet corpus. The second heat refers to the frequency of using the matching words on the Internet.

[0247] In some more specific embodiments, when the second heat of a certain matching word is greater than a preset heat threshold, a trigger word can be determined according to this corresponding matching word; in some other more specific embodiments, when the second heat of a certain matching word ranks among the top predetermined positions among all matching words, a trigger word can be determined according to this corresponding matching word. It should be understood that there are various feasible ways to determine the trigger word, which can include, but are not limited to, the specific examples given above.

[0248] Through the embodiments of the present disclosure shown in steps 610 to 650, a trigger word can be determined according to the second heat of the matching word, which helps to quickly display more effective information and further display the relevant information corresponding to the corpus information.

[0249] Refer to Figure 7A , the figure shows the obtained target Internet corpus. The embodiments of the present disclosure aim to determine trigger words from the target Internet corpus corresponding to the titles in three information columns. Based on this, it is necessary to first determine the target Internet corpus corresponding to the titles in the three information columns, which are "The first batch of pilots began to receive flight training for type A aircraft", "Goods B leaked during transportation", "xxxx agreed that xxx pilots purchase an aircraft after completing type A flight training".

[0250] Refer to Figure 7B , for the three target Internet corpora of "The first batch of pilots began to receive flight training for type A aircraft", "Goods B leaked during transportation", "xxxx agreed that xxx pilots purchase an aircraft after completing type A flight training", and perform word segmentation on them. Then, matching words that determine that the words segmented match the candidate labels in the candidate label set are obtained. The matching words can specifically include "the first batch of pilots", "type A aircraft", "flight training", "goods B", "transportation process", "leakage", "xxx pilots", "type A flight training". Further, the second heat of the matching words is obtained, and a trigger word is determined among the matching words according to the comparison of the second heat. The trigger words can specifically include "type A aircraft", "goods B", "xxx pilots", "type A flight training".

[0251] Refer to Figure 7C , after determining that the trigger words are "type A aircraft", "goods B", "xxx pilots", "type A flight training", the target Internet corpus can be displayed, and the trigger words are displayed in a display manner different from other parts of the target Internet corpus. Specifically, for the trigger words "type A aircraft", "goods B", "xxx pilots", "type A flight training", Figure 7CIt is displayed on the interface of the object terminal in a bold and underlined manner. In addition, for the title containing the trigger word, a "Gain Knowledge" label can also be added around the title in the information column to prompt the object browsing the web page that there is an expandable trigger word in the title.

[0252] Refer to Figure 8 According to some embodiments provided by the present disclosure, obtaining the second heat of the matching word in step 630 may include, but is not limited to, the following steps 810 to 830.

[0253] Step 810, determine the first number of times the matching word is searched by the object from the Internet search records;

[0254] Step 820, determine the second number of times the matching word appears in the search results of the Internet search records;

[0255] Step 830, determine the second heat of the matching word based on the first number and the second number.

[0256] The following describes steps 810 to 830.

[0257] In step 810 of some embodiments, determine the first number of times the matching word is searched by the object from the Internet search records. It should be noted that if the object executing the first trigger is the target object, the first number of times the matching word is searched by the object may specifically refer to the number of times the target object has searched for the matching word; the first number of times the matching word is searched by the object may also specifically refer to the number of times other objects browsing the web page have searched for the matching word except the target object.

[0258] In step 820 of some embodiments, determine the second number of times the matching word appears in the search results of the Internet search records. It should be noted that the second number of times the matching word appears may specifically refer to the number of times the matching word appears in the search results of the target object's previous Internet search records; the second number of times the matching word appears may also specifically refer to the number of times the matching word appears in the search results of other objects' previous Internet search records except the target object.

[0259] In step 830 of some embodiments, based on the first count and the second count, the second popularity of the matching word is determined. It should be noted that after clarifying the first count of the matching word searched by the object and the second count of the appearance of the matching word, it is possible to further determine the second popularity of the matching word based on the first count and the second count. It should be pointed out that in some other embodiments of the present disclosure, the second popularity of the matching word can be determined only based on the first count, or only based on the second count. However, considering that the first count is used to reflect the frequency of the object actively searching for the matching word, and the second count is used to reflect the frequency of the appearance of the matching word in the search results of Internet search records, on this basis, comprehensively considering the factor of the object's active search and the factor of the appearance density of the matching word in the search results can more objectively reflect the frequency of the use of the matching word on the Internet, that is, the second popularity.

[0260] The embodiments of the present disclosure illustrated by steps 810 to 830 determine the second popularity of the matching word by comprehensively considering the factor of the object's active search and the factor of the appearance density of the matching word in the search results. In order to determine the trigger word according to the second popularity of the matching word, it helps to quickly display more effective information and realize the further display of the relevant information corresponding to the corpus information.

[0261] Detailed description of step 320

[0262] Refer to Figure 9 , according to some embodiments provided by the present disclosure, step 320 responds to the first trigger of the trigger word and displays the first information expansion page, which may include, but is not limited to, the following steps 910 to 930.

[0263] Step 910, in response to the first trigger of the trigger word, input the trigger word plus the third guiding text into the large language model to obtain the third detailed information of the trigger word;

[0264] Step 920, input the third detailed information into the Hunyuan model to obtain the third summary information of the trigger word;

[0265] Step 930, in the first area, display the third summary information and the third detailed information as the first explanatory information.

[0266] The following explains steps 910 to 930.

[0267] In step 910 of some embodiments, in response to the first trigger of the trigger word, input the trigger word plus the third guiding text into the large language model to obtain the third detailed information of the trigger word.

[0268] It should be noted that a large language model refers to a language model pre-trained on a large-scale text corpus. These models typically use self-supervised learning methods to train on a large amount of unlabeled text data to learn the language structure and semantic information in the text. These models have powerful representation capabilities and can be applied to various natural language processing tasks, such as text extraction, text generation, text classification, sequence labeling, machine translation, etc. At the same time, large language models can also be adapted to the needs of specific tasks through techniques such as fine-tuning to achieve better performance.

[0269] It should be noted that a guiding text (Prompt), as the name implies, is text information that plays a guiding role. It should be understood that in the process of applying a large language model to the generation of detailed information, in order to obtain the third detailed information for explaining the detailed meaning of a trigger word, a guiding text can be pre-determined to convey the generation requirements of the third detailed information to the large language model. This guiding text is also the third guiding text. If the input text is randomly determined during the process of conveying the preset requirements to the large language model, the input text will be difficult to conform to the expression paradigm of the large language model. Therefore, in some embodiments of the present disclosure, it is necessary to add a third guiding text that conforms to the expression paradigm of the large language model based on the trigger word to further generate the text to be input into the large language model. In this way, higher-quality third detailed information can be obtained in the process of applying the large language model to the generation of the third detailed information. Based on this, in response to the first trigger of the trigger word, the embodiments of the present disclosure can input the trigger word plus the third guiding text into the large language model to obtain the third detailed information of the trigger word.

[0270] In step 920 of some embodiments, the third detailed information is input into the Hunyuan model to obtain the third summary information of the trigger word. It should be noted that the Hunyuan model in the embodiments of the present disclosure is a natural language model used to refine the semantic content of a large text to obtain a concise expression to summarize the main content of the text. It should be pointed out that since the third detailed information is used to explain the detailed meaning of the trigger word, inputting the third detailed information into the Hunyuan model aims to use the Hunyuan model to refine the semantic content of the large text to obtain the third summary information corresponding to the trigger word. Among them, the third summary information is used to relatively concisely summarize the meaning of the trigger word. The Hunyuan model is an existing large language model that can obtain a summary description of the input content.

[0271] In step 930 of some embodiments, in the first region, the third summary information and the third detailed information are displayed as the first explanatory information. It should be emphasized that the first information expansion page has a first region and a second region. The first region contains the first explanatory information about the trigger word, and the first explanatory information is used to reveal the detailed information of the trigger word. Therefore, on the basis that the third summary information is used to relatively concisely summarize the meaning of the trigger word, and the third detailed information is used to explain the detailed meaning of the trigger word, the third summary information and the third detailed information can be displayed in the first region as the first explanatory information.

[0272] In the embodiments of the present disclosure shown via steps 910 to 930, on the basis of generating the third detailed information using a large language model and generating the third summary information using the Hunyuan model, the third summary information and the third detailed information are displayed as the first explanatory information in the first region, which can display the meaning of the trigger word in a well-proportioned manner, and helps the object browsing the web page to understand the information about the person, event, etc. pointed to by the trigger word.

[0273] Refer to Figure 10 , in response to the first trigger of the trigger word "Type A aircraft", the trigger word "Type A aircraft" is added with the third guiding text "Please generate a detailed introduction to []" to form the text "Please generate a detailed introduction to Type A aircraft", and it is input into the large language model to obtain the third detailed information of the trigger word "Type A aircraft is a transport aircraft developed by xxxx company; as a fourth-generation transport aircraft, the advantages of Type A aircraft are high transport capacity, high mobility, short takeoff and landing ability, multi-purpose performance, advanced avionics system, good protection and reliability, and strong cargo hold leak prevention ability". Further, the Hunyuan model processes the third detailed information to extract the main content of the meaning of the trigger word "Type A aircraft" to obtain the third summary information "Type A aircraft is a fourth-generation transport aircraft, a transport aircraft developed by xxxx company". Furthermore, the above-obtained third summary information and third detailed information are displayed in the first region as the first explanatory information.

[0274] Refer to Figure 11 , according to some embodiments provided by the present disclosure, multiple candidate tags to be triggered are selected from the candidate tag set, which may include, but are not limited to, the following.

[0275] Step 1110, for each candidate tag in the candidate tag set, obtain the first distance between the candidate tag and the trigger word in the knowledge graph;

[0276] Step 1120, for each candidate tag, obtain the first popularity of the candidate tag;

[0277] Step 1130, based on the first distance and the first popularity, select the candidate tags to be triggered from the candidate tag set.

[0278] The following describes steps 1110 to 1130.

[0279] A Knowledge Graph (KG) is a structured knowledge base, which is essentially a labeled directed graph. Each node in the graph represents an entity, and each edge represents a relationship, represented by a standard triple (s, r, o), where s and o are the head entity and the tail entity respectively, and r is the relationship between s and o. Knowledge graphs are widely used in many scenarios, such as semantic search, intelligent question answering, and auxiliary decision-making. However, although the creation and maintenance of knowledge graphs require a lot of costs, even the largest knowledge graphs face problems such as data sparsity and data missing. Therefore, in order to make the knowledge graph more complete and accurate, it is necessary to continuously expand and improve it so that the knowledge graph model can be continuously constructed. It should be noted that the trigger words and the words corresponding to the candidate labels in the embodiments of the present disclosure can query the corresponding entities in the pre-set knowledge graph. In order to perform operations such as entity disambiguation, knowledge fusion, and entity disambiguation, the embodiments of the present disclosure can introduce the concept of distance measurement. If there are fewer edges connecting two entities in the knowledge graph, it means that the distance between these two entities is closer. Since the edges between entities in the knowledge graph represent the relationships between entities, fewer edges connecting two entities in the knowledge graph also mean stronger relevance between the two entities.

[0280] In step 1110 of some embodiments, for each candidate label in the candidate label set, obtain the first distance between the candidate label and the trigger word in the knowledge graph. It should be noted that each candidate label in the candidate label set can correspond to an entity in the knowledge graph. Similarly, the trigger word also corresponds to an entity in the knowledge graph. In the embodiments of the present disclosure, for each candidate label in the candidate label set, the first distance between the candidate label and the trigger word in the knowledge graph can be obtained, aiming to quantify the relevance between the candidate label and the trigger word through the first distance. If the first distance is shorter, it means that the relevance between the candidate label and the trigger word is stronger.

[0281] In step 1120 of some embodiments, for each candidate label, obtain the first popularity. It should be noted that the first popularity refers to the frequency of using the vocabulary corresponding to the candidate label on the Internet. If the first popularity is higher, it means that the frequency of using the vocabulary corresponding to the candidate label on the Internet is higher.

[0282] In step 1130 of some embodiments, based on the first distance and the first popularity, select the to-be-triggered label from the candidate label set.

[0283] It should be noted that after determining the first distance between the candidate label and the trigger word in the knowledge graph and the first popularity of the candidate label, it is possible to further determine the second popularity of the candidate label based on the first distance and the first popularity. It should be pointed out that in some other embodiments of the present disclosure, the to-be-triggered label can be selected from the candidate label set only based on the first distance, or the to-be-triggered label can be selected from the candidate label set only based on the first popularity. However, considering that the first distance is used to reflect the relevance between the candidate label and the trigger word, and the first popularity is used to reflect the frequency of the vocabulary corresponding to the candidate label used on the Internet, on this basis, considering both the factor of the relevance between the candidate label and the trigger word and the factor of the frequency of the vocabulary corresponding to the candidate label used on the Internet can more objectively reflect which candidate labels in the candidate label set are more suitable for expanding and displaying the associated information of the trigger word, so as to identify the candidate labels with relatively high expansion value, and then select them as the to-be-triggered labels.

[0284] The embodiments of the present disclosure shown in steps 1110 to 1130 can determine the candidate labels suitable for expanding the key information of the associated words from the candidate label set, and select these candidate labels with relatively high expansion value as the to-be-triggered labels, which helps to further display the associated information corresponding to the corpus information, quickly display more effective information, and improve the display efficiency of the associated information and the effectiveness of the displayed information.

[0285] Refer to Figure 12A, for each candidate tag in the candidate tag set, it is necessary to obtain the first distance between the candidate tag and the trigger word in the knowledge graph, and then compare based on the first distance between each candidate tag and the trigger word in the knowledge graph; in addition, for each candidate tag, it is necessary to obtain the first popularity of the candidate tag, and then compare based on the first popularity of each candidate tag. According to the results of the comparison of the first distance and the results of the comparison of the first popularity, select the tags to be triggered from the candidate tag set. Among them, since the first distance between the candidate tag "xxx transportation leakage event" and the trigger word "Class A aircraft" is relatively short, and the first popularity is the highest, it can be determined as the tag to be triggered; since the first distance between the candidate tag "F16 transport aircraft" and the trigger word "Class A aircraft" is relatively short, and the first popularity ranks third, it can be determined as the tag to be triggered; since the first distance between the candidate tag "xxx group" and the trigger word "Class A aircraft" is relatively short, and the first popularity ranks fourth, it can be determined as the tag to be triggered; since the first distance between the candidate tag "xxx devaluation" and the trigger word "Class A aircraft" is separated by one entity, which is relatively long, but the first popularity ranks fifth, it can be determined as the tag to be triggered. In addition, since the first distance between the candidate tag "xxx opera house" and the trigger word "Class A aircraft" is separated by two entities, which is the longest among multiple candidate tags, although the first popularity ranks second, it is not sufficient to be determined as the tag to be triggered considering comprehensively. It should be understood that based on the first distance and the first popularity, selecting the tags to be triggered from the candidate tag set can be flexibly adjusted according to application requirements, and will not be listed one by one here.

[0286] Refer to Figure 12B , via Figure 12A After determining the four tags to be triggered, namely "xxx transportation leakage event", "F16 transport aircraft", "xxx group" and "xxx devaluation", according to the embodiments shown, these four tags to be triggered can be displayed in the second area of the first information expansion page.

[0287] Refer to Figure 13 , according to some embodiments provided by the present disclosure, step 1130 may include, but is not limited to, the following steps 1310 to 1320.

[0288] Step 1310, based on the first distance, select the first number of tags to be triggered from the candidate tag set, where the first number is obtained based on the number of tags to be triggered and the first ratio;

[0289] Step 1320, based on the first popularity, select the remaining tags to be triggered among the multiple tags to be triggered from the candidate tags that have not been selected in the candidate tag set.

[0290] The following explains steps 1310 to 1320.

[0291] In step 1310 of some embodiments, based on the first distance, the first number of tags to be triggered are selected from the candidate tag set, and the first number is obtained based on the number of tags to be triggered and the first ratio. It should be emphasized that the first distance between the candidate tag and the trigger word in the knowledge graph is used to quantify the relevance between the candidate tag and the trigger word. If the first distance is shorter, it means that the relevance between the candidate tag and the trigger word is stronger. It should be clear that the number of tags to be triggered to be displayed in the second area of the first information expansion page is the first number, and the first number can be one or more than one. Among them, the first number is obtained based on the number of tags to be triggered and the first ratio. The tags to be triggered are selected from the candidate tag set based on the first distance and the first popularity, and the first ratio is used to determine the proportion of the tags to be triggered selected based on the first distance among all the tags to be triggered to be displayed.

[0292] In some more specific embodiments, if 10 tags to be triggered need to be displayed in the second area of the first information expansion page, and the first ratio is 60%, then 6 out of the 10 tags to be triggered are selected from the candidate tag set based on the first distance. It should be understood that in the above embodiments, 6 candidate tags with the shortest first distance can be selected from the candidate tag set as the tags to be triggered according to the first distance between each candidate tag and the trigger word in the knowledge graph.

[0293] In step 1320 of some embodiments, based on the first popularity, the remaining tags to be triggered among the multiple tags to be triggered are selected from the candidate tags that have not been selected in the candidate tag set. It should be emphasized that the first popularity refers to the frequency of using the vocabulary corresponding to the candidate tag on the Internet. If the first popularity is higher, it means that the frequency of using the vocabulary corresponding to the candidate tag on the Internet is higher. It should be clear that the tags to be triggered are selected from the candidate tag set based on the first distance and the first popularity. If among all the tags to be triggered, the first number of tags to be triggered are selected from the candidate tag set based on the first distance, then based on the first popularity, the remaining tags to be triggered among the multiple tags to be triggered can be selected from the candidate tags that have not been selected in the candidate tag set.

[0294] Through the embodiments of the present disclosure shown in steps 1310 to 1320, candidate tags suitable for expanding the key information of associated words can be determined from the candidate tag set, which is convenient for the object browsing the web page to master two types of information and improve the diversity of information display. Selecting these candidate tags with relatively high expansion value as the tags to be triggered helps to further display the associated information corresponding to the corpus information, quickly display more effective information, and improve the display efficiency of associated information and the effectiveness of the displayed information.

[0295] Refer to Figure 14, in some embodiments provided by the present disclosure, step 1130 selects the to-be-triggered tags from the candidate tag set based on the first distance and the first popularity, and may further include, but are not limited to, the following steps 1410 to 1440.

[0296] Step 1410, determine the first score of the to-be-triggered tag based on the first distance;

[0297] Step 1420, determine the second score of the to-be-triggered tag based on the first popularity;

[0298] Step 1430, determine the total score of the to-be-triggered tag based on the first score and the second score;

[0299] Step 1440, select the to-be-triggered tag from the candidate tag set based on the total score.

[0300] The following describes steps 1410 to 1420.

[0301] In step 1410 of some embodiments, the first score of the to-be-triggered tag is determined based on the first distance. It should be emphasized that the first distance between the candidate tag and the trigger word in the knowledge graph is used to quantify the relevance between the candidate tag and the trigger word. If the first distance is shorter, it means that the relevance between the candidate tag and the trigger word is stronger. Therefore, the first score of the to-be-triggered tag can be determined based on the first distance, where the first score is used to measure the relevance between the to-be-triggered tag and the trigger word.

[0302] In step 1420 of some embodiments, the second score of the to-be-triggered tag is determined based on the first popularity. It should be emphasized that the first popularity refers to the frequency of using the vocabulary corresponding to the candidate tag on the Internet. If the first popularity is higher, it means that the frequency of using the vocabulary corresponding to the candidate tag on the Internet is higher. Therefore, the second score of the to-be-triggered tag can be determined based on the first popularity, where the second score is used to measure the frequency of using the to-be-triggered tag on the Internet.

[0303] In steps 1430 to 1440 of some embodiments, the total score of the to-be-triggered tag is determined based on the first score and the second score, and the to-be-triggered tag is selected from the candidate tag set based on the total score. It should be noted that after determining the first score used to measure the relevance between the to-be-triggered tag and the trigger word and the second score used to measure the frequency of using the to-be-triggered tag on the Internet, the total score of each to-be-triggered tag can be determined based on the first score and the second score, and then it can be determined which to-be-triggered tags need to be displayed in the second area of the first information expansion page according to the total scores of the respective to-be-triggered tags.

[0304] In some more specific embodiments, based on the first score and the second score, the total score of each tag to be triggered can be determined by averaging the first score and the second score, or by adding the first score and the second score, or by performing weighted summation on the first score and the second score. Based on the first score and the second score, the total score of each tag to be triggered can also be determined by other methods, which are not listed one by one here.

[0305] It should be understood that there are various embodiments of selecting tags to be triggered from the candidate tag set based on the first distance and the first popularity, not limited to the above examples.

[0306] Refer to Figure 15 , according to some embodiments provided by the present disclosure, the candidate tag set may include candidate tags and display modes corresponding to the candidate tags. The generation of the candidate tag set may include, but is not limited to, the following steps 1510 to step 1540.

[0307] Step 1510, obtain the first corpus sample library;

[0308] Step 1520, segment the first corpus sample library, and each word obtained by segmentation is used as a candidate tag and added to the candidate tag library;

[0309] Step 1530, based on at least one of the knowledge graph, the object behavior data of the corpus sample having the candidate tag in the first corpus sample library, and the third popularity of the corpus sample, determine the display mode of the candidate tag and store it corresponding to the candidate tag in the candidate tag set;

[0310] Step 1540, display the first information expansion page, including: displaying the tags to be triggered in the second area in the display mode corresponding to the candidate tags in the candidate tag set.

[0311] The following explains steps 1510 to 1540.

[0312] In step 1510 of some embodiments, the first corpus sample library is obtained. It should be noted that the first corpus sample library stores corpora corresponding to various contents and is used to provide corpus materials for generating the candidate tag set.

[0313] In step 1520 of some embodiments, the first corpus sample library is segmented, and each word obtained by segmentation is used as a candidate tag and added to the candidate tag library. It should be noted that the purpose of segmenting the first corpus sample library is to divide the corpora in the first corpus sample library into multiple words, and the words obtained by dividing the corpora in the first corpus sample library are also corpus samples. Then each corpus sample is used as a candidate tag and added to the candidate tag library.

[0314] In step 1530 of some embodiments, based on at least one of a knowledge graph, object behavior data of a corpus sample having a candidate label in the first corpus sample library, and the third popularity of the corpus sample, determine the display mode of the candidate label, and store it corresponding to the candidate label in the candidate label set.

[0315] It should be noted that for the word corresponding to the corpus sample in the embodiments of the present disclosure, the corresponding entity can be queried in a pre-set knowledge graph. The fewer the edges connecting two entities in the knowledge graph, the stronger the relevance between the two entities.

[0316] If the object executing the first trigger is the target object, then the object behavior data of the corpus sample in the embodiments of the present disclosure can be the interaction behavior data related to the corpus sample in the interaction behaviors that the target object has ever implemented on the Internet; it can also be the interaction behavior data related to the corpus sample in the interaction behaviors that other objects except the target object have ever implemented on the Internet.

[0317] The third popularity of the corpus sample refers to the frequency of using the corpus sample in the Internet.

[0318] It should be clear that the knowledge graph, object behavior data of a corpus sample having a candidate label in the first corpus sample library, and the third popularity of the corpus sample respectively correspond to three evaluation dimensions of the corpus sample in terms of extended value. Among them, the knowledge graph corresponds to the relevance between the corpus sample and the trigger word, the object behavior data corresponds to the usage habit of the trigger word by the object in Internet interaction operations, and the third popularity corresponds to the frequency of the general public using the corpus sample in the Internet. Based on this, the embodiments of the present disclosure can be based on at least one of a knowledge graph, object behavior data of a corpus sample having a candidate label in the first corpus sample library, and the third popularity of the corpus sample, and according to at least one of the three evaluation dimensions corresponding to the extended value, determine the display mode of the candidate label, and store it corresponding to the candidate label in the candidate label set.

[0319] It should be noted that according to the evaluation dimensions corresponding to the extended values of different candidate labels, a display mode matching the evaluation dimension can be determined for the candidate label. The display modes of the candidate labels selected based on different evaluation dimensions can be different. Specifically, the differences between the display modes of the candidate labels can be reflected in multiple aspects. The candidate labels of different evaluation dimensions can be different in shape and appearance, can also be different in filling mode, and can also be different in other aspects of the display mode. Details are not listed one by one here.

[0320] Step 1540 of some embodiments displays a first information expansion page, including: displaying a to-be-triggered label in a second area in a display manner corresponding to the candidate label in the candidate label set. It should be noted that after the display manner corresponding to the candidate label is determined, it is stored in the candidate label set. When selecting a to-be-triggered label from the candidate label set and displaying the first information expansion page, the to-be-triggered labels can be displayed in the second area of the first information expansion page in their respective corresponding display manners.

[0321] The embodiments of the present disclosure illustrated via steps 1510 to 1540 determine the display manner corresponding to the candidate label by following at least one evaluation dimension among the relevance between the corpus sample and the trigger word, the usage habit of the trigger word by the object in Internet interaction operations, and the frequency of the corpus sample used by the public on the Internet, which helps to distinguish the expansion value of different candidate labels in different evaluation dimensions, so that the object browsing the web page can select the corresponding expansion direction according to the display manner of the candidate label.

[0322] Refer to Figure 16A , in some more specific embodiments, it is necessary to first obtain a first corpus sample library. The corpus included in the first corpus sample library can be "The fourth-generation transport aircraft includes Class A aircraft, F16 transport aircraft", "XXX Group considers using Class A aircraft to undertake the transport tasks of XXX routes due to transport efficiency reasons", "XXX route passes by XXX Opera House, XXXX Bridge", "Transport leakage incident occurred on XXX route", "The existing operation mode of XXX Group results in low transport efficiency on XXX route", "The transport efficiency of XXX Group on XXX route is low, so that XXX depreciates".

[0323] Furthermore, the first corpus sample library is segmented, and the obtained corpus samples include: "F16 transport aircraft", "Transport efficiency of xx route", "XXX depreciation", "XXXX Bridge", "XXX route", "XXX Opera House", "Transport mode", "XXX transport leakage incident", "XXX Group". Then, the above corpus samples are used as candidate labels and added to the candidate label library.

[0324] Still further, based on at least one of the knowledge graph, the object behavior data of the corpus sample having the candidate label in the first corpus sample library, and the third popularity of the corpus sample, determine the display manner corresponding to the candidate label and store it in the candidate label set corresponding to the candidate label.

[0325] It should be noted that based on the knowledge graph, it can be determined that the shape of the candidate label "F16 transport aircraft" is a circle and it is displayed in this way.

[0326] It should be noted that based on the object behavior data of the corpus samples with candidate tags in the first corpus sample library, it can be determined that the shapes of the candidate tags "xxxx Bridge", "xxx Route", and "xxx Opera House" are clouds, and they are displayed in this way.

[0327] It should be noted that based on the third popularity of the corpus samples, it can be determined that the shapes of the candidate tags "xx Route Transportation Efficiency", "xxx Depreciation", "Transportation Mode", and "xxx Transportation Leakage Incident" are rectangles, and they are displayed in this way.

[0328] In addition, in addition to being able to determine the display method of the candidate tags based on any one of the knowledge graph, the object behavior data of the corpus samples with candidate tags in the first corpus sample library, and the third popularity of the corpus samples, it is also possible to combine any two of the knowledge graph, the object behavior data of the corpus samples with candidate tags in the first corpus sample library, and the third popularity of the corpus samples to determine the display method of the candidate tags. For example, by combining the knowledge graph and the object behavior data of the corpus samples with candidate tags in the first corpus sample library, it can be determined that the shape of the candidate tag "xxx Group" is a hexagon, and it is displayed in this way.

[0329] It should be understood that the above embodiments are examples of the display method in terms of shape. In other embodiments of the present disclosure, the differences in the evaluation dimensions corresponding to each candidate tag in terms of extended value can also be reflected in various ways such as the filling color, filling texture, and border line type of the candidate tag.

[0330] Refer to Figure 16B , display the first information expansion page, display the tags to be triggered in the second area, and also include displaying the tags to be triggered in the second area in the display manner corresponding to the candidate tags in the candidate tag set. Among them, multiple tags to be triggered are displayed in the second area of the first information expansion page, including: the tag to be triggered "F16 Transport Aircraft" in a circular shape; the tags to be triggered "xxxx Bridge", "xxx Route", and "xxx Opera House" in a cloud shape; the tags to be triggered "xx Route Transportation Efficiency", "xxx Depreciation", "Transportation Mode", and "xxx Transportation Leakage Incident" in a rectangular shape; the tag to be triggered "xxx Group" in a hexagonal shape.

[0331] Refer to Figure 17 , according to some embodiments provided by the present disclosure, step 1530 may include, but is not limited to, the following steps 1710 to step 1720.

[0332] Step 1710: For any two candidate tags in the candidate tag set, obtain the target corpus in the first corpus sample library that contains both candidate tags.

[0333] Step 1720: If the target corpus has been accessed by the target object that has undergone the first trigger, the third heat of the target corpus is greater than the first threshold, and the first distance between the two candidate tags in the knowledge graph is less than the second threshold, then assign the same display method to the two candidate tags.

[0334] The following explains Steps 1710 to 1720.

[0335] In Step 1710 of some embodiments, for any two candidate tags in the candidate tag set, obtain the target corpus in the first corpus sample library that contains both candidate tags. It should be noted that for any two candidate tags in the candidate tag set, if there is a target corpus in the first corpus sample library that contains both candidate tags, it means that the sources of the two candidate tags are the same.

[0336] For example, if the candidate tag set includes two candidate tags, namely the candidate tag "xxx Group" and the candidate tag "xxx Route". And in the first corpus sample library, there is such a corpus: "xxx Group considers using Class A aircraft to undertake the transportation tasks of xxx Route due to transportation efficiency reasons". Since this corpus contains both the candidate tag "xxx Group" and the candidate tag "xxx Route", this corpus can be obtained as the target corpus.

[0337] In Step 1720 of some embodiments, if the target corpus has been accessed by the target object that has undergone the first trigger, the third heat of the target corpus is greater than the first threshold, and the first distance between the two candidate tags in the knowledge graph is less than the second threshold, then assign the same display method to the two candidate tags. It should be emphasized that the object behavior data of the corpus sample can be the interaction behavior data related to the corpus sample in the interaction behaviors that the target object has implemented on the Internet. Therefore, according to the object behavior data of the corpus sample, it can be determined whether the target corpus has been accessed by the target object that has undergone the first trigger. If the target corpus has been accessed by the target object that has undergone the first trigger, it means that the target object has had an interaction intention with the target corpus. It should be noted that if the third heat of the target corpus is greater than the first threshold, it means that the two candidate tags are used frequently in the Internet. In addition, the first distance between the two candidate tags in the knowledge graph is less than the second threshold, which means that the two candidate tags have a high degree of relevance.

[0338] Based on this, when two candidate tags have the same source of the target corpus, the target object has an interaction intention with the target corpus that is the source of the candidate tags, the frequency of use of the two candidate tags on the Internet is relatively high, and the two candidate tags also have a high correlation in the knowledge graph. It can be clear that the evaluation dimensions of the two candidate tags in terms of expansion value are relatively close, so the same display method can be assigned to the two candidate tags.

[0339] In the embodiment of the present disclosure shown by steps 1710 to 1720, since the display method of the candidate tag matches the evaluation dimension of the candidate tag in terms of expansion value during the formation process, if the evaluation dimensions of the two candidate tags in terms of expansion value are relatively close, assigning the same display method to the two candidate tags helps to distinguish the expansion values of different candidate tags in different evaluation dimensions, so that the object browsing the web page can select the corresponding expansion direction according to the display method of the candidate tag.

[0340] Referring to Figure 18 , according to some embodiments provided by the present disclosure, step 1530 may include, but is not limited to, the following steps 1810 to 1820.

[0341] Step 1810, for any two candidate tags in the candidate tag set, obtain the relationship between the two candidate tags in the knowledge graph;

[0342] Step 1820, based on the type of the relationship, determine the display method of the two candidate tags.

[0343] For steps 1810 to 1820 of some embodiments, for any two candidate tags in the candidate tag set, obtain the relationship between the two candidate tags in the knowledge graph, and based on the type of the relationship, determine the display method of the two candidate tags. It should be noted that the words corresponding to the candidate tags in the embodiments of the present disclosure can query the corresponding entities in the pre-set knowledge graph. If there are fewer edges connecting two entities in the knowledge graph, it means that the distance between the two entities is closer. Since the edges between entities in the knowledge graph represent the relationships between entities, fewer edges connecting two entities in the knowledge graph also means stronger correlation between the two entities. It should be pointed out that the knowledge graph is essentially a labeled directed graph, each node in the graph represents an entity, and each edge represents a relationship. The types of relationships between different entities can include, but are not limited to:

[0344] "Commonality", which means that two entities have the same properties, for example, "Type A aircraft" and "F16 transport aircraft" both belong to the fourth-generation transport aircraft;

[0345] "Subordination", which means that there is a subordination relationship between two entities, for example, "Type A aircraft" belongs to "xxx Group";

[0346] "Extension" means that an entity can be associated with another entity through an extended meaning. For example, an "Aircraft of Type A" can be associated with a "Transportation Mode" through the extension of its transportation function. Another example is that the locations passed by an "xxx Route" can be associated with an "xxxx Bridge" and an "xxx Opera House" through its geographical location extension;

[0347] "Causality" means that there is a causal association between two entities in terms of events. For example, low transportation efficiency is caused by the transportation mode. Based on this, there can be a "causality" relationship between the two candidate labels of "Transportation Mode" and "Low Transportation Efficiency".

[0348] It should be understood that the types of relationships between two candidate labels in a knowledge graph are diverse and can include, but are not limited to, the examples above.

[0349] It should be noted that the display method of a certain candidate label corresponds to the evaluation dimension of the extended value of this candidate label. Based on this, the display methods of two candidate labels can be determined based on the type of relationship between the two candidate labels.

[0350] In some embodiments, after determining the type of relationship between two candidate labels, these two candidate labels can be displayed in the same shape, or these two candidate labels can be displayed in the same color, or these two candidate labels can be displayed in other same appearance types.

[0351] The embodiments of the present disclosure shown from step 1810 to step 1820 can display these two candidate labels in the same appearance type according to the type of relationship between the two candidate labels, so that an object browsing the web page can select the corresponding extension direction according to the display methods of the candidate labels.

[0352] According to some embodiments provided by the present disclosure, step 1530 can include, but is not limited to, the following steps:

[0353] For any two candidate labels in the candidate label set, if the two candidate labels are jointly associated with the same event in the knowledge graph, the time proximity in the two corpora of the two candidate labels in the first corpus sample library meets the first condition, and the two candidate labels have the same type, then the two candidate labels are assigned the same display method.

[0354] It should be noted that if two candidate labels in the knowledge graph are co - related to the same event, it can be determined that these two candidate labels are closely related to this event in terms of content; if the time proximity in the two corpora with the two candidate labels in the first corpus sample library meets the first condition, it means that the two candidate labels are closely related in terms of time; if the two candidate labels have the same type, it means that the two candidate labels are closely related in terms of category attributes. Based on this, the same display method can be assigned to these two candidate labels, so that the object browsing the web page can select the corresponding expansion direction according to the display method of the candidate labels.

[0355] Referring to Figure 19 , according to some embodiments provided by the present disclosure, step 1530 may include, but is not limited to, the following steps 1910 to step 1920.

[0356] Step 1910, obtain the candidate label number threshold for each display method;

[0357] Step 1920, if, based on the knowledge graph, the object behavior data of the corpus sample with the candidate label in the first corpus sample library, and at least one of the third heat of the corpus sample, after assigning a display method to the candidate label, the number of candidate labels of the display method exceeds the candidate label number threshold, cancel the assigned display method for the candidate label.

[0358] For steps 1910 to 1920 of some embodiments, obtain the candidate label number threshold for each display method. If, based on the knowledge graph, the object behavior data of the corpus sample with the candidate label in the first corpus sample library, and at least one of the third heat of the corpus sample, after assigning a display method to the candidate label, the number of candidate labels of the display method exceeds the candidate label number threshold, cancel the assigned display method for the candidate label. It should be noted that each display method has a corresponding candidate label number threshold. It should be pointed out that if, based on the knowledge graph, the object behavior data of the corpus sample with the candidate label in the first corpus sample library, and at least one of the third heat of the corpus sample, after assigning a display method to the candidate label, if the number of candidate labels of the same display method reaches the corresponding candidate label number threshold, then for those candidate labels that exceed the candidate label number threshold under this display method, cancel the assigned display method for them. In this way, it can be avoided that the number of candidate labels under a certain display method exceeds the corresponding candidate label number threshold.

[0359] Through the embodiments of the present disclosure shown in steps 1910 to 1920, the number of candidate labels under the same display method can be limited by the candidate label number threshold, avoiding too many candidate labels of a certain display method in the candidate label set and enriching the diversity of display methods for candidate labels.

[0360] Reference Figure 20 According to some embodiments provided by the present disclosure, in step 320, displaying the first information expansion page may include, but is not limited to, the following steps 2010 to 2020.

[0361] Step 2010, determining a first distance between a to-be-triggered tag and a trigger word in a knowledge graph;

[0362] Step 2020, based on the first distance, displaying a plurality of to-be-triggered tags such that a second distance between a to-be-triggered tag and the trigger word in a second region is associated with the first distance.

[0363] In step 2010 of some embodiments, determining a first distance between a to-be-triggered tag and a trigger word in a knowledge graph. It should be noted that each candidate tag in the candidate tag set may correspond to an entity in the knowledge graph. Similarly, the trigger word also corresponds to an entity in the knowledge graph. In the embodiments of the present disclosure, for each candidate tag in the candidate tag set, the first distance between the candidate tag and the trigger word in the knowledge graph can be obtained, aiming to quantify the relevance between the candidate tag and the trigger word through the first distance. If the first distance is shorter, it means that the relevance between the candidate tag and the trigger word is stronger.

[0364] In step 2020 of some embodiments, based on the first distance, displaying a plurality of to-be-triggered tags such that a second distance between a to-be-triggered tag and the trigger word in a second region is associated with the first distance. It should be noted that the to-be-triggered tags are candidate tags selected from the candidate tag set. The distance between the candidate tag and the trigger word in the knowledge graph is the first distance, and the distance between the to-be-triggered tag and the trigger word in the knowledge graph is the second distance. It should be pointed out that displaying a plurality of to-be-triggered tags based on the first distance means determining the second distance between the to-be-triggered tag and the trigger word in the knowledge graph based on the first distance between the candidate tag and the trigger word in the knowledge graph. In this way, since the second distance is determined based on the first distance, the second distance between the to-be-triggered tag and the trigger word in the second region is associated with the first distance.

[0365] In some more specific embodiments, the to-be-triggered tags can be selected in the candidate tag set according to the length of the first distance from the trigger word. Since the shorter the first distance, the stronger the relevance between the candidate tag and the trigger word, several candidate tags with a shorter first distance can be selected in the candidate tag set and used as the to-be-triggered tags to be displayed in the second region of the first information expansion page.

[0366] The embodiments of the present disclosure shown via step 2010 to step 2020 can select a to-be-triggered tag that can be displayed in the second area of the first information expansion page from a set of candidate tags based on the relevance between the candidate tag and the trigger word reflected by the first distance, so as to display a second information expansion page in response to a second trigger on the target tag among multiple to-be-triggered tags, and the second information expansion page is associated with the target tag. In this way, through the second trigger on the to-be-triggered tag, the second information expansion page can be used to further display the relevant information corresponding to the corpus information, so as to quickly display more effective information and improve the display efficiency and effectiveness of the relevant information.

[0367] According to some embodiments provided by the present disclosure, the first information expansion page may have an update control for multiple to-be-triggered tags. After step 320, the Internet corpus information expansion method may further include:

[0368] Updating multiple to-be-triggered tags in response to a third trigger on the update control. It should be noted that the update control in the first information expansion page is used to update multiple to-be-triggered tags in the second area of the first information expansion page. In this way, the update control can be used to replace multiple to-be-triggered tags in the second area, which helps to display to-be-triggered tags that meet the expansion needs of the object performing the third trigger.

[0369] Refer to Figure 21A , an optional example diagram of the first information expansion page is shown in the figure. It can be clear that the second area of the first information expansion page contains multiple to-be-triggered tags and an update control for updating multiple to-be-triggered tags. It should be pointed out that before the third trigger occurs on the update control of the first information expansion page, the to-be-triggered tags specifically included in the second area of the first information expansion page are: "F16 transport aircraft", "transport efficiency of xx route", "depreciation of xxx", "xxxx bridge", "xxx route", "xxx opera house", "transport mode", "xxx transport leakage event", "xxx group".

[0370] It should be pointed out that when the third trigger occurs on the update control of the first information expansion page, multiple to-be-triggered tags in the second area change.

[0371] Refer to Figure 21B , the figure shows the corresponding first information expansion page after the third trigger occurs on the update control of the first information expansion page. Among them, the to-be-triggered tags specifically included in the second area of the first information expansion page change to: "F22 transport aircraft", "M142 transport aircraft", "xxx group", "new transport model", "xxx's active resignation", "xxx opera house", "transport mode", "B goods", "B-type aircraft", "depreciation of xx assets".

[0372] According to some embodiments provided by the present disclosure, a plurality of to-be-triggered tags can be divided into multiple groups, where the to-be-triggered tags in each group have the same display manner and balanced numbers. To respond to the third trigger on the update control, updating the plurality of to-be-triggered tags may include:

[0373] Updating the to-be-triggered tags in each group so that the number of updated to-be-triggered tags in each group is the same as the number of to-be-triggered tags before the update. It should be noted that in order to enable the first information expansion page to display relatively rich to-be-triggered tags in the second region, the multiple to-be-triggered tags updated in the second region can be divided into multiple groups according to their display manners, and the number of each group is balanced. Based on this, after updating the to-be-triggered tags that need to be displayed in the second region, it is also necessary to divide the multiple to-be-triggered tags into multiple groups according to the corresponding display manners. Therefore, during the process of updating the to-be-triggered tags in each group, it is necessary to make the number of updated to-be-triggered tags in each group the same as the number of to-be-triggered tags before the update.

[0374] Referring to Figure 22A , an optional example diagram of the first information expansion page is shown in the figure. It can be clear that the second region of the first information expansion page contains a plurality of to-be-triggered tags and an update control for updating the plurality of to-be-triggered tags. It should be pointed out that before the third trigger occurs on the update control of the first information expansion page, the to-be-triggered tags specifically included in the second region of the first information expansion page are:

[0375] Capsule-shaped to-be-triggered tags: "xxx transportation leakage event", "xxx Group", and "Fourth-generation transport aircraft";

[0376] Cloud-shaped to-be-triggered tags: "xxxx Bridge", "xxx Shipping Route", and "xxx Opera House";

[0377] Rectangular to-be-triggered tags: "xx shipping route transportation efficiency", "xxx devaluation", and "transportation mode";

[0378] Among them, the multiple to-be-triggered tags are divided into three groups according to their shapes, where the to-be-triggered tags in each group have the same display manner and balanced numbers, and each group has three to-be-triggered tags.

[0379] It should be pointed out that when the third trigger occurs on the update control of the first information expansion page, the multiple to-be-triggered tags in the second region change.

[0380] Referring to Figure 22B, the figure shows the first information expansion page corresponding to the third trigger of the update control on the first information expansion page. It should be noted that after the third trigger of the update control on the first information expansion page, the pending trigger labels included by the first information expansion page in the second area are specifically:

[0381] Capsule-shaped pending trigger labels: "xxx takes the initiative to resign", "xxx route", and "xx group";

[0382] Cloud-shaped pending trigger labels: "xxx route", "xxxx bridge", and "xxx opera house";

[0383] Rectangle-shaped pending trigger labels: "xxx devaluation", "xxx route transportation efficiency", and "xxx airport";

[0384] Among them, the number of updated pending trigger labels is the same as the number of pending trigger labels before the update. In this way, it can be ensured that the first information expansion page can display relatively rich pending trigger labels in the second area. The multiple pending trigger labels displayed in the second area can be divided into multiple groups according to their display methods, and the number of each group is balanced.

[0385] Refer to Figure 23 , according to some embodiments provided by the present disclosure, updating the pending trigger labels of each group so that the number of updated pending trigger labels of each group is the same as the number of pending trigger labels before the update may include, but is not limited to, the following steps 2310 to step 2340.

[0386] Step 2310, for the pending trigger labels of each group, determine the number of pending update labels in the group based on the number of pending trigger labels in the group and the second ratio;

[0387] Step 2320, for each candidate label in the candidate label set, obtain the first distance between the candidate label and the target label in the knowledge graph;

[0388] Step 2330, for each candidate label, obtain the first popularity of the candidate label;

[0389] Step 2340, based on the first distance and the first popularity, select the number of candidate labels equal to the number of pending update labels from the unselected candidate labels in the candidate label set for updating the pending trigger labels of the group.

[0390] The following describes steps 2310 to 2340.

[0391] It should be noted that, in order to enable the first information expansion page to display relatively rich triggerable tags in the second area, the multiple triggerable tags after the update in the second area can be divided into multiple groups according to their display methods, and the number of each group is balanced. However, after the update control is triggered for the third time, not necessarily all triggerable tags need to be replaced, and some of the triggerable tags can be retained to deepen their impression in the minds of viewers.

[0392] In step 2310 of some embodiments, for each group of triggerable tags, the number of tags to be updated in the group is determined based on the number of triggerable tags in the group and the second ratio. It should be noted that during the process of updating the triggerable tags, for each group of triggerable tags, the number of tags to be updated in the group can be determined based on the group and the second ratio. It should be pointed out that the second ratio is used to delimit the proportion of triggerable tags that need to be updated in the total number of triggerable tags in a group. Among them, the second ratio can be preset during the process of generating a group of triggerable tags; the second ratio can also be obtained from the object that triggers the third time when the triggerable tags need to be updated. Based on this, the number of tags to be updated in each group can be determined based on the number of triggerable tags in each group and the second ratio.

[0393] In step 2320 of some embodiments, for each candidate tag in the candidate tag set, the first distance between the candidate tag and the target tag in the knowledge graph is obtained. It should be noted that in the embodiments of the present disclosure, for each candidate tag in the candidate tag set, the first distance between the candidate tag and the trigger word in the knowledge graph can be obtained, aiming to quantify the relevance between the candidate tag and the trigger word through the first distance. If the first distance is shorter, it means that the relevance between the candidate tag and the trigger word is stronger.

[0394] In step 2330 of some embodiments, for each candidate tag, the first popularity of the candidate tag is obtained. It should be noted that the first popularity refers to the frequency of using the vocabulary corresponding to the candidate tag on the Internet. If the first popularity is higher, it means that the frequency of using the vocabulary corresponding to the candidate tag on the Internet is higher.

[0395] In step 2340 of some embodiments, based on the first distance and the first popularity, select a number of candidate tags equal to the number of tags to be updated from the unselected candidate tags in the candidate tag set for the tags to be triggered in the group. It should be noted that after determining the first distance between the candidate tag and the trigger word in the knowledge graph and the first popularity of the candidate tag, it is possible to further select a number of candidate tags equal to the number of tags to be updated from the unselected candidate tags in the candidate tag set for the tags to be triggered in the group based on the first distance and the first popularity. It should be pointed out that in some other embodiments of the present disclosure, it is possible to select a number of candidate tags equal to the number of tags to be updated from the unselected candidate tags in the candidate tag set based only on the first distance, or select a number of candidate tags equal to the number of tags to be updated from the unselected candidate tags in the candidate tag set based only on the first popularity. However, considering that the first distance is used to reflect the relevance between the candidate tag and the trigger word, and the first popularity is used to reflect the frequency of the vocabulary corresponding to the candidate tag used on the Internet, on this basis, considering both the factor of the relevance between the candidate tag and the trigger word and the factor of the frequency of the vocabulary corresponding to the candidate tag used on the Internet can more objectively reflect which candidate tags in the candidate tag set are more suitable for expanding and displaying the associated information of the trigger word, so as to identify the candidate tags with relatively high expansion value and then select them as the candidate tags to be updated.

[0396] According to some embodiments provided by the present disclosure, after step 320 of displaying the first information expansion page in response to the first trigger of the trigger word, the Internet corpus information expansion method may further include:

[0397] In response to a sliding operation on the second area, update a plurality of tags to be triggered. It should be noted that the tags to be triggered are candidate tags in the candidate tag set that have expansion value for expanding and displaying the associated information of the trigger word. Therefore, when the number of tags to be triggered that have expansion value for expanding and displaying the associated information of the trigger word is relatively large, it is difficult to fully display all of them in the second area of the first information expansion page within the fixed display range of the interface window. Based on this, the second area of the first information expansion page can be configured as an interface window that can be slid to update the plurality of tags to be triggered displayed in the interface window.

[0398] According to some embodiments provided by the present disclosure, the sliding operation is a sliding operation along the screen width direction of the interface displaying the second area;

[0399] Updating a plurality of tags to be triggered in response to a sliding operation on the second area includes:

[0400] Update the second number of to-be-triggered tags among multiple to-be-triggered tags, where the second number is associated with the ratio of the sliding path length of the sliding operation to the screen width. It should be noted that if the sliding path length of the sliding operation is longer, it means that for the object performing the third trigger, the number of to-be-triggered tags to be updated in the second region is more. Based on this, the second number can be associated with the ratio of the sliding path length of the sliding operation to the screen width, and on this basis, update the second number of to-be-triggered tags among multiple to-be-triggered tags to meet the corresponding update requirements of the third trigger for the to-be-triggered tags.

[0401] Refer to Figure 24A , which shows an example diagram of the first information expansion page in some embodiments of the present disclosure. It should be noted that there are multiple to-be-triggered tags in the second region of the first information expansion page;

[0402] Refer to Figure 24B , which shows an example diagram of the first information expansion page in some embodiments of the present disclosure. Since the second region of the first information expansion page is configured as an interface window where a sliding operation can be performed. Based on this, a sliding operation can be performed along the screen width direction of the interface displaying the second region to change the to-be-triggered tags displayed in the second region;

[0403] Refer to Figure 24C , which shows an example diagram of the first information expansion page in some embodiments of the present disclosure. After performing the sliding operation on the second region, the to-be-triggered tags displayed in the second region can be changed. In this way, when the number of to-be-triggered tags with extended value for the associated information of the trigger word is large, these to-be-triggered tags can be displayed in a way of being updated by the sliding operation.

[0404] Refer to Figure 25 , according to some embodiments provided by the present disclosure, updating the second number of to-be-triggered tags among multiple to-be-triggered tags may include, but is not limited to, the following steps 2510 to step 2540.

[0405] Step 2510, determine the second number based on the ratio of the sliding path length of the sliding operation to the screen width;

[0406] Step 2520, for each candidate tag in the candidate tag set, obtain the first distance between the candidate tag and the target tag in the knowledge graph;

[0407] Step 2530, for each candidate tag, obtain the first popularity of the candidate tag;

[0408] Step 2540, based on the first distance and the first popularity, select the second number of candidate tags from the unselected candidate tags in the candidate tag set to replace the second number of to-be-triggered tags among multiple to-be-triggered tags.

[0409] The following describes steps 2510 to 2540.

[0410] In step 2510 of some embodiments, a second number is determined based on the ratio of the sliding path length of the sliding operation to the screen width. It should be noted that the larger the ratio of the sliding path length of the sliding operation to the screen width, the more the number of tags to be triggered that need to be updated in the second region for the object performing the third trigger. In this case, the second number of tags to be triggered can be determined based on the ratio of the sliding path length of the sliding operation to the screen width.

[0411] In step 2520 of some embodiments, for each candidate tag in the candidate tag set, the first distance between the candidate tag and the target tag in the knowledge graph is obtained. It should be noted that since the first distance is used to quantify the relevance between the candidate tag and the trigger word, the shorter the first distance, the stronger the relevance between the candidate tag and the trigger word. Therefore, obtaining the first distance between the candidate tag and the target tag in the knowledge graph aims to use the relevance between the candidate tag and the trigger word as a consideration factor to update the tags to be triggered.

[0412] In step 2530 of some embodiments, for each candidate tag, the first popularity of the candidate tag is obtained. It should be noted that the first popularity refers to the frequency of using the vocabulary corresponding to the candidate tag on the Internet. The higher the first popularity, the higher the frequency of using the vocabulary corresponding to the candidate tag on the Internet. Therefore, for each candidate tag, obtaining the first popularity of the candidate tag aims to use the frequency of using the vocabulary corresponding to the candidate tag on the Internet as a consideration factor to update the tags to be triggered.

[0413] In step 2540 of some embodiments, based on the first distance and the first popularity, from the candidate tags that have not been selected in the candidate tag set, second number of candidate tags are selected to replace the second number of tags to be triggered among the multiple tags to be triggered. It should be noted that after determining the first distance and the first popularity, second number of candidate tags can be selected from the candidate tags that have not been selected in the candidate tag set on the basis of considering the relevance between the candidate tag and the trigger word and the frequency of using the vocabulary corresponding to the candidate tag on the Internet, and replace the second number of tags to be triggered among the multiple tags to be triggered. In this way, it can more objectively reflect which candidate tags in the candidate tag set are more suitable for updating the tags to be triggered.

[0414] Detailed description of step 330

[0415] Refer to Figure 26, according to some embodiments provided by the present disclosure, step 330 may include, but is not limited to, the following steps 2610 to 2620.

[0416] Step 2610, if the display mode of the target label in the second area is the same as that of the trigger word, display the common associated information of the target label and the trigger word in the third area of the second information expansion page;

[0417] Step 2620, display multiple to-be-triggered labels in the second area in the fourth area of the second information expansion page.

[0418] The following explains steps 2610 to 2620.

[0419] For step 2610 of some embodiments, if the display mode of the target label in the second area is the same as that of the trigger word, display the common associated information of the target label and the trigger word in the third area of the second information expansion page. It should be noted that the common associated information is the associated information formed to reflect the common characteristics between two words. If the display mode of the target label in the second area is the same as that of the trigger word, it means that there is a commonality between the target label and the trigger word. Based on this, the common characteristics between the target label and the trigger word can be displayed in the third area in the form of common associated information.

[0420] For step 2620 of some embodiments, display multiple to-be-triggered labels in the second area in the fourth area of the second information expansion page. It should be noted that in response to the second trigger of the target label among the multiple to-be-triggered labels, during the process of displaying the second information expansion page, the third area is used to display the common associated information of the target label and the trigger word, and the fourth area of the second information expansion page can inherit the second area of the first information expansion page to display the multiple to-be-triggered labels in the second area. In this way, other to-be-triggered labels that have expansion value for the trigger word can still be retained, facilitating the object browsing the web page to select from them, so as to further expand the understanding of other associated information of the trigger word, and improve the display efficiency of associated information and the effectiveness of displayed information. Other to-be-triggered labels that have expansion value for the trigger word can be retained, facilitating the object browsing the web page to select from them, so as to further expand the understanding of other associated information of the trigger word, and improve the display efficiency of associated information and the effectiveness of displayed information.

[0421] In some other embodiments of the present disclosure, during the process of displaying the second information expansion page, the associated information that is associated with both the target label and the trigger word can also be displayed through the fourth area of the second information expansion page. In this way, it is possible to further expand the understanding of other associated information that is associated with both the trigger word and the target label, and improve the display efficiency of associated information and the effectiveness of displayed information.

[0422] Refer toFigure 27A , which shows an example diagram of the first information expansion page in an embodiment of the present disclosure. It should be noted that in the second area of the first information expansion page, the target label "F-16 transport aircraft" and the trigger word "Type A aircraft" are displayed in the same way, both are displayed with a capsule-shaped outer frame.

[0423] Refer to Figure 27B , which shows an example diagram of the second information expansion page in an embodiment of the present disclosure. After the target label "F-16 transport aircraft", which is displayed in the same way as the trigger word "Type A aircraft", is subjected to the second trigger, the second information expansion page is immediately displayed. It should be noted that in the third area of the second information expansion page, the common associated information of the target label "F-16 transport aircraft" and the trigger word "Type A aircraft" is displayed: "Fourth-generation transport aircraft; both the Type A aircraft and the F-16 transport aircraft are fourth-generation multi-purpose high-performance transport aircraft, which have advanced electronic systems, radars, cargo hold configurations, and flight performance. These two transport aircraft have many commonalities in flight performance and transport functions, but there are also certain differences, such as R & D background and actual application scenarios, etc.". During the process of displaying the second information expansion page, the fourth area of the second information expansion page inherits the second area of the first information expansion page and displays multiple pending trigger labels in the second area. In this way, other pending trigger labels that are still valuable for expanding the trigger word "Type A aircraft" can be retained, facilitating the object browsing the web page to select from them to further expand the understanding of other associated information of the trigger word "Type A aircraft" and improve the display efficiency and effectiveness of the associated information.

[0424] Refer to Figure 27C , which shows an example diagram of the second information expansion page in an embodiment of the present disclosure. It should be noted that during the process of displaying the second information expansion page, the associated information that is associated with both the target label "F-16 transport aircraft" and the trigger word "Type A aircraft" can also be displayed through the fourth area of the second information expansion page. In this way, it is possible to further expand the understanding of other associated information that is associated with both the trigger word "Type A aircraft" and the target label "F-16 transport aircraft", and improve the display efficiency and effectiveness of the associated information.

[0425] Refer to Figure 28 , according to some embodiments provided by the present disclosure, in step 2610, displaying the common associated information of the target label and the trigger word in the third area of the second information expansion page may include, but is not limited to, the following steps 2810 to 2840.

[0426] Step 2810, determine the first guiding text according to the types of the target label and the trigger word;

[0427] Step 2820: Add the target label and the trigger word with the first guiding text and input them into the large language model to obtain the first detailed information of the common associated information;

[0428] Step 2830: Input the first detailed information into the Hunyuan model to obtain the first summary information of the common associated information;

[0429] Step 2840: In the third area, display the first summary information and the first detailed information as the common associated information.

[0430] The following explains Steps 2810 to 2840.

[0431] It should be noted that the common associated information is used to reflect the associated information formed by the common characteristics between two words. Since the large language model has powerful representation capabilities, the large language model can be used to process the target label and the trigger word to generate the common associated information corresponding to the target label and the trigger word.

[0432] In Steps 2810 to 2820 of some embodiments, according to the types of the target label and the trigger word, determine the first guiding text, add the target label and the trigger word with the first guiding text, and input them into the large language model to obtain the first detailed information of the common associated information. It should be noted that in order to generate the common associated information corresponding to the target label and the trigger word, the first guiding text can be determined according to the types of the target label and the trigger word. If the input text is randomly formulated during the process of conveying the preset requirements to the large language model, the input text will be difficult to conform to the expression paradigm of the large language model. Therefore, in some embodiments of the present disclosure, it is necessary to add the first guiding text that conforms to the expression paradigm of the large language model based on the target label and the trigger word, and further generate the text to be input into the large language model. In this way, a higher-quality first detailed information can be obtained during the process of applying the large language model to generate the first detailed information. Based on this, in order to respond to the first trigger of the trigger word, the embodiments of the present disclosure can add the first guiding text to the target label and the trigger word and input them into the large language model to obtain the first detailed information of the trigger word. Among them, the first detailed information is used to explain the commonality between the trigger word and the target label in detail.

[0433] In step 2830 of some embodiments, the first detailed information is input into the Hunyuan model to obtain the first summary information of the common association information. It should be noted that the Hunyuan model in the embodiments of the present disclosure is a natural language model used to extract the semantic content of a large text and obtain a refined expression to summarize the main content of the text. It should be pointed out that since the first detailed information is used to explain the commonality between the trigger word and the target label in detail, inputting the first detailed information into the Hunyuan model aims to use the Hunyuan model to extract the semantic content of the large text and obtain the first summary information corresponding to the trigger word. Among them, the first summary information is used to relatively refinedly summarize the detailed explanation of the commonality between the trigger word and the target label.

[0434] In step 2840 of some embodiments, in the third region, the first summary information and the first detailed information are displayed as the common association information. It should be noted that the first summary information is used to relatively refinedly summarize the detailed explanation of the commonality between the trigger word and the target label, and the first detailed information is used to explain the commonality between the trigger word and the target label in detail. Therefore, using the first summary information and the first detailed information as the common association information between the trigger word and the target label and displaying them in the third region can quickly display more effective information and improve the display efficiency and effectiveness of the association information.

[0435] Referring to Figure 29 , the trigger word "Type A aircraft" and the target label "F-16 transport aircraft" are combined with the first guiding text "Please explain the common features between [] and [] in detail" to form the text "Please explain the common features between 'Type A aircraft' and 'F-16 transport aircraft' in detail", which is input into the large language model to obtain the first detailed information of the trigger word: "Both Type A aircraft and F-16 transport aircraft are fourth-generation multi-purpose high-performance transport aircraft. They have advanced electronic systems, radars, cargo hold configurations, and flight performance. These two types of transport aircraft have many commonalities in flight performance and transport functions, but there are also certain differences, such as R & D background and actual application scenarios, etc.". Further, the first detailed information is processed by the Hunyuan model to extract the main content of the commonality between the trigger word "Type A aircraft" and the target label "F-16 transport aircraft", and the first summary information "Fourth-generation transport aircraft" is obtained. Furthermore, the above-obtained first summary information and first detailed information are displayed in the third region as the common association information between the trigger word "Type A aircraft" and the target label "F-16 transport aircraft".

[0436] Referring to Figure 30 , according to some embodiments provided by the present disclosure, the step of displaying the common association information of the target label and the trigger word in step 2610 may include, but is not limited to, the following steps 3010 to 3030.

[0437] Step 3010, if the types of the target label and the trigger word are both people or items, display the common characteristics of the people or items as the common association information;

[0438] Step 3020, if the types of the target label and the trigger word are both events, display the relationship between the events as the common association information;

[0439] Step 3030, if one of the types of the target label and the trigger word is people or items and the other is an event, display the role of the people or items in the event as the common association information.

[0440] The following describes Steps 3010 to 3030.

[0441] In Step 3010 of some embodiments, if the types of the target label and the trigger word are both people or items, display the common characteristics of the people or items as the common association information. It should be noted that if the types of the target label and the trigger word are both people or items, the common characteristics of the people corresponding to the target label and the trigger word, or the common characteristics of the items corresponding to the target label and the trigger word, can be obtained and these common characteristics can be displayed as the common association information.

[0442] In Step 3020 of some embodiments, if the types of the target label and the trigger word are both events, display the relationship between the events as the common association information. It should be noted that if the types of the target label and the trigger word are both events, the relationship between the events can be obtained and the relationship between the events can be displayed as the common association information.

[0443] In Step 3030 of some embodiments, if one of the types of the target label and the trigger word is people or items and the other is an event, display the role of the people or items in the event as the common association information. It should be noted that if one of the types of the target label and the trigger word is people or items while the other is an event, it means that the people or items exist in the corresponding event. On this basis, the role of the people or items in the event can be obtained and the role of the people or items in the event can be displayed as the common association information.

[0444] The embodiments of the present disclosure shown via Steps 3010 to 3030 determine the common association information between the target label and the trigger word based on the characteristics of the types of the target label and the trigger word. In this way, the determined common association information can more accurately reflect the common characteristics between the target label and the trigger word.

[0445] Refer to Figure 31 , according to some embodiments provided by the present disclosure, Step 330 may further include, but is not limited to, the following Steps 3110 to 3120.

[0446] Step 3110, if the display mode of the target label in the second area is different from that of the trigger word, display the second interpretation information of the target label in the third area of the second information expansion page;

[0447] Step 3120, display the updated multiple to-be-triggered labels in the fourth area of the second information expansion page.

[0448] In step 3110 of some embodiments, if the display mode of the target label in the second area is different from that of the trigger word, display the second interpretation information of the target label in the third area of the second information expansion page. It should be noted that if the display mode of the target label in the second area is different from that of the trigger word, it means that there is no commonality between the target label and the trigger word. Therefore, in order to explain the corresponding meaning of the target label, the second interpretation information can be determined based on the target label and then displayed in the third area of the second information expansion page. Among them, the second interpretation information is used to reveal the detailed information of the target label.

[0449] In step 3120 of some embodiments, display the updated multiple to-be-triggered labels in the fourth area of the second information expansion page. It should be noted that through the fourth area of the second information expansion page, the multiple to-be-triggered labels can be updated based on the target label, so that the updated multiple to-be-triggered labels are generated based on the associated information that can be extended corresponding to the target label.

[0450] In this way, through the second trigger of the to-be-triggered label, the second information expansion page can be used to further display the associated information of the corpus information, so as to quickly display more effective information and improve the display efficiency of the associated information and the effectiveness of the displayed information.

[0451] Refer to Figure 31A , which shows an example diagram of the first information expansion page in the embodiments of the present disclosure. It should be noted that in the second area of the first information expansion page, the display modes of the target label "xx route transportation efficiency" and the trigger word "Type A aircraft" are different. Among them, the target label "xx route transportation efficiency" has a rectangular outer shell, and the trigger word "Type A aircraft" has a capsule-shaped outer frame.

[0452] Refer to Figure 31B, which shows an example diagram of the second information expansion page in an embodiment of the present disclosure. After the target label "xx route transportation efficiency", which is different from the display method of the trigger word "Type A aircraft", is second-triggered, the second information expansion page is immediately displayed. It should be noted that in the third area of the second information expansion page, the second explanatory information of the target label "xx route transportation efficiency" is displayed: a schematic diagram of the recent monthly transportation efficiency of the xx route, and it is shown that "the transportation efficiency dropped to 0.0766 on June 23, and xxx Group is facing increasingly low transportation efficiency and decides to improve the transportation mode by purchasing Type A aircraft for operation". During the process of displaying the second information expansion page, multiple to-be-triggered labels updated based on the target label "xx route transportation efficiency" are displayed through the fourth area of the second information expansion page.

[0453] Referring to Figure 32 , according to some embodiments provided by the present disclosure, in step 3110, displaying the second explanatory information of the target label in the third area of the second information expansion page may include, but is not limited to, the following steps 3210 to 3230.

[0454] Step 3210, add the second guiding text to the target label and input it into the large language model to obtain the second detailed information of the target label;

[0455] Step 3220, input the second detailed information into the Hunyuan model to obtain the second summary information of the target label;

[0456] Step 3230, in the third area, display the second summary information and the second detailed information as the second explanatory information.

[0457] The following explains steps 3210 to 3230.

[0458] In step 3210 of some embodiments, add the second guiding text to the target label and input it into the large language model to obtain the second detailed information of the target label. It should be noted that if the input text is randomly formulated during the process of conveying the preset requirements to the large language model, the input text will be difficult to conform to the expression paradigm of the large language model. Therefore, in some embodiments of the present disclosure, it is necessary to add the first guiding text that conforms to the expression paradigm of the large language model on the basis of the target label to further generate the text to be input into the large language model. In this way, higher-quality second detailed information can be obtained when the large language model is applied to the generation of the second detailed information. Based on this, in order to respond to the first trigger of the target label, embodiments of the present disclosure may add the first guiding text to the target label and input it into the large language model to obtain the second detailed information of the target label.

[0459] In step 3220 of some embodiments, the second detailed information is input into the Hunyuan model to obtain the second summary information of the target label. It should be noted that the Hunyuan model in the embodiments of the present disclosure is a natural language model used to refine the semantic content of a large text to obtain a concise expression to summarize the main content of the text. It should be pointed out that since the second detailed information is used to explain the detailed meaning of the target label, inputting the second detailed information into the Hunyuan model aims to use the Hunyuan model to refine the semantic content of the large text to obtain the second summary information corresponding to the target label. Among them, the second summary information is used to relatively concisely summarize the meaning of the target label.

[0460] In step 3230, in the third area, the second summary information and the second detailed information are displayed as the second explanatory information. It should be emphasized that the second information expansion page has a third area and a fourth area. The third area contains the second explanatory information of the target label, and the second explanatory information is used to reveal the detailed information of the target label. Therefore, based on the fact that the second summary information is used to relatively concisely summarize the meaning of the target label and the second detailed information is used to explain the detailed meaning of the target label, the second summary information and the second detailed information can be displayed in the third area as the second explanatory information.

[0461] In the embodiments of the present disclosure shown by steps 3210 to 3230, on the basis of using the large language model to generate the second detailed information and using the Hunyuan model to generate the second summary information, the second summary information and the second detailed information are displayed as the second explanatory information in the first area, which can display the meaning of the target label in a detailed and appropriate manner, and helps the object browsing the web page to understand the information about the person, event, etc. pointed to by the target label.

[0462] Refer to Figure 33 , according to some embodiments provided by the present disclosure, step 3120 may include, but is not limited to, the following steps 3310 to 3330.

[0463] In step 3310, for each candidate label in the candidate label set, obtain the first distance between the candidate label and the target label in the knowledge graph;

[0464] In step 3320, for each candidate label, obtain the first popularity of the candidate label;

[0465] In step 3330, based on the first distance and the first popularity, select the updated to-be-triggered label from the candidate labels that have not been selected in the candidate label set.

[0466] The following explains steps 3310 to 3330.

[0467] In step 3310 of some embodiments, for each candidate label in the candidate label set, obtain the first distance between the candidate label and the target label in the knowledge graph. It should be noted that each candidate label in the candidate label set can correspond to an entity in the knowledge graph. Similarly, the target label also corresponds to an entity in the knowledge graph. In the embodiments of the present disclosure, for each candidate label in the candidate label set, the first distance between the candidate label and the target label in the knowledge graph can be obtained, aiming to quantify the relevance between the candidate label and the target label through the first distance. If the first distance is shorter, it means that the relevance between the candidate label and the target label is stronger.

[0468] In step 3320 of some embodiments, for each candidate label, obtain the first popularity of the candidate label. It should be noted that the first popularity refers to the frequency of using the vocabulary corresponding to the candidate label on the Internet. If the first popularity is higher, it means that the frequency of using the vocabulary corresponding to the candidate label on the Internet is higher.

[0469] In step 3330 of some embodiments, based on the first distance and the first popularity, select the updated trigger label to be triggered from the candidate labels that have not been selected in the candidate label set. It should be noted that after clarifying the first distance between the candidate label and the target label in the knowledge graph and clarifying the first popularity of the candidate label, it is possible to further determine the second popularity of the candidate label based on the first distance and the first popularity. It should be pointed out that in some other embodiments of the present disclosure, the trigger label to be triggered can be selected from the candidate label set only based on the first distance, or only based on the first popularity. However, considering that the first distance is used to reflect the relevance between the candidate label and the target label, and the first popularity is used to reflect the frequency of using the vocabulary corresponding to the candidate label on the Internet, on this basis, comprehensively considering the factor of the relevance between the candidate label and the target label and the factor of the frequency of using the vocabulary corresponding to the candidate label on the Internet can more objectively reflect which candidate labels in the candidate label set are more suitable for expanding and displaying the associated information of the target label, so as to clarify the candidate labels with more expansion value, and then select them as the updated trigger label to be triggered.

[0470] The embodiments of the present disclosure shown in steps 3310 to 3330 can determine the candidate labels suitable for expanding the key information of the associated words from the candidate label set, select these candidate labels with more expansion value as the updated trigger label to be triggered, which helps to further display the associated information corresponding to the corpus information, quickly display more effective information, and improve the display efficiency and effectiveness of the associated information.

[0471] Refer to Figure 34, step 3330 selects an updated to-be-triggered tag from the unselected candidate tags in the candidate tag set based on the first distance and the first heat, which may include, but are not limited to, the following steps 3410 to 3420.

[0472] Step 3410 selects a first number of to-be-triggered tags from the unselected candidate tags in the candidate tag set based on the first distance.

[0473] Step 3420 selects the remaining to-be-triggered tags among the multiple to-be-triggered tags from the unselected candidate tags in the candidate tag set based on the first heat.

[0474] The following describes steps 3410 to 3420.

[0475] In step 3410 of some embodiments, a first number of to-be-triggered tags are selected from the unselected candidate tags in the candidate tag set based on the first distance. It should be emphasized that the first distance between a candidate tag and a target tag in the knowledge graph is used to quantify the relevance between the candidate tag and the target tag. If the first distance is shorter, it means that the relevance between the candidate tag and the target tag is stronger. It should be clear that the number of to-be-triggered tags to be displayed in the fourth area of the second information expansion page is the first number, and the first number can be one or more than one. Among them, the first number is obtained based on the number of to-be-triggered tags and the first ratio. The to-be-triggered tags are selected from the candidate tag set based on the first distance and the first heat, and the first ratio is used to determine the proportion of the to-be-triggered tags selected based on the first distance among all the to-be-triggered tags to be displayed.

[0476] In some more specific embodiments, if 10 to-be-triggered tags need to be displayed in the fourth area of the second information expansion page, and the first ratio is 60%, then 6 out of the 10 to-be-triggered tags are selected from the candidate tag set based on the first distance. It should be understood that in the above embodiments, 6 candidate tags with the shortest first distance can be selected as the to-be-triggered tags according to the first distance between each candidate tag and the target tag in the knowledge graph.

[0477] In step 3420 of some embodiments, based on the first popularity, the remaining to-be-triggered tags among the multiple to-be-triggered tags are selected from the unselected candidate tags in the candidate tag set. It should be emphasized that the first popularity refers to the frequency of using the words corresponding to the candidate tags in the Internet. The higher the first popularity, the higher the frequency of using the words corresponding to the candidate tags in the Internet. It should be clear that the to-be-triggered tags are selected from the candidate tag set based on the first distance and the first popularity. If among all the to-be-triggered tags, a first number of to-be-triggered tags are selected from the candidate tag set based on the first distance, then based on the first popularity, the remaining to-be-triggered tags among the multiple to-be-triggered tags can be selected from the unselected candidate tags in the candidate tag set.

[0478] Through the embodiments of the present disclosure shown in steps 3410 to 3420, candidate tags suitable for expanding the key information of associated words can be determined from the candidate tag set, facilitating the object browsing the web page to master two types of information and improving the diversity of information display. Selecting these candidate tags with relatively high expansion value as the to-be-triggered tags helps to further display the associated information corresponding to the corpus information, quickly display more effective information, and improve the display efficiency of associated information and the effectiveness of displayed information.

[0479] Referring to Figure 35 , the first information expansion page further includes a fifth area for displaying the target tag sequence historically triggered by the target object that performs the first trigger. Among them, in the target tag sequence, the target tag most recently triggered by the target object is displayed in a first display manner different from the display manners of other target tags in the target tag training. After step 330 displays the second information expansion page in response to the second trigger of the target tag among the multiple to-be-triggered tags, the Internet corpus information expansion method may further include, but is not limited to, the following steps 3510 to 3520.

[0480] Step 3510, perform a sliding operation on the target tag sequence so that in the fifth area, the target tags in the target tag sequence are sequentially displayed in the first display manner;

[0481] Step 3520, display the second information expansion page corresponding to the target tag displayed in the first display manner in the fifth area.

[0482] The following explains steps 3510 to 3520.

[0483] It should be noted that the target tag sequence includes multiple target tags historically triggered by the target object, and the display manners of different target tags in the target tag sequence may be different. Among them, for different target tags with different display manners, the corresponding second information expansion pages are also different.

[0484] In step 3510 of some embodiments, a sliding operation is performed on the target label sequence, such that in the fifth region, the target labels in the target label sequence are sequentially displayed in the first display manner. It should be noted that the first information expansion page further includes a fifth region, in which a target label sequence of the target object's first trigger in the past is displayed. Based on this, by performing a sliding operation on the target label sequence, the target labels in the target label sequence can be sequentially displayed in the first display manner in the fifth region.

[0485] In step 3520 of some embodiments, a second information expansion page corresponding to the target label displayed in the first display manner in the fifth region is displayed. It should be noted that after the second trigger of the target label among the multiple to-be-triggered labels, the target object can trigger the target label displayed in the first display manner from the fifth region, and then display the corresponding second information expansion page.

[0486] In this way, the target object can quickly find the second information expansion page browsed in the past in the fifth region, so as to quickly display more effective information, improving the display efficiency of associated information and the effectiveness of the displayed information.

[0487] Referring to Figure 36A , an example diagram of the first information expansion page in some embodiments of the present disclosure is shown. It should be noted that a target label sequence of the target object's first trigger in the past is displayed in the fifth region of the first information expansion page. Among them, the target label sequence includes a target label "Type A aircraft & F16 transport aircraft" with a different display manner and a target label "Transport efficiency of xx route".

[0488] By performing a sliding operation on the target label sequence ( Figure 36A the arrow in

[0489] Referring to Figure 36B , an example diagram of the first information expansion page in some embodiments of the present disclosure is shown. In the order of the target label sequence, referring to Figure 36A the arrow in

[0490] Further, by continuing to perform a sliding operation on the target label sequence ( Figure 36B the arrow in

[0491] Referring toFigure 36C , which shows an example diagram of the first information expansion page in some embodiments of the present disclosure. In the order of the target tag sequence, referring to Figure 36A the direction of the sliding operation shown by the arrow in, the target tag "xx route transportation efficiency" is sequentially displayed. Based on this, the second information expansion page corresponding to the target tag "xx route transportation efficiency" is immediately retrieved and displayed.

[0492] Referring to Figure 37 , according to some embodiments provided by the present disclosure, step 3520 of displaying the second information expansion page corresponding to the target tag displayed in the fifth area in the first display manner may include, but is not limited to, the following steps 3710 to 3720.

[0493] Step 3710, obtain a browsing history database, where the browsing history database stores the second information expansion page corresponding to each target tag in the target tag sequence;

[0494] Step 3720, obtain from the browsing history database the second information expansion page corresponding to the target tag displayed in the fifth area in the first display manner and display it.

[0495] The following describes steps 3710 to 3720.

[0496] Steps 3710 to 3720 of some embodiments, obtain a browsing history database, where the browsing history database stores the second information expansion page corresponding to each target tag in the target tag sequence, and obtain from the browsing history database the second information expansion page corresponding to the target tag displayed in the fifth area in the first display manner and display it. It should be noted that the browsing history database is used to store the second information expansion page corresponding to each target tag in the target tag sequence. Therefore, in order to quickly display the second information expansion page corresponding to the target tag, the second information expansion page corresponding to the target tag displayed in the fifth area in the first display manner can be conveniently and quickly obtained from the browsing history database and displayed.

[0497] Description of the device embodiment of the present disclosure

[0498] Referring to Figure 38 , according to the Internet corpus information expansion device 3800 provided by some embodiments of the present disclosure, may include, but is not limited to:

[0499] A first display unit 3810, configured to display a target Internet corpus, where the target Internet corpus has a trigger word, and the trigger word is displayed in a display manner different from other parts of the target Internet corpus;

[0500] The second display unit 3820 is configured to display a first information expansion page in response to a first trigger on a trigger word. The first information expansion page has a first area and a second area. The first area contains first explanatory information about the trigger word, and the second area contains the trigger word and a plurality of to-be-triggered tags.

[0501] The third display unit 3830 is configured to display a second information expansion page in response to a second trigger on a target tag among the plurality of to-be-triggered tags. The second information expansion page is associated with the target tag.

[0502] Optionally, the third display unit 3830 is specifically configured to:

[0503] If the display manner of the target tag in the second area is the same as that of the trigger word, in a third area of the second information expansion page, display the common associated information of the target tag and the trigger word;

[0504] In a fourth area of the second information expansion page, display the plurality of to-be-triggered tags in the second area.

[0505] Optionally, the third display unit 3830 is specifically configured to:

[0506] Determine a first guiding text according to the types of the target tag and the trigger word;

[0507] Input the target tag, the trigger word, and the first guiding text into a large language model to obtain first detailed information about the common associated information;

[0508] Input the first detailed information into a Hunyuan model to obtain a first summary information about the common associated information;

[0509] In the third area, display the first summary information and the first detailed information as the common associated information.

[0510] Optionally, the third display unit 3830 is specifically configured to:

[0511] If the types of both the target tag and the trigger word are people or items, display the common characteristics of the people or items as the common associated information;

[0512] If the types of both the target tag and the trigger word are events, display the relationship between the events as the common associated information;

[0513] If one of the target tag and the trigger word is of the type of people or items and the other is of the type of events, display the role of the people or items in the events as the common associated information.

[0514] Optionally, the third display unit 3830 is specifically configured to:

[0515] If the display mode of the target label in the second area is different from that of the trigger word, in the third area of the second information expansion page, display the second explanation information of the target label;

[0516] In the fourth area of the second information expansion page, display the updated multiple tags to be triggered.

[0517] Optionally, the third display unit 3830 is specifically configured to:

[0518] Add the second guiding text to the target label, input it into the large language model, and obtain the second detailed information of the target label;

[0519] Input the second detailed information into the Hunyuan model to obtain the second summary information of the target label;

[0520] In the third area, display the second summary information and the second detailed information as the second explanation information.

[0521] Optionally, the third display unit 3830 is specifically configured to:

[0522] For each candidate label in the candidate label set, obtain the first distance between the candidate label and the target label in the knowledge graph;

[0523] For each candidate label, obtain the first popularity of the candidate label;

[0524] Based on the first distance and the first popularity, select the updated tags to be triggered from the candidate labels that have not been selected in the candidate label set.

[0525] Optionally, the third display unit 3830 is specifically configured to:

[0526] Based on the first distance, select the first number of tags to be triggered from the candidate labels that have not been selected in the candidate label set;

[0527] Based on the first popularity, select the remaining tags to be triggered among the multiple tags to be triggered from the candidate labels that have not been selected in the candidate label set.

[0528] Optionally, the first information expansion page has an update control for the multiple tags to be triggered; the Internet corpus information expansion device further includes:

[0529] The first update unit is configured to update the multiple tags to be triggered in response to a third trigger on the update control.

[0530] Optionally, the multiple tags to be triggered are divided into multiple groups, where the display modes of the tags to be triggered in each group are the same and the numbers are balanced;

[0531] The first update unit is specifically configured to:

[0532] Update the to-be-triggered tags of each group so that the number of updated to-be-triggered tags in each group is the same as the number of to-be-triggered tags before the update.

[0533] Optionally, the first update unit is specifically configured to:

[0534] For the to-be-triggered tags of each group, determine the number of to-be-updated tags in the group based on the number of to-be-triggered tags of the group and the second ratio;

[0535] For each candidate tag in the candidate tag set, obtain the first distance between the candidate tag and the target tag in the knowledge graph;

[0536] For each candidate tag, obtain the first popularity of the candidate tag;

[0537] Based on the first distance and the first popularity, select the number of candidate tags equal to the number of to-be-updated tags from the unselected candidate tags in the candidate tag set for updating the to-be-triggered tags of the group.

[0538] Optionally, the Internet corpus information expansion device further includes:

[0539] A second update unit, configured to update a plurality of to-be-triggered tags in response to a sliding operation on a second area.

[0540] Optionally, the sliding operation is a sliding operation along the screen width direction of the interface displaying the second area;

[0541] The second update unit is specifically configured to:

[0542] Update the second number of to-be-triggered tags among the plurality of to-be-triggered tags, where the second number is associated with the ratio of the sliding path length of the sliding operation to the screen width.

[0543] Optionally, the second update unit is specifically configured to:

[0544] Determine the second number based on the ratio of the sliding path length of the sliding operation to the screen width;

[0545] For each candidate tag in the candidate tag set, obtain the first distance between the candidate tag and the target tag in the knowledge graph;

[0546] For each candidate tag, obtain the first popularity of the candidate tag;

[0547] Based on the first distance and the first popularity, select the second number of candidate tags from the unselected candidate tags in the candidate tag set to replace the second number of to-be-triggered tags among the plurality of to-be-triggered tags.

[0548] Optionally, the first information expansion page further includes a fifth area for displaying the target label sequence triggered by the target object for the first trigger, where, in the target label sequence, the target label triggered by the target object most recently is displayed in a first display manner different from the display manners of other target labels in the target label training;

[0549] The Internet corpus information expansion device further includes:

[0550] A sliding operation unit for performing a sliding operation on the target label sequence, so that in the fifth area, the target labels in the target label sequence are sequentially displayed in the first display manner;

[0551] A fourth display unit for displaying a second information expansion page corresponding to the target label displayed in the first display manner in the fifth area.

[0552] Optionally, the fourth display unit is specifically configured to:

[0553] Obtain a browsing history database, where the browsing history database stores second information expansion pages corresponding to each target label in the target label sequence;

[0554] Obtain, from the browsing history database, the second information expansion page corresponding to the target label displayed in the first display manner in the fifth area and display it.

[0555] Optionally, the target Internet corpus is an Internet search result;

[0556] The first display unit 3810 is specifically configured to:

[0557] Obtain an Internet search request;

[0558] Display multiple Internet search results corresponding to the Internet search request, and the Internet search results are sorted according to the number of trigger words included.

[0559] Optionally, the first display unit 3810 is specifically configured to:

[0560] Obtain the target Internet corpus;

[0561] Segment the target Internet corpus and determine matching words that match the candidate labels in the candidate label set; [[ID=4l]]

[0562] Obtain the second popularity of the matching words;

[0563] Determine trigger words among the matching words based on the second popularity;

[0564] Display the target Internet corpus and display the trigger words in a display manner different from other parts of the target Internet corpus.

[0565] Optionally, the first display unit 3810 is specifically configured to:

[0566] Determine the first occurrence number of the matching word searched by the object from the Internet search records;

[0567] In the search results of the Internet search records, determine the second occurrence number of the matching word;

[0568] Based on the first occurrence number and the second occurrence number, determine the second popularity of the matching word.

[0569] Optionally, the second display unit 3820 is specifically configured to:

[0570] In response to the first trigger of the trigger word, input the trigger word plus the third guiding text into the large language model to obtain the third detailed information of the trigger word;

[0571] Input the third detailed information into the Hunyuan model to obtain the third summary information of the trigger word;

[0572] In the first area, display the third summary information and the third detailed information as the first explanatory information.

[0573] Optionally, the Internet corpus information expansion device further includes:

[0574] A to-be-triggered label selection unit, configured to select multiple to-be-triggered labels from the candidate label set in the following manner:

[0575] For each candidate label in the candidate label set, obtain the first distance between the candidate label and the trigger word in the knowledge graph;

[0576] For each candidate label, obtain the first popularity of the candidate label;

[0577] Based on the first distance and the first popularity, select the to-be-triggered labels from the candidate label set.

[0578] Optionally, the to-be-triggered label selection unit is specifically configured to:

[0579] Based on the first distance, select the first number of to-be-triggered labels from the candidate label set, where the first number is obtained based on the number of to-be-triggered labels and the first ratio;

[0580] Based on the first popularity, select the remaining to-be-triggered labels among the multiple to-be-triggered labels from the candidate labels not selected in the candidate label set.

[0581] Optionally, the to-be-triggered label selection unit is specifically configured to:

[0582] Based on the first distance, determine the first score of the to-be-triggered label;

[0583] Based on the first popularity, determine the second score of the to-be-triggered label;

[0584] Determine the total score of the label to be triggered based on the first score and the second score;

[0585] Select the label to be triggered from the candidate label set based on the total score.

[0586] Optionally, the candidate label set includes candidate labels and display modes corresponding to the candidate labels;

[0587] The Internet corpus information expansion device further includes:

[0588] A candidate label set generation unit, configured to generate a candidate label set in the following manner:

[0589] Obtain a first corpus sample library;

[0590] Segment the first corpus sample library, and use each word obtained as a candidate label and add it to the candidate label library;

[0591] Determine the display mode of the candidate label based on at least one of the knowledge graph, the object behavior data of the corpus sample having the candidate label in the first corpus sample library, and the third popularity of the corpus sample, and store it in the candidate label set corresponding to the candidate label;

[0592] Display a first information expansion page, including: displaying the label to be triggered in a second area in the display mode corresponding to the candidate label in the candidate label set.

[0593] Optionally, the candidate label set generation unit is specifically configured to:

[0594] For any two candidate labels in the candidate label set, obtain the target corpus that commonly contains the two candidate labels in the first corpus sample library;

[0595] If the target corpus has been accessed by the target object that has been first triggered, the third popularity of the target corpus is greater than the first threshold, and the first distance between the two candidate labels in the knowledge graph is less than the second threshold, then assign the same display mode to the two candidate labels.

[0596] Optionally, the candidate label set generation unit is specifically configured to:

[0597] For any two candidate labels in the candidate label set, obtain the relationship between the two candidate labels in the knowledge graph;

[0598] Determine the display mode of the two candidate labels based on the type of the relationship.

[0599] Optionally, the candidate label set generation unit is specifically configured to:

[0600] For any two candidate tags in the candidate tag set, if the two candidate tags are associated with the same event in the knowledge graph, the temporal proximity of the two corpora with the two candidate tags in the first corpus sample library meets the first condition, and the two candidate tags have the same type, then the two candidate tags will be assigned the same display mode.

[0601] Optionally, the candidate tag set generating unit is specifically used to:

[0602] Get the threshold number of candidate tags for each display mode;

[0603] If, based on at least one of the knowledge graph, the object behavior data of the corpus sample with the candidate label in the first corpus sample library, and the third heat of the corpus sample, after the display mode is assigned to the candidate label, the number of candidate labels in the display mode exceeds the candidate label number threshold, the display mode assigned to the candidate label is cancelled.

[0604] Optionally, the first display unit 3810 is specifically configured to:

[0605] Determine the first distance between the tag to be triggered and the trigger word in the knowledge graph;

[0606] Based on the first distance, a plurality of to-be-triggered tags are displayed, so that a second distance between the to-be-triggered tag and the trigger word in the second area is associated with the first distance.

[0607] Reference Figure 39 , Figure 39 The following is a block diagram of the structure of the target terminal 110 for implementing the Internet corpus information expansion method according to an embodiment of the present disclosure. The terminal includes: a radio frequency (RF) circuit 3910, a memory 3915, an input unit 3930, a display unit 3940, a sensor 3950, an audio circuit 3960, a wireless fidelity (WiFi) module 3970, a processor 3980, and a power supply 3990. It will be understood by those skilled in the art that Figure 39 The structure of the target terminal 110 shown does not constitute a limitation on a mobile phone or a computer, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0608] The RF circuit 3910 may be used for receiving and sending signals during information transmission or calls. In particular, after receiving downlink information from the base station, it is sent to the processor 3980 for processing. In addition, the designed uplink data is sent to the base station.

[0609] The memory 3915 can be used to store software programs and modules. The processor 3980 executes various functional applications and data processing of the content terminal by running the software programs and modules stored in the memory 3915.

[0610] The input unit 3930 can be used to receive input digital or character information, and generate key signal inputs related to the settings and function controls of the content terminal. Specifically, the input unit 3930 can include a touch panel 3931 and other input devices 3932.

[0611] The display unit 3940 can be used to display the input information or provided information, as well as various menus of the content terminal. The display unit 3940 can include a display panel 2941.

[0612] The audio circuit 3960, speaker 3961, and microphone 3962 can provide an audio interface.

[0613] In this embodiment, the processor 3980 included in the object terminal 110 can execute the Internet corpus information expansion method of the previous embodiment.

[0614] The object terminal 110 of the embodiments of the present disclosure includes, but is not limited to, mobile phones, computers, intelligent voice interaction devices, intelligent home appliances, vehicle-mounted terminals, aircraft, etc. The embodiments of the present disclosure can be applied to various scenarios, including but not limited to content recommendation, data screening, etc.

[0615] Figure 40 It is a structural block diagram of a part of the Internet corpus information expansion server 140 for implementing the Internet corpus information expansion method of the embodiments of the present disclosure. The Internet corpus information expansion server 140 can vary greatly due to configuration or performance differences, and can include one or more central processing units (CPUs) 4022 (for example, one or more processors) and a memory 4032, and one or more storage media 4030 (for example, one or more mass storage devices) storing application programs 4042 or data 4044. Among them, the memory 4032 and the storage media 4030 can be transient storage or persistent storage. The programs stored in the storage media 4030 can include one or more modules (not shown in the figure), and each module can include a series of instruction operations on the server. Further, the central processor 4022 can be set to communicate with the storage media 4030 and execute a series of instruction operations in the storage media 4030 on the server.

[0616] The Internet corpus information expansion server 140 may also include one or more power supplies 4026, one or more wired or wireless network interfaces 4050, one or more input / output interfaces 4058, and / or one or more operating systems 4041, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSD TM, and so on.

[0617] The central processing unit 2722 in the Internet corpus information expansion server 140 may be used to execute the Internet corpus information expansion method of the embodiments of the present disclosure.

[0618] The embodiments of the present disclosure also provide a computer-readable storage medium for storing program codes for executing the Internet corpus information expansion method of the foregoing various embodiments.

[0619] The embodiments of the present disclosure also provide a computer program product including a computer program. The processor of the computer device reads and executes the computer program, so that the computer device implements the above-mentioned Internet corpus information expansion method.

[0620] Terms such as "first", "second", "third", "fourth", etc. (if any) in the specification of the present disclosure and the above-mentioned drawings are used to distinguish similar contents and do not necessarily need to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present disclosure described here can be implemented in an order other than those illustrated or described here. In addition, the terms "include" and "comprise" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0621] It should be understood that in the present disclosure, "at least one (item)" means one or more, and "a plurality" means two or more. "And / or" is used to describe the relationship between related content, indicating that there can be three relationships. For example, "A and / or B" can mean: only A exists, only B exists, and both A and B exist at the same time. Here, A and B can be singular or plural. The character " / " generally indicates that the related content before and after is an "or" relationship. "At least one (one) of the following" or a similar expression means any combination of these items, including any combination of single items (ones) or plural items (ones). For example, at least one (one) of a, b, or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0622] It should be understood that in the description of the embodiments of the present disclosure, the meaning of "a plurality (or multiple items)" is more than two. Understandings such as greater than, less than, exceeding, etc. do not include the present number, and understandings such as above, below, within, etc. include the present number.

[0623] In several embodiments provided by the present disclosure, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of units is only a logical function division. In actual implementation, there can be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection between each other can be through some interfaces, and the indirect coupling or communication connection of devices or units can be in electrical, mechanical, or other forms.

[0624] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0625] In addition, in the embodiments of the present disclosure, the term "module" or "unit" refers to a computer program with a predetermined function or a part of a computer program, and works with other related parts to achieve a predetermined goal, and can be fully or partially implemented by using software, hardware (such as a processing circuit or a memory), or a combination thereof. Similarly, one processor (or multiple processors or memories) can be used to implement one or more modules or units. In addition, each module or unit can be a part of the overall module or unit that includes the function of the module or unit.

[0626] When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present disclosure, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present disclosure. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical discs that can store program codes.

[0627] It should also be understood that the various embodiments provided in the present disclosure can be combined arbitrarily to achieve different technical effects.

[0628] The above is a specific description of the embodiments of the present disclosure, but the present disclosure is not limited to the above embodiments. Those skilled in the art can also make various equivalent deformations or substitutions without departing from the spirit of the present disclosure, and these equivalent deformations or substitutions are all included within the scope defined by the claims of the present disclosure.

Claims

1. An Internet corpus information expansion method, characterized in that The method includes: Displaying a target Internet corpus, where the target Internet corpus has a trigger word, and the trigger word is displayed in a display manner different from other parts of the target Internet corpus; In response to a first trigger on the trigger word, displaying a first information expansion page, where the first information expansion page has a first area and a second area, the first area contains first explanatory information about the trigger word, and the second area contains the trigger word and a plurality of to-be-triggered tags; In response to a second trigger on a target tag among the plurality of to-be-triggered tags, displaying a second information expansion page, where the second information expansion page is associated with the target tag.

2. The method for expanding Internet corpus information according to claim 1, wherein The step of, in response to a second trigger on a target tag among the plurality of to-be-triggered tags, displaying a second information expansion page includes: If the display manner of the target tag in the second area is the same as that of the trigger word, in a third area of the second information expansion page, displaying common associated information of the target tag and the trigger word; In a fourth area of the second information expansion page, displaying the plurality of to-be-triggered tags in the second area.

3. The method for expanding Internet corpus information according to claim 2, wherein The step of displaying the common associated information of the target tag and the trigger word includes: If the types of both the target tag and the trigger word are a person or an item, displaying common characteristic features of the person or the item as the common associated information; If the types of both the target tag and the trigger word are events, displaying the relationship between the events as the common associated information; If one of the target tag and the trigger word is of the type of a person or an item and the other is of the type of an event, displaying the role of the person or the item in the event as the common associated information.

4. The method for expanding Internet corpus information according to claim 1, wherein The step of, in response to a second trigger on a target tag among the plurality of to-be-triggered tags, displaying a second information expansion page further includes: If the display manner of the target tag in the second area is different from that of the trigger word, in a third area of the second information expansion page, displaying second explanatory information about the target tag; In a fourth area of the second information expansion page, displaying the updated plurality of to-be-triggered tags.

5. The method for expanding Internet corpus information according to claim 1, wherein The first information expansion page has an update control for the plurality of to-be-triggered tags; After displaying the first information expansion page in response to the first trigger on the trigger word, the Internet corpus information expansion method further includes: in response to a third trigger on the update control, updating the plurality of to-be-triggered tags.

6. The method for expanding Internet corpus information according to claim 5, characterized in that The plurality of to-be-triggered tags are divided into multiple groups, where the display manner of the to-be-triggered tags in each group is the same and the number is balanced; The step of updating the plurality of to-be-triggered tags includes: updating the to-be-triggered tags in each group so that the number of the updated to-be-triggered tags in each group is the same as the number of the to-be-triggered tags before the update.

7. The method for expanding Internet corpus information according to claim 1, wherein After displaying the first information expansion page in response to the first trigger on the trigger word, the Internet corpus information expansion method further includes: In response to a sliding operation on the second area, updating the plurality of to-be-triggered tags.

8. The method for expanding Internet corpus information according to claim 7, wherein The sliding operation is a sliding operation along the screen width direction of the interface where the second area is displayed; In response to a sliding operation on the second region, updating a plurality of the tags to be triggered includes: Updating a second number of the tags to be triggered among the plurality of the tags to be triggered, where the second number is associated with a ratio of a sliding path length of the sliding operation to a screen width.

9. The method for expanding Internet corpus information according to claim 1, wherein The first information expansion page further includes a fifth region for displaying the target tag sequence historically triggered by the target object for the first trigger, where, in the target tag sequence, the target tag most recently triggered by the target object is displayed in a first display manner different from that of other target tags in the target tag training; After displaying a second information expansion page in response to a second trigger on a target tag among the plurality of the tags to be triggered, the internet corpus information expansion method further includes: Performing a sliding operation on the target tag sequence such that, in the fifth region, the target tags in the target tag sequence are sequentially displayed in the first display manner; Displaying the second information expansion page corresponding to the target tag displayed in the first display manner in the fifth region.

10. The method for expanding Internet corpus information according to claim 1, wherein The displaying of the target internet corpus includes: Obtaining the target internet corpus; Segmenting the target internet corpus and determining matching words among the segmented words that match candidate tags in a candidate tag set; Obtaining a second popularity of the matching words; Determining the trigger words among the matching words based on the second popularity; Displaying the target internet corpus and displaying the trigger words in a display manner different from other parts of the target internet corpus.

11. The method for expanding Internet corpus information according to claim 1, wherein The displaying of the first information expansion page in response to a first trigger on the trigger words includes: In response to a first trigger on the trigger words, inputting the trigger words plus a third guiding text into a large language model to obtain third detailed information of the trigger words; Inputting the third detailed information into a Hunyuan model to obtain third summary information of the trigger words; In the first region, displaying the third summary information and the third detailed information as the first explanatory information.

12. The method for expanding Internet corpus information according to claim 1, wherein The plurality of the tags to be triggered are selected from the candidate tag set in the following manner: For each candidate tag in the candidate tag set, obtaining a first distance between the candidate tag and the trigger word in a knowledge graph; For each candidate tag, obtaining a first popularity of the candidate tag; Selecting the tags to be triggered from the candidate tag set based on the first distance and the first popularity.

13. The method for expanding Internet corpus information according to claim 12, wherein The selecting of the tags to be triggered from the candidate tag set based on the first distance and the first popularity includes: Selecting a first number of the tags to be triggered from the candidate tag set based on the first distance, where the first number is obtained based on a number of the tags to be triggered and a first ratio; Selecting the remaining tags to be triggered among the plurality of the tags to be triggered from the candidate tags not selected from the candidate tag set based on the first popularity.

14. The method for expanding Internet corpus information according to claim 12, wherein The candidate tag set includes the candidate tags and display manners corresponding to the candidate tags; The candidate tag set is generated in the following manner: Obtain a first corpus sample library; Segment the first corpus sample library, and each of the segmented words is used as the candidate tag and added to the candidate tag library; Based on at least one of the knowledge graph, the object behavior data of the corpus samples having the candidate tags in the first corpus sample library, and the third popularity of the corpus samples, determine the display mode of the candidate tags, and store it corresponding to the candidate tags in the candidate tag set; The display of the first information expansion page includes: displaying the to-be-triggered tag in the second area in the display mode corresponding to the candidate tag in the candidate tag set.

15. The method for expanding Internet corpus information according to claim 14, wherein The determining the display mode of the candidate tags based on at least one of the knowledge graph, the object behavior data of the corpus samples having the candidate tags in the first corpus sample library, and the third popularity of the corpus samples includes: For any two candidate tags in the candidate tag set, obtain the target corpus in the first corpus sample library that contains both candidate tags; If the target corpus has been accessed by the target object that performs the first trigger, the third popularity of the target corpus is greater than the first threshold, and the first distance between the two candidate tags in the knowledge graph is less than the second threshold, then assign the same display mode to the two candidate tags.

16. The method for expanding Internet corpus information according to claim 14, wherein The determining the display mode of the candidate tags based on at least one of the knowledge graph, the object behavior data of the corpus samples having the candidate tags in the first corpus sample library, and the third popularity of the corpus samples includes: For any two candidate tags in the candidate tag set, if the two candidate tags are co-associated with the same event in the knowledge graph, the time proximity in the two corpus samples respectively having the two candidate tags in the first corpus sample library meets the first condition, and the two candidate tags have the same type, then assign the same display mode to the two candidate tags.

17. An Internet corpus information expansion device, characterized in that, Includes: A first display unit for displaying a target Internet corpus, where the target Internet corpus has a trigger word, and the trigger word is displayed in a display mode different from other parts of the target Internet corpus; A second display unit for, in response to a first trigger on the trigger word, displaying a first information expansion page, where the first information expansion page has a first area and a second area, the first area contains first explanation information about the trigger word, and the second area contains the trigger word and a plurality of to-be-triggered tags; A third display unit for, in response to a second trigger on a target tag among the plurality of to-be-triggered tags, displaying a second information expansion page, where the second information expansion page is associated with the target tag.

18. An electronic device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the Internet corpus information expansion method according to any one of claims 1 to 16.

19. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the Internet corpus information expansion method according to any one of claims 1 to 16.

20. A computer program product, which includes a computer program that is read and executed by a processor of a computer device, so that the computer device executes the method for expanding Internet corpus information according to any one of claims 1 to 16.

Images