Information search methods, apparatus, electronic device and storage medium

Abstract

The present application relates to the technical field of Internet, in particular to information search methods, an apparatus, an electronic device and a storage medium, which are used for improving the information search efficiency. A method comprises: displaying a content detail interface of target content; in response to a tagging operation triggered for at least one content fragment within the target content, displaying within the content detail interface tagging identifiers respectively corresponding to the at least one content fragment, and at least one piece of recommended search information corresponding to a current tagging operation, the recommended search information being generated on the basis of fragment detail information contained in content fragments tagged by current and previous tagging operations; and in response to a search operation triggered on the basis of target recommended search information among the at least one recommended search information, displaying a search result corresponding to the target recommended search information. The present application supports content fragment tagging by subjects, and intelligently generates recommended search information on this basis, so as to satisfy search requirements of subjects, thus improving the search efficiency.

Description

Information search method, device, electronic device and storage medium

[0001] This application claims priority to the Chinese patent application filed with the China Patent Office on September 8, 2023, with application number 2023111626535 and application name “An information search method, device, electronic device and storage medium”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present application relates to the field of Internet technology, and in particular to information search technology. Background Art

[0003] With the popularization of computers and the development of the Internet, people use the Internet more and more frequently. Computers and the Internet have gradually become indispensable tools in people's daily lives. Internet search services have been widely used in people's daily lives because they can provide objects with all kinds of information and data, bringing great convenience to people.

[0004] Taking the browser scenario as an example, in the current browser search environment, the object needs to manually enter the search keyword, and the browser search engine mainly relies on the search keyword entered by the object to provide search results.

[0005] In this method, the subject needs to spend time thinking and entering search keywords, which may cause information delays when the subject needs to obtain information quickly. Secondly, because the selection and input method of search keywords may affect the accuracy of search results, the subject may need to try entering different search keywords multiple times and perform multiple searches before searching for information that meets their needs. It can be seen that the information search method in the related art is overly dependent on the search keywords entered by the subject, resulting in low search efficiency.

[0006] In summary, how to improve the efficiency of information search is an urgent problem to be solved.

[0007] Summary of the Invention

[0008] Embodiments of the present application provide an information search method, device, electronic device, and storage medium to improve information search efficiency.

[0009] An information search method provided in an embodiment of the present application is performed by an electronic device and includes:

[0010] Display the content details interface of the target content;

[0011] In response to a mark operation triggered on at least one content segment in the target content, displaying a mark identifier corresponding to each of the at least one content segment and at least one recommended search information corresponding to the current mark operation in the content details interface; the recommended search information is generated based on the segment detail information included in each content segment marked by the current and previous mark operations;

[0012] In response to a search operation triggered based on target recommended search information in the at least one recommended search information, search results corresponding to the target recommended search information are displayed.

[0013] Another information search method provided in an embodiment of the present application is performed by an electronic device and includes:

[0014] Obtaining content segments marked in target content; the content segments are determined by the client in response to a marking operation triggered on the target content, and each marking operation corresponds to marking at least one content segment;

[0015] Generating at least one recommended search information corresponding to the marking operation by performing semantic analysis on the fragment details information contained in each content fragment marked this time and before this time;

[0016] The at least one recommended search information is sent to the client, so that the client displays the mark identifier corresponding to each of the at least one content segment and the at least one recommended search information corresponding to the marking operation in the content details interface of the target content.

[0017] An information search device provided in an embodiment of the present application includes:

[0018] A display unit, used to display a content details interface of the target content;

[0019] a first response unit configured to, in response to a mark operation triggered on at least one content segment in the target content, display, in the content details interface, a mark identifier corresponding to each of the at least one content segment and at least one recommended search information corresponding to the current mark operation; the recommended search information being generated based on segment detail information included in each content segment marked by the current and previous mark operations;

[0020] The second response unit is configured to, in response to a search operation triggered based on target recommended search information in the at least one recommended search information, display search results corresponding to the target recommended search information.

[0021] Another information search device provided in an embodiment of the present application includes:

[0022] a determining unit configured to obtain content segments marked in the target content; the content segments are determined by the client in response to a marking operation triggered on the target content, and each marking operation corresponds to marking at least one content segment;

[0023] A generating unit, which generates at least one recommended search information corresponding to the marking operation by performing semantic analysis on the segment detail information contained in each content segment marked this time and before this time;

[0024] The feedback unit sends the at least one recommended search information to the client, so that the client displays the mark identifier corresponding to each of the at least one content segment and the at least one recommended search information corresponding to the marking operation in the content details interface of the target content.

[0025] An electronic device provided by an embodiment of the present application includes a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of any one of the above-mentioned information search methods.

[0026] An embodiment of the present application provides a computer-readable storage medium, which includes a computer program. When the computer program is run on an electronic device, the computer program is used to enable the electronic device to perform the steps of any one of the above-mentioned information search methods.

[0027] An embodiment of the present application provides a computer program product, which includes a computer program stored in a computer-readable storage medium; when a processor of an electronic device reads the computer program from the computer-readable storage medium, the processor executes the computer program, causing the electronic device to perform the steps of any one of the above-mentioned information search methods.

[0028] The beneficial effects of this application are as follows:

[0029] The present application provides an information search method, device, electronic device, and storage medium. In the present application, when browsing the content details of target content, the subject can directly mark content segments in the content details interface. For example, the subject can mark content segments that he or she is interested in and wants to explore, or content segments that he or she is confused about and wants to explore, etc., according to his or her needs. Furthermore, the present application can intelligently generate recommended search information based on the segment details information contained in each content segment that the subject has marked in this browsing process and before, and display it to the subject. Since these recommended search information are generated based on the analysis of the content segments marked by the subject, they can effectively reflect the subject's search needs. In addition, the subject can further search for information based on the target recommended search information in the displayed recommended search information. In this search process, the subject only needs to mark the content segments according to his or her own search needs, without having to spend time thinking and manually entering search keywords, which can effectively improve the efficiency of the search. In addition, in this way, on the basis of better meeting the subject's search needs, it can also help the subject discover and obtain more relevant information, improve the subject's search experience, and increase the richness of the information obtained by the subject.

[0030] Other features and advantages of the present application will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present application. The purposes and other advantages of the present application can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0032] FIG1 is a schematic diagram of an application scenario of an embodiment of the present application;

[0033] FIG2 is a flowchart of an implementation of an information search method provided in an embodiment of the present application;

[0034] FIG3 is a schematic diagram of a content details interface in an embodiment of the present application;

[0035] FIG4 is a schematic diagram of another content details interface in an embodiment of the present application;

[0036] FIG5 is a schematic diagram of a first process of marking content segments in an embodiment of the present application;

[0037] FIG6 is a schematic diagram of a second process of marking content segments in an embodiment of the present application;

[0038] FIG7 is a schematic diagram of a third process of marking content segments in an embodiment of the present application;

[0039] FIG8 is a schematic diagram of recommended search information in an embodiment of the present application;

[0040] FIG9 is a schematic diagram of another method of marking a content segment in an embodiment of the present application;

[0041] FIG10 is a schematic diagram of a process of performing information search based on target recommendation search information in an embodiment of the present application;

[0042] FIG11 is a schematic diagram of a first search information set provided in an embodiment of the present application;

[0043] FIG12 is a schematic diagram of another first search information set provided in an embodiment of the present application;

[0044] FIG13 is a schematic diagram of a process of performing information search based on target first search information in an embodiment of the present application;

[0045] FIG14 is a schematic diagram of second search information provided in an embodiment of the present application;

[0046] FIG15 is a schematic diagram of a process of performing autonomous splicing search based on second search information provided by an embodiment of the present application;

[0047] FIG16 is a schematic diagram of a third search information in an embodiment of the present application;

[0048] FIG17 is a schematic diagram of another third search information in an embodiment of the present application;

[0049] FIG18 is a schematic diagram of yet another third search information in an embodiment of the present application;

[0050] FIG19 is a schematic diagram of presenting search results based on third search information in an embodiment of the present application;

[0051] FIG20 is a schematic diagram of another content details interface in an embodiment of the present application;

[0052] FIG21 is a schematic diagram of a switch control in an embodiment of the present application;

[0053] FIG22 is a flowchart of an implementation of an information search method in an embodiment of the present application;

[0054] FIG23 is a schematic diagram of a logic for generating recommended search information in an embodiment of the present application;

[0055] FIG24 is a schematic diagram of the structure of an information search device according to an embodiment of the present application;

[0056] FIG25 is a schematic diagram of the structure of another information search device in an embodiment of the present application;

[0057] FIG26 is a schematic diagram of a hardware structure of an electronic device to which an embodiment of the present application is applied;

[0058] FIG27 is a schematic diagram of a hardware structure of another electronic device to which an embodiment of the present application is applied. DETAILED DESCRIPTION

[0059] To make the purpose, technical solutions, and advantages of the embodiments of this application more clear, the technical solutions of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of the technical solutions of this application, but not all of them. Based on the embodiments described in this application document, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the technical solutions of this application.

[0060] The following is an introduction to some concepts involved in the embodiments of this application.

[0061] Marking identifier: In the present application, when an object browses the content details of the target content, it can directly mark it in real time in the content details interface. For example, the object can mark the content segments that it is interested in, or the content segments that it is confused about, etc. according to its own needs; the marking identifier refers to the identifier that appears in the content details interface when the object marks the content segment. It is used to indicate that the content segment has been marked. The identifier corresponds to the content segment. For example, the marking identifier may include at least one of the following: a selection mark for identifying the selected content segment, and a marking serial number.

[0062] Recommended search information: refers to search information generated based on the content details contained in the content fragments marked by the object, which is used as basic information for subsequent searches. Specifically, it can be search keywords, search statements, etc. The object can select recommended search information for intelligent search and view the corresponding search results.

[0063] Switch control: refers to the control used to turn on or off the smart tag search function. The smart tag search function is a function that generates recommended search information based on the content fragments tagged by the object, and then performs intelligent search based on this. After the control is turned on, the content details interface can support object tagging, and intelligently generate recommended search information based on the content fragments tagged by the object. Specifically, after the control is turned on, the search sub-interface can be presented in the content details interface, and the object can be tagged in the content details interface. After the object completes at least one tag, the search sub-interface will present the corresponding recommended search information; after the control is turned off, the search sub-interface can be canceled in the content details interface, and all tag information of the object can be cleared.

[0064] First search information set: In this application, it refers to a set of intelligent search information (which may be referred to as the intelligent search information set). Each content segment marked by the object in the target content corresponds to a first search information set. Each first search information set includes at least one first search information. These first search information are intelligently generated based on the content segment marked by the object. Specifically, each time the object marks a content segment, a corresponding first search information set is generated. The first search information in the first search information set includes: at least one recommended search information and candidate search information generated based on the content segments marked by the object this time and before.

[0065] Second search information: In this application, it refers to the search information that can be independently spliced ​​by the object. This search information is also intelligently generated based on the content fragments marked by the object. However, the second search information is different from the first search information. The second search information is recommended search information generated based on the content fragments marked by the object this time.

[0066] Third search information: in this application, refers to the reverse search information corresponding to the first search information. Specifically, the third search information and the first search information, which are reverse search information, have opposite semantics in the search results they correspond to.

[0067] In an embodiment of the present application, semantic analysis can be performed on the content detail information contained in the tag information based on natural language processing technology and machine learning technology to generate corresponding recommended search information, first search information, second search information, and third search information, etc., so that the object can perform intelligent search based on these automatically generated search information to improve information search efficiency.

[0068] Furthermore, the information search methods in the embodiments of the present application also involve database technology. For example, in the embodiments of the present application, when verifying the authenticity of search results corresponding to recommended search information, an authoritative database may be queried. Furthermore, the various types of recommended search information generated in the embodiments of the present application may also be stored in a database for subsequent use, and so on.

[0069] The following is a brief introduction to the design concept of the embodiment of this application:

[0070] In current online search services, almost all search engines support a search method in which the subject enters search keywords, and the search engine returns results related to the search keywords. The search engine uses natural language processing technology to understand the subject's query input, helping the search engine to better understand the subject's needs, and sort the search results based on various factors (such as the authority of the page, the relevance of the content, the subject's search history, etc.), thereby forming a relevance ranking. Under this method, the search engine mainly relies on the search keywords entered by the subject to provide search results, and the subject needs to spend time thinking and entering the search keywords. Secondly, since the choice and input method of the search keywords may affect the accuracy of the search results, the subject may need to try to enter different search keywords multiple times and make multiple attempts to find the information that meets the needs, resulting in low search efficiency.

[0071] Furthermore, if subjects encounter multiple interesting contents while browsing a web page or reading a document, they may need to remember these contents and search for them separately at a later time, which is neither convenient nor efficient.

[0072] At the same time, users may have new needs in this situation. For example, they may want to be able to directly mark text of interest while browsing content, and automatically save this text as a search request (query, also known as recommended search information) so that they can search it when needed. They may also want to be able to search multiple queries at the same time to obtain more relevant information at once.

[0073] In related technologies, to improve search efficiency, some search engines combine recommendation systems and object modeling techniques to provide personalized search results based on a subject's search history and behavior data. These search engines analyze a subject's search behavior to generate keyword recommendation models, thereby more accurately meeting the subject's search needs, shortening search time, and reducing search costs. This is primarily based on two types of search behavior records: the first is searches conducted after the subject browses content, which records the browsed content and search keywords; the second is the subject's consecutive search keywords and the search order between them. Using these records, the system generates two types of associations: the first is the association between content and search keywords, and the second is the association between search keywords. However, this approach still requires the subject to leave the current browsing page after browsing content to search, and the subject must analyze the browsed content and enter keywords on their own.

[0074] In view of this, the embodiments of the present application provide an information search method, device, electronic device and storage medium. In the embodiments of the present application, when an object browses the content details of a target content, it can directly mark content segments in the content details interface. For example, the object can mark content segments of interest or content segments that it has doubts about according to its own needs. Furthermore, the present application can intelligently generate recommended search information based on the segment details information contained in each content segment that the object has marked this time and before based on the marking operation of the object during the browsing process, and display it to the object. Since these recommended search information are generated based on the analysis of the content segments marked by the object, they can effectively reflect the search needs of the object. In addition, the object can further search for information based on the target recommended search information in the displayed recommended search information. In this search process, the object only needs to mark the content segments according to its own search needs, and does not need to spend time thinking and manually entering search keywords, which can effectively improve the efficiency of the search. In addition, in this way, on the basis of better meeting the search needs of the object, it can also help the object discover and obtain more relevant information, improve the search experience of the object, and increase the richness of the information obtained by the object.

[0075] The preferred embodiments of the present application are described below in conjunction with the drawings in the specification. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present application and are not used to limit the present application. In addition, the embodiments and features in the embodiments of the present application can be combined with each other if there is no conflict.

[0076] As shown in Figure 1, which is a schematic diagram of an application scenario of an embodiment of the present application, the application scenario diagram includes two terminal devices 110 and a server 120.

[0077] In the embodiment of the present application, the terminal device 110 includes but is not limited to mobile phones, tablet computers, laptop computers, desktop computers, e-book readers, intelligent voice interaction devices, smart home appliances, car terminals and other devices; the terminal device 110 may be installed with a client related to information search, which may be software (such as a browser, media software, etc.), or a web page, applet, etc. The server 120 is a background server corresponding to the software or web page, applet, etc., or a server specifically used for information search, which is not specifically limited in this application. The server 120 may be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0078] It should be noted that the information search method in each embodiment of the present application can be executed by an electronic device, which may be a terminal device 110 or a server 120, that is, the method can be executed by the terminal device 110 or the server 120 alone, or can be executed jointly by the terminal device 110 and the server 120. For example, when the terminal device 110 and the server 120 execute together, a client related to the information search service (such as a browser) is deployed on the terminal device 110. When the client receives a marking operation triggered by the object for the target content in the content details interface of the target content, the corresponding marking fragment can be sent to the server 120 through the terminal device 110. Then, the server 120 performs semantic analysis on the content details information contained in the marking fragment to generate at least one recommended search information. Thereafter, the at least one recommended search information is sent to the terminal device 110. The terminal device 110 presents at least one recommended search information in the content details interface of the target content through the client. When the object selects a target recommended search information in the at least one recommended search information, the client responds to the search operation triggered based on the target recommended search information and sends a search request generated based on the target recommended search information to the server 120 through the terminal device 110. The server 120 obtains the search results corresponding to the target recommended search information and feeds the search results back to the terminal device 110. The terminal device 110 presents these search results through the client.

[0079] In an optional embodiment, the terminal device 110 and the server 120 can communicate via a communication network. In an optional embodiment, the communication network can be a wired network or a wireless network.

[0080] It should be noted that what is shown in FIG1 is only an example. In fact, the number of terminal devices and servers is not limited and is not specifically limited in the embodiments of the present application.

[0081] In an embodiment of the present application, when there are multiple servers, the multiple servers can be formed into a blockchain, and the servers are nodes on the blockchain; the target content and related search information involved in the information search method disclosed in the embodiment of the present application can be saved on the blockchain, for example, the target content, the content fragments marked in the target content, the first search information, the second search information, the third search information, etc.

[0082] In addition, the embodiments of the present application can be applied to various scenarios, including but not limited to cloud technology, artificial intelligence, smart transportation, assisted driving and other scenarios.

[0083] It is understandable that in the specific implementation of this application, when the above embodiments of this application are applied to specific products or technologies, data related to target content, marking behavior of objects (such as objects), marked content fragments, etc., must be obtained from the object, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions.

[0084] The following describes the information search method provided by the exemplary embodiment of the present application in combination with the application scenarios described above and with reference to the accompanying drawings. It should be noted that the above application scenarios are only shown to facilitate understanding of the spirit and principles of the present application, and the implementation of the present application is not limited in this respect.

[0085] 2 is a flowchart of an implementation of an information search method provided in an embodiment of the present application. Taking a client (such as a browser) installed in an electronic device as an example, the specific implementation process of the method is as follows S21 to S23:

[0086] S21: The client displays a content details interface of the target content.

[0087] In the embodiment of the present application, the target content refers to the electronic content currently viewed by the subject, which may be, for example, articles (such as film reviews, entertainment news, etc.), news information, etc., and may include text, pictures and other information.

[0088] The target content's content details interface is an object-oriented interface used to present the target content's corresponding content details to the target object. Specifically, it presents the target content's text, images, and other content details. See Figure 3 for a schematic diagram of a content details interface in accordance with an embodiment of the present application. The target content shown in Figure 3 is a film review article, specifically presenting the publisher's evaluation of Movie C, produced by Director B.

[0089] In an embodiment of the present application, a convenient and clear content marking function is provided to the object so that the object can freely mark the content segments of interest while browsing the target content and sliding down.

[0090] Based on this function, objects can be marked in real time in the content details interface of the target content, and then relevant recommended search information can be intelligently generated based on the object's mark. This function can be referred to as the smart mark search function in the embodiment of this application.

[0091] For example, if the target content is text, users can mark words, phrases, sentences, and paragraphs that interest or have questions about them in the content details interface. Each marked word, phrase, sentence, or paragraph is considered a content snippet. After marking, recommended search information for the marked content snippet can be displayed in the content details interface.

[0092] Of course, the target content may include not only text content but also other content such as pictures. For example, when the marked content segment contains pictures, the pictures can be parsed first, and then relevant recommended search information can be generated based on the parsing results. Alternatively, the text in the pictures can be extracted to directly generate relevant recommended search information, etc. This article does not make specific restrictions.

[0093] The following uses the example of target content containing text content to illustrate the information search method in the embodiment of the present application:

[0094] Optionally, the aforementioned smart tag search function can be enabled by default, meaning that in the content details interface of any content, the subject can directly tag the content and intelligently generate recommended search information based on the tag. Alternatively, the subject can enable or disable the smart tag search function in the client. In this case, the smart tag search function is enabled or disabled by default for all content details interfaces in the client. For example, by configuring the corresponding control in the settings bar of the browser, the subject can choose to enable or disable the smart tag search function for all content details interfaces based on the control.

[0095] Optionally, the smart tag search feature can be enabled or disabled using a switch, and this control can be applied to individual content details screens. For example, a switch control can be set on the content details screen to enable or disable the smart tag search feature. Objects can then use this switch control to enable or disable the smart tag search feature on the content details screen corresponding to a particular piece of content.

[0096] Continuing with Figure 3 as an example, S31 in Figure 3 illustrates a switch control in an embodiment of the present application. When a subject enters the content details interface, a smart tag search switch appears. Currently, the switch control shown in S31 is off. The subject can adjust the switch control's state by clicking, for example.

[0097] As shown in Figure 4, which is a schematic diagram of another content details interface in an embodiment of the present application, the switch control shown in S41 in Figure 4 is in the on state. In this state, objects can mark content segments in the content details interface, and then intelligently generate recommended search information based on the object tags.

[0098] Optionally, the recommended search information can be directly presented in a partial area of ​​the content details interface (such as the bottom or other areas), or a search sub-interface (such as a floating layer, semi-floating layer, pop-up window, etc.) can be presented in the content details interface, and the recommended search information can be presented in the search sub-interface.

[0099] An optional implementation manner is: the client displays a search sub-interface at the bottom of the content details interface in response to the start operation triggered by the switch control.

[0100] Still using Figure 4 as an example, after the object clicks the switch control, the switch control switches from the closed state shown in interface 41 of Figure 4 to the open state shown in interface 42 of Figure 4, as shown in S41. In the open state, the search sub-interface shown in S42 of Figure 4 is presented on the content details interface. A blank text display box (also called a search box) is presented in this search sub-interface. Subsequently, recommended search information can be intelligently generated based on the content fragments marked by the object and presented in this blank text display box.

[0101] Optionally, when the object is not marked in the content details interface, the function buttons (including the search button and the all button) in the search sub-interface shown in S42 of Figure 4 are in an inactive state and cannot be clicked (as shown in the gray state of interface 42 of Figure 4, indicating that they are currently not clickable). However, when the object marks at least one content fragment (such as a simple text such as a word or phrase), the relevant function button area is activated (as shown in the black state of Figure 5, indicating that they are currently clickable).

[0102] It should be noted that the presentation form of the search sub-interface can be a semi-floating layer as shown in S42 in Figure 4. Of course, it can also be other forms, such as a pop-up window, etc. This application does not make specific limitations.

[0103] In the above embodiment, the subject can enable the smart tag search function in the content details interface to facilitate subsequent subject tagging of content segments that interest the subject or that have doubts about the subject.

[0104] S22: In response to a mark operation triggered on at least one content segment in the target content, the client displays, in the content details interface, a mark identifier corresponding to each of the at least one content segment and at least one recommended search information corresponding to the current mark operation; the recommended search information is generated based on the segment details information included in each content segment marked by the current and previous mark operations. When an object marks a content segment, a corresponding mark identifier is generated and displayed at the content segment, and the mark identifier is used to indicate the content marked by the object in the content details interface.

[0105] In the embodiments of the present application, the marking identifier includes, but is not limited to, some or all of the following: a check mark for the marked content segment and a mark sequence number. The mark sequence number indicates the order in which the object marked the content segment (i.e., the number of the marked content segment in the target content), and the selection mark can be achieved by gray filling, bolding, highlighting, underlining, changing the font color, etc., which are not specifically limited in this article.

[0106] In an embodiment of the present application, after an object marks a content segment, it is necessary to perform semantic analysis on the segment detail information contained in the content segment marked by the object in real time. During this process, the speed of the semantic analysis will also affect the presentation of the corresponding recommended search information.

[0107] The fragment details information included in the content fragment is the content of the content fragment marked by the object, such as text, picture and other information.

[0108] Specifically, after an object marks a content segment, the relevant function button area is activated. The search sub-interface in the content details interface will generate at least one recommended search information (query) that is strongly related to the marked content segment based on semantic understanding and integrated sorting of relevance a short period of time (i.e., a preset time range) after the object marks the content segment, so that the object can conduct timely searches based on his or her own doubts or interests.

[0109] Among them, the length of the above-mentioned preset time range can also further reflect the number of times the recommended search information is presented. For example, when the semantic analysis speed is relatively ideal, that is, when the recommended search information can be generated relatively quickly, each time the object marks a content segment, the recommended search information can be promptly presented once in the content details interface. For another example, when the semantic analysis speed is not ideal, that is, when the speed of generating recommended search information is slow, the object needs to wait for a certain period of time after the marking is completed. During the waiting process, the object may mark again. In this case, the recommended search information can be presented once in the content details interface each time the object marks at least one content segment (including one content segment and multiple content segments).

[0110] In an embodiment of the present application, based on the above idea, it can be set that if the marking time of at least one content segment is within a preset time range, then every time at least one content segment is marked, at least one recommended search information corresponding to this marking will be displayed in the search sub-interface of the content details interface.

[0111] Among them, the preset time range is set according to the speed of "generating recommended search information based on the fragment detail information contained in each content fragment marked this time and before this time". When generating recommended search information based on the fragment detail information contained in each content fragment marked by the object, semantic analysis is performed based on the fragment detail information, and the corresponding recommended search information is generated based on the results of the semantic analysis, that is, the speed of generating recommended search information can be considered as the speed of semantic analysis. If it is set to 2 seconds, it means that when the object continuously marks two or more content fragments within 2 seconds, the semantic analysis speed is not as fast as the object marking speed. At this time, based on all content fragments marked by the object within 2 seconds and the previously marked content fragments, only one recommended search information can be generated and presented. When the marking time interval of two content fragments marked continuously by the object is greater than 2 seconds, each time the object marks, recommended search information can be generated and presented based on the content fragment marked this time and the previously marked content fragment.

[0112] In an embodiment of the present application, after an object is tagged with a content segment, it is necessary to generate recommended search information corresponding to the tagged content segment in real time based on the segment detail information contained in the tagged content segment, and present the recommended search information in a content details interface, such as in a search sub-interface within the content details interface, or in the bottom area of ​​the content details interface, etc. There are often more than one recommended search information generated based on semantic analysis. Therefore, the content details interface can present recommended search information corresponding to a single tagged content segment, or it can present recommended search information corresponding to multiple tagged content segments, which is not specifically limited in this document.

[0113] The above two situations are briefly described below with reference to the accompanying drawings:

[0114] As shown in Figure 5, it is a schematic diagram of the process of marking the first content segment in an embodiment of the present application. Among them, the object marks the content segment "Director B", as shown in S51, and the content segment is marked in the content details interface with a gray background fill style, and the corresponding marking order is presented in the upper right corner of the content segment, as shown in Figure 5, indicating that the content segment is the first marked segment. If the object is not marked for the second time within 2 seconds, the recommended search information generated based on the mark can be directly presented in the search sub-interface, as shown in Figure 5 S52 in the search sub-interface, where a generated recommended search information "Director B" is presented in the search box.

[0115] Furthermore, when recommended search information is presented, the relevant function button area is activated, indicating that it is currently clickable. For example, the subject can click the search button to search for "Director B" as the search term; for example, the subject can click the All button to view more candidate search information, and so on.

[0116] As shown in Figure 6, it is a schematic diagram of the second process of marking content segments in an embodiment of the present application. Among them, interface 61 in Figure 6 illustrates that the subject first marks the first content segment "Director B". As shown in S61, the subject needs to wait for a certain period of time after the marking is completed. The recommended search information cannot be temporarily presented in the search box of the search sub-interface, but a prompt "Generating..." as shown in S62 is presented. Within 2 seconds after the first marking, the subject marks the second content segment "Movie C" again, as shown in S63 in interface 62 in Figure 6. After marking the second content segment, the subject waits for a certain period of time, and then the search sub-interface shown in S64 in interface 62 in Figure 6 presents the recommended search information "Box Office of Director B's Movie C in L Year" generated based on the first and second content segments marked by the subject.

[0117] It should be noted that the situation listed in Figure 6 above indicates that the time interval between the first mark and the second mark is within 2 seconds. In this case, before the recommended search information is generated based on the first mark of the object (prompt S62 shows "Generating..."), the object immediately performs a second mark. In this case, recommended search information can be generated based on these two marks at the same time, and then the recommended search information shown in S64 is presented.

[0118] If the marking time interval between two markings of the object is not within the preset time range listed above, the client may present the marking identifier corresponding to the content segment in the content details interface every time the object marks a content segment, as well as at least one recommended search information corresponding to this marking.

[0119] Refer to Figure 7, which is a schematic diagram of the third process of marking content segments in an embodiment of the present application. Figure 7 is an example of presenting recommended search information every time the object marks a content segment. For example, if the object first marks the first content segment "Director B", the recommended search information "Director B" generated based on the content segment is presented in the search box of the search sub-interface shown in S71; then, the object marks the second content segment "Movie C" again, and the recommended search information "Director B's L-year Movie C box office" generated based on the first and second content segments is presented in the search box of the search sub-interface shown in S72; then, the object marks the third content segment "Director F" again, and the recommended search information "Director B's Movie C plagiarized Director F" generated based on the first, second and third content segments is presented in the search box of the search sub-interface shown in S73.

[0120] It should be noted that the examples listed above all use the example of presenting only one recommended search information at a time. In practice, multiple recommended search information may be presented at a time. The number of recommended search information presented each time may be the same or different, and this document does not impose any specific restrictions.

[0121] The following example presents two recommended search information at a time: FIG8 is a schematic diagram of a recommended search information example in an embodiment of the present application. As shown in FIG8 , the object has five content segments marked sequentially, numbered 1 through 5, in the content details interface. Due to the limited display size of the content details interface, FIG8 only shows the fourth content segment, "G Movie," and the fifth content segment, "Country Z Version of O Movie."

[0122] After the object is marked for the fifth time, semantic analysis is performed on the segment detail information contained in the five content segments, and recommended search information is generated based on the results of the semantic analysis. The recommended search information is presented in the search sub-interface. For example, in the search sub-interface shown in S81 in Figure 8, two recommended search information items, "Box office of Director B's "Movie C"" and "Screening schedule of Director B's "Movie C"", are presented.

[0123] In the above embodiment, the object can arbitrarily mark the content segments of interest or content segments of doubt through the smart tag search function. After semantic analysis of the segment detail information contained in the at least one content segment, at least one recommended search information is presented in the search sub-interface. The object can click on the recommended search information to directly present the search results, so that the object can quickly generate recommended search information based on the object's search needs while browsing the content details interface, avoiding the problem of the object summarizing the keyword search by itself, improving the search efficiency, and carrying the recommended search information through the search sub-interface can easily distinguish the recommended search information from the currently browsed target content, so that the object can intuitively understand the recommended search information and conduct subsequent searches based on it.

[0124] In an embodiment of the present application, the marking capability provided by the client is not limited to the structural form of the marked content. Taking text content as an example, the content segment marked by the object can be a simple word, word, phrase, or a sentence, a large paragraph of text, etc. If the content segment marked by the object in the content details interface is a simple word, word, phrase, etc., it is only necessary to directly generate relevant recommended search information based on the simple word, word, phrase. The recommended search information can be the simple word, word, phrase itself, or it can be the recommended search information generated by associating the simple word, word, phrase with other content segments. If the content segment marked by the object in the content details interface is a sentence, a large paragraph of text, etc., at this time, based on the semantic understanding of the content segment, the core content or viewpoint can be extracted, and then the extracted content or viewpoint can be associated with other content segments marked by the object in the content details interface to generate recommended search information.

[0125] As shown in Figure 9, it is a schematic diagram of another type of marked content fragment in an embodiment of the present application. In interface 91 of Figure 9, the object has marked two content fragments with marking serial numbers 6 and 7 in the target content of the content details interface. Both content fragments are large paragraphs of text. Among them, the content fragment with marking serial number 6 is "From the perspective of narrative tone, there is indeed a suspicion of reference to these two movies. In addition, the plot is familiar, the processing level is low, and the viewing process has a sense of copycat imitation from time to time." The content fragment with marking serial number 7 is "The suspense of the plot is actually not strong, the narrative tension is far less than the first "K Movie", and the charm of the characters cannot be compared with the first "M Movie"."

[0126] Taking the seventh marking as an example, after marking the above-mentioned content segments, by performing semantic analysis on the segment detail information contained in the seven content segments with marking serial numbers 1-7, the recommended search information "Director B's "C Movie" is not as good as "M Movie"" is obtained, and the recommended search information "Movie review of Director B's plagiarized work "C Movie"" is presented in the search sub-interface of the content details interface 91 in Figure 9.

[0127] In interface 92 of Figure 9, the subject marks the content segment with the mark number 8 in the target content of the content details interface. The content segment is a long text "The reason why "M Movie" is so good is that the father played by actor N gives the audience the duality of a good father and a highly intelligent XX. The charm of the character and the charm of the performance are both very eye-catching. The key to the popularity of "M Movie" is that it treats the audience as God, playing with wisdom and suspense from the audience's God's perspective. In contrast, "C Movie" treats the audience as fools. Both camps are acting for the sake of acting and deceiving. By the time their two sides are finally revealed through a forced reversal, it's too late. All that remains is clumsy and mechanical, devoid of any character charm. "The content extracted from this text is "The difference between 'Movie C' and 'Movie M'." This is then associated with the other seven content segments marked by the subject in the target content. Finally, semantic analysis is performed based on the segment details contained in these eight content segments. Based on the results of this semantic analysis, the recommended search information "Director B's 'Movie C' is not as good as 'Movie M'" is presented in the search sub-interface of the content details interface in Figure 9."

[0128] In an embodiment of the present application, query information (query) based on the object's custom annotation content can be generated for the search, without relying on the object's input. That is, the object can freely mark existing content and convert it into a single query for search. The object only needs to mark the content that it is interested in or has doubts about. Based on the marked content and the object's marking behavior, the system intelligently generates search queries that the object may be interested in and have a high correlation with all the marked content fragments (the system sorts the queries based on the object's intention prediction and the comprehensive correlation of all the marked content fragments). The remaining unused queries (those ranked lower) are uniformly collected into the query library for use. The object ultimately selects the content in the query library to search.

[0129] Among them, some queries in the query library can be presented directly in S22, and the remaining queries can be presented when the object chooses to view the first search information set (this will be explained in detail below and will not be explained in detail here). Queries with a higher relevance to the currently marked content fragment will be presented to the object first, further improving the efficiency of information search.

[0130] Based on the above, you can directly search based on the recommended search information generated by the tag intelligence. For example, S23:

[0131] S23: In response to the search operation triggered by the target recommended search information in the at least one recommended search information, the client displays the search results corresponding to the target recommended search information.

[0132] In the content details interface, the user can select a target recommended search information from at least one recommended search information presented and trigger a search operation based on the target recommended search information. After the search operation is triggered, the client sends a search request generated based on the target recommended search information to the server via the terminal device. After receiving the search request, the server retrieves content matching the target recommended search information from the content pool as the search result corresponding to the target recommended search information. Furthermore, after obtaining the search results corresponding to the target recommended search information, the server feeds them back to the terminal device, which then presents these search results through the client.

[0133] FIG10 is a schematic diagram illustrating a process for performing an information search based on target recommended search information in an embodiment of the present application. For example, if only one recommended search information is currently displayed in the search box of the search sub-interface, such as "Director B's 'C Movie' is not as good as 'M Movie'" as shown in interface 101, the user can click the "Search" button S101 in the search box to use this recommended search information as the target recommended search information and search for related content, thereby presenting interface 102, which displays the search results corresponding to "Director B's 'C Movie' is not as good as 'M Movie'".

[0134] In the above implementation, the subject can further search for information based on the recommended search information. During this search process, the subject only needs to tag the information based on their search needs, eliminating the need for the subject to spend time thinking about or manually entering search keywords, effectively improving search efficiency. Furthermore, this approach not only better meets the subject's search needs, but also helps the subject discover and obtain more relevant information, enhancing the subject's search experience and increasing the richness of the information obtained.

[0135] Optionally, the above-mentioned generation process of recommended search information can be implemented using some methods based on natural language processing and machine learning, such as using natural language processing technology through the server to understand and identify the marked content fragments, sorting according to the correlation between the generated recommended search information and the marked content fragments, determining the recommended search information within the specified order range, and returning it to the terminal device, and presenting the recommended search information in the content details interface (such as presented in the search sub-interface) through the client installed on the terminal device. For details, please refer to the following embodiments, and the repetitions will not be repeated.

[0136] In the above-mentioned embodiment, artificial intelligence technology can be used to quickly and accurately generate recommended search information, so that when the subject browses the target content (especially news and current affairs content with complex characters and character relationships), in addition to exploring the content through the search information recommended by the system, the subject can also mark the content and events of interest through his own operations, and conduct a more in-depth information search based on this, providing a more personalized search experience and greatly satisfying the subject's search needs.

[0137] In the embodiment of the present application, there may be a situation where the recommended search information presented in the search sub-interface cannot meet the current search needs of the subject, or the subject wants to view more selectable recommended search information. In these cases, an optional implementation method is:

[0138] Set the option view control in the content details interface, such as in the bottom area of ​​the content details interface, or in the search sub-interface. The following still takes the search sub-interface as an example:

[0139] The object can trigger the option viewing control of the search sub-interface in the content details interface. In response to the viewing operation triggered by the object based on the option viewing control, the client displays the first search information set corresponding to each marked content segment in the content details interface.

[0140] Optionally, the first search information set corresponding to each marked content segment may be presented in the search sub-interface of the content details interface.

[0141] Among them, each first search information set contains one or more first search information, which are intelligently generated based on the content fragments marked by the object, and at least include the "at least one recommended search information" presented in the above S22. In addition, it can also include candidate search information generated based on the various content fragments marked this time and before, and the correlation between at least one recommended search information and each content fragment meets the preset correlation condition, for example, the ranking of the correlation with each content fragment is within the specified order range, while the correlation between the candidate search information and each content fragment does not meet the preset correlation condition, for example, the ranking of the correlation with each content fragment is not within the specified order range.

[0142] The designated order range may refer to a designated order range when sorting relevance from smallest to largest or from largest to smallest, such as "topN" representing the most relevant search results, where N is a positive integer. For example, when N is 1, the most relevant search result will be directly presented in S22. For example, when N is 2, the most relevant and second most relevant search results will be directly presented in S22, and so on.

[0143] Optionally, relevance ranking may refer to ranking based on at least one of the content relevance between the respective content segments and the intention relevance to the search intent of the object. The ranking rule may be to sort from highest to lowest relevance. The at least one recommended search information within the specified order range may be the most highly relevance recommended search information in the first search information set, or the most highly relevance and the second most highly relevance recommended search information, etc., without further limitation herein.

[0144] Optionally, the option viewing control is the "All" control of the search sub-interface in the content details interface. When the object marks at least one content segment, the mark identifier corresponding to at least one content segment is presented in the content details interface. At this time, the search sub-interface in the content details interface presents at least one corresponding recommended search information. The object clicks the "All" control in the search sub-interface to view the first search information set presented in the search sub-interface, including at least one recommended search information (which can be recorded as the optimal query) and other candidate search information (which can be recorded as the query history that is not on the list).

[0145] For example, the query details include an intelligent search module. The queries generated by intelligent search dynamically change as the object's tagged content and its weighting relative to the text length change. This intelligent search module includes the query previously output in the search box at the bottom (including the optimal query and unranked queries), as well as the optimal query up to the last tagged content and related unranked queries, sorted by time and content.

[0146] As shown in Figure 11, it is a schematic diagram of a first search information set provided in an embodiment of the present application. The object has marked eight content segments in the content details interface. Interface 111 in Figure 11 indicates that the content segment marked by the object with the tag number 8, and the recommended search information "Director B's "C Movie" is not as good as "M Movie"" generated based on the above eight content segments, is presented in the search sub-interface in interface 111 in Figure 11. If the recommended search information cannot meet the search needs of the object, at this time, the object clicks the "All" control shown in S111 in Figure 11, and the content details interface will be as shown in interface 112 in Figure 11, increasing the semi-floating range of the search sub-interface, and in the semi-floating layer, presenting the intelligent search module shown in S112 in Figure 11, and the intelligent search module presents the first search information set associated with the content segments with tag numbers 1 to 8. Each first search information set includes all recommended search information generated based on the association of the object's current mark and previously marked content segments, such as "8-1", "7-1", "6-1", "5-1", and "4-1" in S112, which are the identifiers of these first search information sets.

[0147] Among them, all the recommended search information in 8-1, that is, the first search information set corresponding to the content segment marked with sequence number 8, the first search information in this first search information set is all the search information generated based on the marked content segments 1 to 8 (also called the first search information), and the search information is sorted from high to low according to relevance, with the most relevant "Director B's 'C Movie' is not as good as 'M Movie'" ranked first. At the same time, this most relevant search information will be presented as recommended search information in the search sub-interface 111 of Figure 11. Similarly, all the search information in 7-1, that is, the first search information set corresponding to the content segment marked with sequence number 7, the first search information in this first search information set is all the search information generated based on the marked content segments 1 to 7, and the search information is sorted from high to low according to relevance, with the most relevant "Director B's 'C Movie' plagiarized Director F's Movie" ranked first. Similarly, after the content segment 7 is marked, this search information will also be presented as recommended search information in the search box of the search sub-interface, and so on.

[0148] In an embodiment of the present application, the object can view more recommended search information by sliding operations. Specifically, the object can view more first search information sets that are not currently displayed by sliding up and down. For example, the object can further view the first search information sets "3-1", "2-1", and "1-1" by swiping up; the object can also view more first search information that is not currently displayed in each first search information set by sliding left and right.

[0149] As shown in Figure 12, it is a schematic diagram of another first search information set provided in an embodiment of the present application. Interface 121 in Figure 12 indicates that after the object triggers the "All" control of the search sub-interface in the content details interface, the homepage of the smart search module is presented, which contains part of the first search information. If the object needs to view more first search information, it can swipe left in the area where the smart search module is located. At this time, the smart search module is shown in interface 122 in Figure 12, and interface 122 presents more first search information such as "Evaluation of Director B's Film Works", "What are the original films of Director B's "C Movie", "Word-of-mouth Rating of "C Movie", "Schedule of Director B's "C Movie", "Is "C Movie" Worse than "M Movie"", etc.

[0150] In the above embodiment, the object can freely mark the content in the target content that he is interested in or has doubts about, without having to convert it into keywords for search. A first search information set is generated based on the content marked by the object, so that the object can select any search information in the first search information set according to his own search needs, shortening the object's search path and providing the object with richer search suggestions.

[0151] Based on the above, the object can be searched directly based on the first search information intelligently generated according to the tag. An optional implementation is as follows:

[0152] The client displays the search results corresponding to any one of the first search information in response to the search operation triggered by the object based on any one of the first search information.

[0153] In the search sub-interface, the user can select a target first search information from multiple presented first search information and trigger a search operation based on the target first search information. After the search operation is triggered, the terminal device sends a search request generated based on the target first search information to the server. After receiving the search request, the server retrieves content matching the target first search information from the content pool as the search results corresponding to the target first search information. Furthermore, after obtaining the search results corresponding to the target first search information, the server feeds them back to the terminal device, which then presents these search results through the client.

[0154] FIG13 is a schematic diagram illustrating an information search process based on target first search information in an embodiment of the present application. If the intelligent search module in the current search sub-interface presents multiple first search information, the subject can select one of the first search information and use the selected first search information as the target first search information to search for related content, presenting interface 132, i.e., presenting search results corresponding to "What are the original films of Director B's 'Movie C'?"

[0155] In the above embodiment, the object selects the first search information in the first search information set according to the search needs, and directly views the search results by clicking on the first search information, so that the object can gain a deeper understanding of the relevant information according to its own search needs, shorten the search path, and improve search efficiency.

[0156] In an embodiment of the present application, if the recommended search information presented in the search sub-interface and the first search information presented in the intelligent search module still cannot meet the search needs of the object, then when the object triggers the option view control of the search sub-interface in the content details interface, in response to the view operation triggered by the object based on the option view control, in addition to displaying the first search information set, at least one second search information corresponding to each marked content segment is also displayed in the content details interface.

[0157] Optionally, at least one second search information corresponding to each marked content segment may be displayed in the search sub-interface of the content details interface.

[0158] Specifically, the option viewing control is the "All" control of the search sub-interface in the content details interface. When the object marks at least one content segment, at least one mark identifier corresponding to each content segment is presented in the content details interface. At this time, the search sub-interface in the content details interface presents at least one corresponding recommended search information. When the object clicks the "All" control in the search sub-interface, in addition to viewing the first search information set introduced above, at least one second search information corresponding to each marked content segment can also be viewed.

[0159] The second search information is search information generated based on the content segment marked this time. Each content segment may generate at least one second search information, and the number of second search information corresponding to different content segments may be the same or different.

[0160] In an embodiment of the present application, unlike the first search information, the second search information can be used by the object to conduct autonomous splicing search. Therefore, the second search information only needs to be generated based on the content fragment marked this time, without considering the impact of other content fragments marked before on this marking.

[0161] Specifically, after the object clicks the "All" control in the search sub-interface, the query details include an autonomous word search module in addition to the intelligent search modules listed above. Due to the system's ability to understand the tagged content and weight calculation, it may not be able to meet the ever-changing objects. Therefore, this autonomous word search module can be used to provide the object with the ability to autonomously splice the tagged content and adjust the order to integrate the output query. If the first search information in the intelligent search module in Figures 11 to 13 does not meet the object's expectations, the object can also use the autonomous word search module to search, as shown in Figure 14.

[0162] FIG14 is a schematic diagram of a second search information provided in an embodiment of the present application. An object has tagged eight content segments in the content details interface. Interface 141 in FIG14 displays the tag identifier for the content segment tagged with tag number 8, as well as the recommended search information "Director B's 'Movie C' is not as good as 'Movie M'" generated based on the eight content segments, presented in the search sub-interface of interface 141. If the recommended search information cannot meet the object's search needs, at this time, the object clicks the "All" control shown in S141 in Figure 14, and the content details interface 142 will present the independent spelling search module shown in S142 in Figure 14. There are eight second search information in the independent spelling search module, such as: the second search information "1. B Director" generated based on content segment 1, the second search information "2. "C Movie"" generated based on content segment 2, the second search information "3. F Director" generated based on content segment 3, the second search information "4. "M Movie"" generated based on content segment 4, the second search information "5. "O Movie" Chinese version" generated based on content segment 5, the second search information "6. "C Movie" plot copycat" generated based on content segment 6, the second search information "7. "C Movie" plot bad review" generated based on content segment 7, and the second search information "8. "M Movie" "C Movie" comparison" generated based on content segment 8.

[0163] In the above embodiment, at least one corresponding second search information is generated for all content segments currently marked by the object, so that the object can search for any one or more currently marked content segments at any time, and facilitate subsequent objects to conduct independent splicing searches on the second search information, thereby improving the freedom of search and ensuring that the basic information based on the search is more in line with the object's needs.

[0164] In an embodiment of the present application, an object triggers an option viewing control of a search sub-interface in a content details interface. After displaying the second search information, a search operation can also be triggered based on at least one second search information. An optional implementation is as follows:

[0165] The client displays the search results corresponding to the spliced ​​search information in response to the search operation triggered by the object based on the at least one second search information.

[0166] The spliced ​​search information is obtained by splicing at least one second search information.

[0167] Specifically, the object triggers the option viewing control of the search sub-interface in the content details interface. After the second search information is presented, you can click on at least one second search information in sequence. According to the click order of the object, the at least one second search information selected by the object can be spliced ​​to form spliced ​​search information. Information search is performed based on the spliced ​​search information, and corresponding search results are presented.

[0168] FIG15 illustrates a schematic diagram of an autonomous splicing search based on secondary search information provided by an embodiment of the present application. Interface 151 in FIG15 illustrates that after a subject triggers the "All" control in the search sub-interface of the content details interface, an autonomous splicing search module is presented in the content details interface. This autonomous splicing search module contains eight secondary search information, such as "1. Director B." The subject can click on at least one of these eight secondary search information to generate spliced ​​search information. For example, if the subject clicks on the three secondary search information in the autonomous splicing search module, namely "1. Director B," "5. 'O Movie China Version,'" and "'C Movie' Plot Bad Reviews," the client can send these three secondary search information to the server, which will perform splicing to generate the spliced ​​search information, "'O Movie' Directed by Director B, Z Country Version, 'C Movie' Plot Bad Reviews." Based on this spliced ​​search information, the server searches the content pool for matching search results, returns the matching search results to the client, and presents the spliced ​​search results to the subject via the client, as shown in interface 152 in FIG15.

[0169] In the above embodiment, the second search information is flexibly combined and sorted by the object, so that the object can obtain a more personalized search experience, helping the object to explore the content of interest more deeply and accurately.

[0170] In an embodiment of the present application, the positive and negative tendencies of the output query can be adjusted in advance, such as approval (wanting to see more, such as checking the first search information and the second search information mentioned above), disapproval (seeing negative evidence, such as checking the third search information below), etc.; when the query provided by the system does not meet the object's requirements, the object can click on the generated query library and search independently according to the query provided by the system.

[0171] An optional implementation is that the embodiment of the present application also supports reverse search of objects. Specifically, each time an object is marked, at least one recommended search information can be presented in the content details interface, and reverse search prompt information can be presented at the same time. Alternatively, when the recommended search information presented in S22 meets the preset reverse conditions, the reverse search prompt information can also be presented when the recommended search information is presented. An optional implementation is as follows:

[0172] In an embodiment of the present application, based on a natural language understanding model and search result calculations, if it is determined that the recommended search information contains search information with semantically opposite search results, a reverse search prompt message may be further displayed in the search sub-interface. If the subject agrees with the content of the recommended search information in the search box of the current search sub-interface, the search can be performed directly; if the subject disagrees or wants to see a contrary opinion, the reverse search prompt message can also be used to directly and quickly adjust the search results in the content pool.

[0173] Specifically, reverse search prompt information is displayed in the content details interface, and the reverse search prompt information is used to prompt a method of searching for content with semantically opposite search results corresponding to at least one recommended search information. The reverse search prompt information can be in the form of a text link or a button, etc., which is not specifically limited in this document. The object can click (or other operations such as long press, which are not specifically limited in this document) the reverse search prompt information to trigger a reverse search operation. In response to the reverse search operation triggered based on the reverse search prompt information, the client displays the third search information associated with each of the at least one recommended search information in the content details interface.

[0174] The third search information and the associated recommended search information have opposite content semantics for the search results. The reverse search prompt information can be presented below the search box in the search sub-screen or elsewhere in the content details screen. It is not required to be presented in the search sub-screen or below the search box in the search sub-screen, and this document does not impose specific restrictions.

[0175] For example, when the object does not recognize the recommended search information in the search sub-interface and the search sub-interface presents reverse search prompt information, the object can click on the reverse search prompt information to trigger the presentation of third search information corresponding to the recommended search information below at least one recommended search information in the search sub-interface, and the search result corresponding to the third search information is semantically opposite to the search result corresponding to the recommended search information.

[0176] Furthermore, in addition to presenting the third search information opposite to the at least one recommended search information in S22, the third search information corresponding to other first search information may also be presented in the search sub-interface. That is, an optional implementation is as follows:

[0177] In response to the reverse search operation triggered by the reverse search prompt information, the client displays the third search information associated with at least one recommended search information and the third search information associated with each candidate search information in the search sub-interface of the content details interface.

[0178] For example, after the object clicks on the reverse search prompt information, the client responds to the reverse search operation triggered based on the reverse search prompt information, and presents the third search information associated with at least one recommended search information and the third search information associated with each candidate search information in the smart search module in the search sub-interface of the content details interface.

[0179] The candidate search information is generated based on the content segments marked this time and before, and the relevance between the candidate search information and the content segments is lower than the relevance between the recommended search information and the content segments.

[0180] As shown in Figure 16, it is a schematic diagram of a third search information in an embodiment of the present application. Interface 161 in Figure 16 indicates that the object has marked the content segment of interest in the content details interface. For example, after marking content segment 8, the corresponding recommended search information "Director B's "C Movie" is not as good as "M Movie"" is presented in the search box of the search sub-interface, and based on the natural language understanding model and search result calculation, it is determined that the recommended search information has a search result with the opposite semantics to its corresponding search result "Director B's "C Movie" has a great reputation", and then the reverse search prompt information "Do not agree with the above statement? Click to view related opposite results" as shown in S161 in Figure 16 is presented in the search sub-interface. If the object does not agree with the recommendation To search for the information "Director B's "C Movie" is not as good as "M Movie"," you can click "Click to view related opposite results" in the reverse search prompt information to enter interface 162 in Figure 16. The intelligent search module in interface 162 contains the recommended search information in 8-1, which now includes both the original first search information and the third search information associated with the first search information. For example, the first search information "Director B's "C Movie" is not as good as "M Movie"" is associated with the third search information "Director B's "C Movie" has a great reputation", and the search result corresponding to the third search information is semantically opposite to the search result corresponding to the first search information.

[0181] As shown in FIG17 , it is a schematic diagram of another third search information in an embodiment of the present application. Interface 171 in FIG. 17 indicates that the subject has marked a content segment of interest in the content details interface. For example, after marking content segment 7, the corresponding recommended search information "Director B's plagiarized work "C Movie" viewing review" is presented in the search box of the search sub-interface. Based on the natural language understanding model and search result calculation, it is determined that the recommended search information has a search result with the opposite semantics to its corresponding search result. Then, the reverse search prompt information "Disagree with the above statement? Click to view related opposite results" as shown in FIG. 17 is presented in the search sub-interface. If the subject disagrees with the recommended search information "Director B's plagiarized work "C Movie" viewing review", he or she can click "Click to view related opposite results" in the reverse search prompt information to enter interface 172 in FIG. The intelligent search module in interface 172 includes the recommended search information in 7-1. At this time, it includes both the original first search information and the third search information associated with the first search information. For example, the first search information "Director B's plagiarized work "C Movie" viewing review" is associated with the third search information "Director B's "C Movie" plagiarism clarification". The search result corresponding to the third search information is semantically opposite to the search result corresponding to the first search information.

[0182] In addition, after the object clicks the reverse search prompt information, the client can respond to the reverse search operation triggered by the object based on the reverse search prompt information, and present the third search information associated with each first search information in the smart search module in the search sub-interface of the content details interface.

[0183] Refer to FIG18, which is a schematic diagram of another third search information in an embodiment of the present application. Among them, the interface 181 in FIG18 indicates that the object has marked the content segment of interest in the content details interface. For example, after marking the content segment 8, the corresponding recommended search information "Director B's "C Movie" is not as good as "M Movie"" is presented in the search box of the search sub-interface, and based on the natural language understanding model and the search result calculation, it is determined that the recommended search information has search results with opposite semantics to its corresponding search results, and then the reverse search prompt information "Do not agree with the above statement? Click to view related opposite results" as shown in FIG18 is presented in the search sub-interface. If the object does not agree with the recommended search information "Director B's "C Movie" is not as good as "M Movie"," the object can click "Click to view related opposite results" in the reverse search prompt information to enter the interface 182 in FIG18. The intelligent search module in the interface 182 includes 8-1 to 1-1. The recommended search information, that is, the recommended search information corresponding to all the content segments currently marked by the object, includes both the original first search information and the third search information associated with the first search information, as shown in part S181: the first search information "Director B's plagiarized work "C Movie" viewing reviews" in 6-1 is associated with the third search information "Director B's "C Movie" viewing reviews"; the first search information "Director B's plagiarized work "C Movie" viewing reviews" in 7-1 is associated with the third search information "Director B's "C Movie" plagiarism clarification"; the first search information "Director B's "C Movie" is not as good as "M Movie"" in 8-1 is associated with the third search information "Director B's "C Movie" has a great reputation", etc. The search results corresponding to the above third search information have the opposite semantics to the search results corresponding to the first search information.

[0184] In the above embodiment, the present application can better meet the reverse search needs of the object by generating the third search information, which can not only improve the efficiency of the search, but also help the object discover and obtain more relevant information, improve the object's search experience, and obtain other information relative to the marked content fragment.

[0185] In an optional embodiment, if the object is interested in the third search information, the object can directly click on the third search information and search for information based on the third search information. Specifically, the client responds to the search operation triggered by the object based on any third search information and displays the search results corresponding to any third search information.

[0186] Specifically, in response to a search operation triggered by an object based on any third search information (also referred to as target third search information), the client sends a corresponding search request to the server. Upon receiving the search request, the server retrieves content matching the target third search information from the content pool as search results corresponding to the target third search information. Furthermore, the server obtains the search results corresponding to the target third search information and feeds them back to the terminal device, which then presents these search results through the client.

[0187] Refer to Figure 19, which is a schematic diagram of presenting search results based on the third search information in an embodiment of the present application. Interface 191 in Figure 19, the intelligent search module in interface 191 contains the recommended search information from 8-1 to 1-1, that is, the recommended search information corresponding to all the content segments currently marked by the object. At this time, it contains both the original first search information and the third search information associated with the first search information. For example, the first search information "Director B's "C Movie" plagiarized Director F's movie" in 7-1 is associated with the third search information "Director B's "C Movie" plagiarism clarification". The search results corresponding to the third search information are semantically opposite to the search results corresponding to the first search information. The object clicks on the third search information to use the third search information as the target third search information to search for related content, and presents interface 192, which presents the search results corresponding to "Director B's "C Movie" plagiarism clarification".

[0188] In the above embodiment, when the object clicks on the third search information in the smart search interface, the search results with the opposite semantics to the content segment marked by the object are directly displayed, which better meets the reverse search needs of the object, not only improves the efficiency of the search, but also helps the object discover and obtain more relevant information, thereby improving the object's search experience.

[0189] Optionally, before presenting the reverse search prompt in the content details interface, the recommended search information presented in S22 may be analyzed to determine whether it meets a preset reverse condition. One optional implementation is as follows: Among at least one recommended search information displayed in the content details interface, determine whether there is any controversial recommended search information. Only if it is determined that there is controversial recommended search information is the reverse search prompt displayed in the content details interface.

[0190] Specifically, after the system determines the intelligent query based on all content fragments marked by the object, and judges that the query is "controversial" based on database retrieval and understanding of semantics, it activates the reverse search function (such as presenting reverse search prompt information), that is, after the original output of the "controversial" query, the reverse search button enters the details page of the "controversial" query. The object can quickly obtain the opposite query of the "controversial" query (that is, the third search information in this article) in the query details page to quickly open the search purpose.

[0191] In the embodiment of the present application, when analyzing whether the recommended search information is controversial, the optional analysis methods include but are not limited to the following two:

[0192] Analysis method one is to determine whether the recommended search information is controversial by verifying the authenticity of the search results corresponding to the recommended search information.

[0193] Specifically, when generating at least one recommended search information based on the object's tag information, the authenticity of the at least one recommended search information may also be verified in advance. Some search results corresponding to the recommended search information may contain both true and false information. For example, a search result corresponding to a recommended search information may contain a fact about an event or person, but this fact may have been incorrectly reported or deliberately distorted. In this case, the recommended search information is controversial.

[0194] In this case, fact-checking technology can be used to verify the authenticity of search results. Specifically, this can be achieved by comparing information from multiple sources, querying authoritative databases, using pre-trained fact-checking models, etc., to analyze whether a fact has been incorrectly reported or deliberately distorted, and then determine whether the corresponding recommended search information (query) is controversial.

[0195] The second analysis method is to use text classification technology to identify opposing viewpoints on the search results corresponding to the recommended search information, and determine whether the recommended search information is controversial based on the identification results.

[0196] Specifically, when generating at least one recommended search information based on the object's tag information, opposing viewpoints may be identified for the at least one recommended search information. Some search results corresponding to the recommended search information may contain opposing viewpoints. For example, a search result corresponding to a recommended search information may contain opinions on a certain policy or theory, which may vary from person to person. In this case, if different subjects have different views on the policy or theory, then the recommended search information is controversial.

[0197] In this case, text classification technology can be used to identify different viewpoints. If it can be identified that there are multiple viewpoints or positions in the search results of a certain recommended search information (query), and their proportions are similar, then it can be determined that the recommended search information is controversial.

[0198] Then, you can provide search results that include multiple viewpoints based on the subject's needs and interests, or provide search results that match the subject's position based on the subject's position.

[0199] In an embodiment of the present application, the controversialness of the recommended search information can be judged by any at least one of the above-mentioned methods. If there is controversial recommended search information in at least one of the recommended search information presented in S22, reverse search prompt information can be presented in the search sub-interface. For example, based on the above-mentioned analysis method one, if it is determined that there is no controversialness in at least one of the recommended search information presented in S22, no prompt may be given, or further combined with the analysis method two for re-judgment. For another example, based on the above-mentioned analysis method two, if it is determined that there is no controversialness in at least one of the recommended search information presented in S22, no prompt may be given, or further combined with the analysis method one for re-judgment; and so on.

[0200] Based on the above two aspects, the controversial nature of the recommended search information is determined. If at least one of the recommended search information currently displayed on the content details interface is determined to be controversial, a reverse search prompt is displayed on the search sub-interface. If at least one of the recommended search information currently displayed on the content details interface is determined to be non-controversial, a reverse search prompt is not displayed on the search sub-interface.

[0201] In the above embodiment, by combining the authenticity verification and the identification of opposing viewpoints to present the third search information, more real and multi-angle information is provided to the object, thereby meeting the diverse search needs of the object.

[0202] In an embodiment of the present application, the system can flexibly combine and sort search results based on the content of the object tags, through correlation analysis and object search tendencies, thereby providing a more personalized search experience.

[0203] Specifically, when the object wants to obtain different recommended search information, the object can also change the marking order. In addition, the object can also cancel some marks based on the current marks. When the object cancels at least one mark in the marked content fragments that have been searched or the marked content fragments that are not of interest, the mark cancellation operation is triggered for the at least one content fragment, and the recommended search information displayed in the content details interface is updated. The updated recommended search information is generated based on the currently marked content fragments.

[0204] Among them, after the object cancels the marking operation of at least one content segment, if the at least one content segment that was canceled is the most recently marked, there is no need to re-perform semantic analysis in this case. New recommended search information is directly presented in the content details interface based on the previous semantic analysis results. This recommended search information is essentially the same as the recommended search information that has been presented before. For example, currently 8 content segments are marked, and the object cancels the 8th marked content segment, then new recommended search information can be directly presented in the content details interface based on the previous semantic analysis results of the segment details information contained in the 1-7 content segments. This new recommended search information is essentially the same as the recommended search information presented when the object marked the 7th content segment.

[0205] If the at least one content segment that was removed was not the most recently marked, semantic analysis will be performed again on the segment details contained in each currently marked content segment. New recommended search information will be generated based on the new semantic analysis results, and this recommended search information will be presented in the search sub-interface of the content details interface. For example, if eight content segments are currently marked and the subject removes the sixth marked content segment, semantic analysis will be performed again on the segment details contained in the remaining seven marked content segments. Based on the new semantic analysis results, new recommended search information will be presented in the content details interface.

[0206] As shown in Figure 20, it is a schematic diagram of another content details interface in an embodiment of the present application. Interface 201 in Figure 20 indicates that the object has marked three content segments in the current content details interface: content segment "Director B" with tag number 1, content segment "Movie C" with tag number 2, and content segment "Director F" with tag number 3. In the search sub-interface shown in S201 in Figure 20, the recommended search information corresponding to the above three content segments is presented, "Director B's Movie C plagiarized Director F". When the object has completed the search based on the above recommended search information, or is not interested in the recommended search information, or finds that a mark is incorrect, the mark can be canceled. For example, after the object cancels the mark of content segment "Director F" with tag number 3, interface 202 in Figure 20 may be presented. At this time, semantic analysis is performed based on the current marked content segments "Director B" and "Movie C", and new recommended search information "Director B's Movie C box office in L year" is generated based on the results of the semantic analysis, which is presented in the search sub-interface shown in S202 in interface 202.

[0207] In the above embodiment, after the object turns on the smart tag search function, he or she can arbitrarily mark the content fragments that he or she is interested in or has questions about, and recommended search information is generated in the search sub-interface based on the content fragments that the object has marked. After the object cancels the search of the marked content fragments or at least one mark of the marked content fragments that he or she is not interested in, the recommended search information in the search sub-interface is also updated accordingly, and is updated in time according to the marked content of the object, providing the object with a more personalized search experience.

[0208] In an embodiment of the present application, after the object completes the search for content segments of interest or content segments of doubt according to its own needs, it can also trigger a closing operation based on the switch control in the content details interface. At this time, the client responds to the closing operation triggered by the object based on the switch control, turns off the smart tag search function, cancels the display of the search sub-interface, and clears the tag identifiers corresponding to each marked content segment displayed in the content details interface.

[0209] As shown in Figure 21, it is a schematic diagram of a switch control in an embodiment of the present application. Interface 211 in Figure 21 shows that the object has triggered the on operation of the switch control in the content details interface. As shown in S211 in interface 211, the switch control is in the on state. The object has marked the content segment "Director B" with a tag number 1 and the content segment "Movie C" with a tag number 2. After semantic analysis of these two content segments, the generated recommended search information "Director B's L-year Movie C box office" is presented in the search sub-interface of the content details interface of interface 211. After the object completes the search for relevant content of these two content segments according to its own needs, the object triggers the off operation of the switch control in the content details interface, presenting interface 212 in Figure 21. As shown in S212 in interface 212, the switch control is in the off state. At this time, the smart tag search function is disabled, the search sub-interface in the content details interface is no longer displayed, and the tag information of the two content segments marked by the object is cleared simultaneously.

[0210] In addition, when the object has no marked content, after turning off the switch control, the bottom search sub-interface is retracted, that is, the search sub-interface is canceled.

[0211] In the above embodiment, when the object needs to be searched, the smart tag search function can be directly turned on through the switch control, providing the object with a more efficient search experience; when the object does not need to be searched, the smart tag search function can be directly turned off through the switch control, which does not prevent the object from browsing the content details interface later.

[0212] It should be noted that the above mainly illustrates the information search method in the embodiment of the present application from the client side. The following further illustrates the information search method in the embodiment of the present application from the server side:

[0213] 22 is a flowchart of an implementation of an information search method in an embodiment of the present application. Taking the server as the execution subject as an example, the specific implementation process of the method is as follows S221 to S223:

[0214] S221: The server obtains the content segment marked in the target content.

[0215] S222: The server generates at least one piece of recommended search information corresponding to the current marking operation by performing semantic analysis on the segment detail information contained in each content segment marked this time and before this time.

[0216] The content segment is determined by the client in response to a marking operation triggered on the target content. Each marking operation corresponds to marking at least one content segment. For details, please refer to the above embodiment, and the repeated parts will not be repeated.

[0217] In an embodiment of the present application, when an object browses a target content such as an information text or a web page, the client provides a tool to support the object to arbitrarily mark content segments in the target content and form a number (i.e., a mark sequence number in this article). The client collects the text and numbers of these marks, as well as the time sequence of the marks, and then sends this information to the server. Therefore, when the server obtains a content segment from the client, it also receives the mark information, time sequence, and other information corresponding to the content segment sent by the client.

[0218] Specifically, the client first provides an intelligent tag search function, that is, it provides a tool for the object to arbitrarily mark content fragments (such as text) in the content details interface and form a tag identifier. The client sends the object marked content fragments, corresponding tag identifiers, time sequence and other information to the server. After the server receives the marked content fragments, corresponding tag identifiers and time sequence and other information sent by the client, it performs semantic analysis on the content fragments marked this time and before this time sent by the client, and generates at least one recommended search information based on the results of the semantic analysis.

[0219] Specifically, semantic analysis is performed on the fragment details information contained in the content fragment, that is, understanding and integrating the content fragment, including analyzing the theme, keywords, relevance, etc. of the marked content fragment. This process can use natural language processing technology, including but not limited to word meaning understanding, entity recognition, keyword extraction and topic modeling, etc., to understand the theme, keywords, etc. of the marked content fragment. At the same time, it may also involve machine learning technology, such as cluster analysis, to integrate similar marked content fragments, so as to identify key concepts and themes, extract keywords and phrases, and thus generate at least one recommended search information corresponding to the content fragment. The following is a brief description of these natural language processing technologies:

[0220] Word meaning understanding: If the received content is a large text segment, it is first divided into a sequence of text units, specifically into several words, and each word is assigned a part of speech. This process can be achieved through machine learning methods. The meaning of each word can then be understood using models such as word embeddings (such as Word2Vec or GloVe). Word embedding models convert each word into a vector that represents the word's semantic information.

[0221] Based on the understanding of the meaning of each word in the content segment in the above-mentioned manner, words can be selected therefrom to construct recommended search information.

[0222] Entity Recognition: Based on the semantics of the words in the marked content segments obtained through the above word meaning understanding, named entity recognition (NER) technology can be further used to identify entities in the content segments, such as names of people, places, and organizations. In the embodiment of the present application, the content segments listed in Figures 3 to 15 include entities such as names of people (such as Director B) and movie names (such as "Movie C").

[0223] Keyword extraction: Next, you can use the Term Frequency-Inverse Document Frequency (TF-IDF) algorithm or the TextRank algorithm to evaluate the importance of any word in the content segment to the content segment. The more important a word is to the content segment, the greater the probability that the word is a keyword. The importance of a word is positively correlated with the number of times the word appears in the content segment. Based on the above algorithm, extract the keywords in the content segment.

[0224] Topic Modeling: Finally, topics within a content segment can be found through topic modeling techniques such as Latent Dirichlet Allocation (LDA).

[0225] Based on the keywords extracted, entities identified, topics, etc. in the above steps, at least one recommended search information of the content fragment can be generated, and then the correlation between these recommended search information and the content fragment can be calculated using metrics such as cosine similarity, and these recommended search information can be sorted in order of relevance from high to low.

[0226] It should be noted that the above-mentioned natural language processing technology process needs to be adjusted according to the actual content fragments. If the processing effect of a certain step is not good, other technologies or algorithms can be selected for supplementary processing. This article does not make specific restrictions.

[0227] S223: The server sends at least one recommended search information to the client, so that the client displays the mark identifier corresponding to each of the at least one content segment and the at least one recommended search information corresponding to the current marking operation in the content details interface of the target content.

[0228] For example, if the server only feeds back one recommended search information to the client after each object is marked, the server can deeply understand the specific content of the marked content segment and conduct a correlation analysis, combine all texts, and search for related content based on the database (all resources other than the current content), and infer the query with the strongest correlation (i.e., the highest semantic match) and the highest number of possible search results as the recommended search information. The server then feeds back the recommended search information to the client, so that the client presents at least one recommended search information in S22. Generally speaking, the more specific the source of the input content, the more biased the search results will be towards the source of the input content.

[0229] The specific presentation method on the client side can be found in the above embodiments, and the repeated parts will not be repeated.

[0230] Optionally, the at least one recommended search information may be displayed in a search sub-interface in the content details interface, such as in a search box in the search sub-interface.

[0231] It's important to note that the search box doesn't accept manual text entry; it only displays at least one query with the highest overall ranking. When a query matches an object, the object can be searched directly using that query. Compared to traditional input-based searches, users don't need to memorize or summarize content while browsing. Instead, they can search directly based on tagged content combined with machine learning to predict the object's intent.

[0232] Optionally, the content details interface also includes an option viewing control, such as S111 shown in Figure 11. After the object triggers the option viewing control, it can view the various recommended search information corresponding to this marking generated by the server through semantic analysis of the fragment details information contained in each content fragment marked this time and before, such as the at least one recommended search information obtained above, as well as other candidate search information.

[0233] In addition to sending the at least one recommended search information obtained above to the client, the server may also send candidate search information to the client, so that the client can respond to the viewing operation triggered by the option viewing control and display the first search information set corresponding to each marked content segment in the content details interface.

[0234] Among them, the first search information in each first search information set is at least one recommended search information and candidate search information generated based on each content segment marked this time and before this time, and the correlation between at least one recommended search information and each content segment meets the preset correlation condition, for example, the ranking of the correlation between at least one recommended search information and each content segment is within the specified order range, while the correlation between the candidate search information and each content segment does not meet the preset correlation condition, for example, the ranking of the correlation between the candidate search information and each content segment is not within the specified order range.

[0235] Specifically, based on the content segments sent by the client, the corresponding tag identifiers and the time sequence, the server performs semantic analysis on the segment details information contained in the content segments marked this time and before, and determines the core ideas and relevance of each content segment. After identifying and sorting the core ideas and relevance information, the server will generate multiple search information with search results, and sort these search information from high to low according to the relevance based on the relevance of these search information and the content segments marked by the object. At least one search information in the specified order range will be sent to the client as the at least one recommended search information recommended to the object, and the client will present it in the search sub-interface of the content details interface. At the same time, the server also needs to send other search information (i.e., candidate search information) to the client. The client will present the at least one recommended search information and candidate search information in the first search information set. The object can view the first search information set after triggering the option view control in the content details interface.

[0236] For example, after the object marks content segment 1, the server can generate multiple search information based on content segment 1. According to the relevance of these search information to content segment 1, these search information can be sorted and fed back to the client. Based on the sorting results, the client can select at least one to be directly displayed on the content details interface, and the remaining search information (i.e., candidate search information) will be displayed again when the object views the first search information set. Among them, when sorted in a specified order range such as TOP1, that is, the search information with the strongest relevance to content segment 1 can be used as recommended search information directly presented in the content details interface, and the remaining search information with lower rankings will be further presented when the object views the first search information set.

[0237] Similarly, after the object marks content fragment 2, the server can generate multiple search information based on content fragment 1 and content fragment 2. According to the correlation between these search information and content fragment 1 and content fragment 2, these search information can be sorted and fed back to the client. Based on the sorting results, the client can select at least one to directly present in the content details interface, and present the remaining search information (that is, candidate search information) in the first search information set of the object view.

[0238] The same principle applies to object marking of other content fragments. You can follow the same logic and will not repeat it here.

[0239] In an embodiment of the present application, when clustering and sorting the marked content segments based on the degree of association, first, each marked content segment is regarded as a node, and clustered based on the degree of association between all the marked content segments, those with stronger association are divided into one group, and those with weaker association are divided into different groups. Secondly, the clustering results are sorted based on the search needs of the object, so that the object can more easily find the content of interest, thereby improving the relevance of the search results and the object's search satisfaction.

[0240] For example, after an object tags content segments A, B, C, and D, content segment D is the object's tagged content segment. Among them, content segments A, B, and C are highly correlated and grouped together. Keywords and other information extracted from these three content segments are combined to generate search information 1. Content segments A and D are less correlated and grouped together. Keywords and other information extracted from these two content segments are combined to generate search information 2. When search information 1 and search information 2 are presented in the first search information set, based on the object's search needs, search information 1 is ranked ahead of search information 2, allowing the object to find the content they need more quickly.

[0241] Optionally, the relevance between the search information and the content segment includes at least one of the following:

[0242] The semantic relevance between search information and content snippets;

[0243] The temporal correlation between the search information and the marked time of the content segment.

[0244] Among them, the marking time sequence of each content segment of the object can reflect the search intention of the object.

[0245] In the embodiment of the present application, when the above two types are included, the relevance between a certain search information and a content segment can be the result of a weighted sum of the semantic relevance and the temporal relevance.

[0246] The semantic relevance between the search information and the content segment can be calculated using a metric such as the aforementioned cosine similarity. When the server calculates that the relevance of the object-tagged content segment to at least one corresponding search information is within a specified order range, such as the search information with the highest relevance or the search information with the highest and second highest relevance, the server can determine the at least one search information as recommended search information and send it to the client for presentation in the search sub-interface of the content details interface.

[0247] Among them, the temporal correlation between the search information and the marking time of the content segment is determined based on the object's marking behavior of the content segment. For example, if the time (timestamp) when the object marks a content segment is closer to the object's search operation, it can be considered that the search information generated by the server based on the content segment is more important. Therefore, when sorting the search information by correlation, the search information corresponding to the content segment most recently marked by the object will be given a higher weight.

[0248] In an embodiment of the present application, the object can freely mark content fragments that are of interest or doubt to the existing content, without having to convert them into keywords for search. Instead, the system generates a first search information set based on the marked content fragments, and the object selects at least one search information in the first search information set. The server then generates search results based on the at least one search information, thereby shortening the search path and improving search efficiency.

[0249] If the subject does not find search information that meets their search requirements within the first search information set, the server will perform semantic analysis on the segment details contained in each currently marked content segment within the target content to generate at least one second search information corresponding to the content segment. Each generated second search information is sent to the client, so that the client, in response to a viewing operation triggered by the option viewing control, displays the at least one second search information corresponding to each marked content segment in the content details interface. The subject can then stitch these second search information together in the content details interface to generate stitched search information.

[0250] The server generates at least one corresponding second search information according to the content segment, and generates a search result according to the object splicing operation, specifically including the following steps:

[0251] Processing object-tagged content: Using text processing and natural language processing techniques and pre-trained language models, perform semantic analysis and entity recognition on the fragment details contained in the object-tagged content fragments.

[0252] Generating second search information: For example, algorithms such as TF-IDF can be used to extract keywords from the object-tagged content segment to generate second search information. The extracted second search information can then be stored in a database for subsequent use.

[0253] Unlike the first search information described above, the second search information is generated based solely on the currently marked content segment, without considering any previously marked content segments. For example, if the object is tagged with content segment 1 as "Director B," the corresponding second search information "Director B" will be generated. If the object is tagged with content segment 2 as "Movie C," the corresponding second search information "Movie C" will be generated. If the object is tagged with content segment 3 as "Director F," the second search information "Director F" will be generated, and so on.

[0254] Second search information selection: The subject selects the required second search information on the content details interface.

[0255] Second search information splicing: The object will splice the selected second search information according to the order and combination of its own search requirements. This step involves client-side development technology, such as using JavaScript to implement the second search information splicing.

[0256] Generating spliced ​​search information: Finally, the system converts the spliced ​​second search information into spliced ​​search information. This step involves server development technology, such as using Python, Java, or Node.js to generate the spliced ​​search information.

[0257] In an embodiment of the present application, by flexibly combining and sorting the second search information, a more personalized search experience can be provided to the object, helping the object to explore the content of interest more deeply.

[0258] In this embodiment of the application, the server uses a natural language understanding model and search results to calculate the reverse search information. When the recommended search information contains content that is semantically opposite to the corresponding search result, a reverse search prompt button will be activated. If the subject agrees with the content of the recommended search information, they can search directly. If they disagree or want to see a contrary opinion, they can also directly and quickly retrieve relevant search results from the content pool.

[0259] Specifically, controversial recommended search information is determined in at least one recommended search information, and for each recommended search information in the at least one recommended search information, third search information is generated that has an opposing viewpoint to the search result corresponding to the recommended search information.

[0260] The generated third search information is sent to the client, so that the client displays the third search information associated with the at least one recommended search information in the content details interface in response to the reverse search operation triggered by the reverse search prompt information.

[0261] The reverse search prompt information is used to prompt a search for content that is contrary to the viewpoint of the search result corresponding to at least one recommended search information.

[0262] Whether the recommended search information is controversial is determined by at least one of the following methods:

[0263] Analysis method 1: By verifying the authenticity of the search results corresponding to the recommended search information, it is determined whether the recommended search information is controversial. The specific analysis process can be referred to the above embodiment, and the repeated parts will not be repeated here.

[0264] Analysis method 2 uses text classification technology to identify opposing viewpoints in the search results corresponding to the recommended search information, and based on the identification results, determines whether the recommended search information is controversial. The specific analysis process can be found in the above embodiment, and the repeated parts will not be repeated here.

[0265] In this embodiment of the present application, the server may generate the third search information through the following steps:

[0266] Understanding the original recommended search information: First, natural language processing (NLP) technology is used to understand the meaning and intent of the original recommended search information, such as using a pre-trained language model for semantic analysis.

[0267] Identifying Controversy: First, we identify the controversial nature of the search results for a recommended search by querying search engines and databases. Then, we use text classification techniques to identify different viewpoints or positions within the search results. If a recommended search result contains multiple viewpoints or positions with similar proportions, the recommended search is considered controversial.

[0268] Generating third-party search information: For controversial recommended search information, a semantically contradictory third-party search information can be generated by modifying keywords in the original recommended search information, using antonyms, adding negation words, etc. For example, for the recommended search information "the impact of global warming," a semantically contradictory third-party search information "the benefits of global warming" can be generated.

[0269] Verify the third search information: Finally, the server needs to verify whether the generated third search information is valid. This can be achieved by inputting the third search information into a search engine or database, and then checking whether the corresponding search results are semantically opposite to the search results corresponding to the original recommended search information.

[0270] By generating the third search information, the present application can better meet the forward and reverse search needs of the object, which can not only improve the efficiency of the search, but also help the object discover and obtain more relevant information, thereby improving the object's search experience.

[0271] Refer to FIG. 23 , which is a schematic diagram of a logic for generating recommended search information in an embodiment of the present application.

[0272] When the subject browses target content such as news text or web pages, he or she clicks to open the content details interface of the news or web pages;

[0273] At this time, the client can enable the smart tag search function, which provides a tool for the object to mark text at will and form a tag identifier. The object can continuously mark content fragments.

[0274] In response to the marking operation triggered by the object, the client can send the object marked content segment, marking identifier and marking time sequence to the server. After receiving the marked content segment, marking information and time sequence sent by the client, the server stores this information in the database.

[0275] Then, the server uses natural language processing technology to understand and identify the marked content fragments, outputs recommended search information within the specified order range based on the relevance ranking, and controls the client to present the recommended search information in the search sub-interface of the content details interface. The object-marked content fragments will be understood in segments to generate short recommended search information.

[0276] If the subject recognizes the recommended search information, the search result can be directly obtained based on the generated recommended search information with the highest correlation with the marked content segment.

[0277] If the object does not agree with the recommended search information, the object may enter the first search information set, which includes the above-mentioned recommended search information and other candidate search information (i.e., search information with relatively low correlation with the content segment marked by the object), or, the object may freely splice and combine the second search information to form spliced ​​search information based on the second search information. The object clicks on the candidate search information or the second search information in the first search information set to control the server to retrieve the corresponding search results.

[0278] At the same time, the server can also determine whether the generated recommended search information is "controversial." If a recommended search information is determined to be controversial, the server will generate a related third search information, which is intended to provide content that is semantically opposite to the original recommended search information.

[0279] In addition, when the object enters the first search information set, it can use the forward original search information to search, or use the third search information to quickly search and consume content from the opposite perspective.

[0280] It should be noted that the execution logic diagram of the information search method listed in Figure 23 above is only a simple example. In addition, the execution logic of other related information search methods is also applicable to the embodiments of this application, and will not be described in detail herein.

[0281] In summary, the embodiments of the present application provide a method of marking content fragments in multiple places, predicting object search tendencies, personalized sorting and rich search suggestions. The system will flexibly combine and sort the search results based on the content fragments marked by the object through correlation analysis and object search tendencies, thereby providing a more personalized search experience; in addition, compared with the previous long search process, the present invention gives the object the ability to search for multi-point marked content, shortening the search path, and the system can generate richer search suggestions and related content recommendations, helping the object to explore topics of interest more deeply, and providing the object with a more accurate and personalized search experience, thereby improving the object's satisfaction and reputation.

[0282] Currently, articles, news, and other information available on the market are extremely rich in content, describing numerous and complex events and involving a wide range of people and perspectives. This makes it difficult for the system to determine what the subject is interested in, and the resulting recommendations are not very accurate. The tagging function provided by this application is not limited to the structure of the tagged content. Regardless of the content structure, the system can output a query based on the tagged original content, whether it is strongly or weakly related to the original main content (although the tagged text is derived from the original text, it may not be the main event). This is similar to the prompt generated by artificial intelligence (a generative model based on a language model). The subject can use this query to search for the extended content that most interests them, allowing them to quickly select the ideal search results. With minimal input cost (tags), the subject gains very convenient search capabilities and can search for other related content through customized content. In contrast, current mainstream search methods still output the original content even if the subject enters long text. This innovative solution helps to improve subject stickiness and create higher commercial value through content-driven content.

[0283] Based on the same inventive concept, the present application also provides an information search device. As shown in FIG24 , it is a schematic diagram of the structure of the information search generation device 2400, which may include:

[0284] Display unit 2401, used to display the content details interface of the target content;

[0285] First responding unit 2402 is configured to, in response to a mark operation triggered on at least one content segment in the target content, display, in the content details interface, a mark identifier corresponding to each of the at least one content segment and at least one recommended search information corresponding to the current mark operation; the recommended search information is generated based on the segment details information included in each content segment marked by the current and previous mark operations;

[0286] The second responding unit 2403 is configured to display search results corresponding to the target recommended search information in response to a search operation triggered based on the target recommended search information in the at least one recommended search information.

[0287] Optionally, the first responding unit 2402 is specifically configured to:

[0288] In response to a marking operation triggered on the at least one content segment, at least one recommended search information corresponding to the marking operation is displayed in a search sub-interface of the content details interface.

[0289] Optionally, the content details interface further includes a switch control for smart search control; in response to a mark operation triggered on at least one content segment in the target content, before displaying the mark identifier corresponding to each of the at least one content segment and at least one recommended search information corresponding to the mark operation in the content details interface, the first responding unit 2402 is further configured to:

[0290] In response to an on-operation triggered by the switch control, displaying a search sub-interface in the content details interface;

[0291] Furthermore, the first response unit 2402 is specifically configured to:

[0292] In the search sub-interface, at least one recommended search information corresponding to this marking operation is displayed.

[0293] Optionally, the first responding unit 2402 is further configured to:

[0294] In response to a closing operation triggered by the switch control, the search sub-interface is canceled, and the mark identifiers corresponding to the respective content segments marked by the marking operation displayed in the content details interface are cleared.

[0295] Optionally, the content details interface further includes an option viewing control; the device further includes:

[0296] The third responding unit 2404 is configured to display, in response to a viewing operation triggered by the option viewing control, the first search information set corresponding to each content segment marked by the marking operation in the content details interface;

[0297] Among them, the first search information in the first search information set is at least one recommended search information and candidate search information generated based on each content segment marked through this and previous marking operations, and the correlation between at least one recommended search information and each content segment meets the preset correlation conditions, and the correlation between the candidate search information and each content segment does not meet the preset correlation conditions.

[0298] Optionally, the third responding unit 2404 is further configured to:

[0299] In response to a search operation triggered based on any one of the first search information, search results corresponding to any one of the first search information are displayed.

[0300] Optionally, the third responding unit 2404 is further configured to:

[0301] At least one second search information corresponding to each content segment marked by the marking operation is displayed in the content details interface, and the second search information is recommended search information generated based on the content segment marked by this marking operation.

[0302] Optionally, the third responding unit 2404 is further configured to:

[0303] In response to a search operation triggered based on at least one second search information, search results corresponding to the spliced ​​search information are displayed, where the spliced ​​search information is obtained by splicing the at least one second search information.

[0304] Optionally, the device further comprises:

[0305] The fourth response unit 2405 is configured to display reverse search prompt information in the content details interface, where the reverse search prompt information is used to prompt a search for content that is semantically opposite to the search result corresponding to the at least one recommended search information;

[0306] In response to a reverse search operation triggered based on the reverse search prompt information, third search information associated with at least one recommended search information is displayed in the content details interface, and the content semantics of the third search information and the search results corresponding to the associated recommended search information are opposite.

[0307] Optionally, the fourth responding unit 2405 is specifically configured to:

[0308] In response to a reverse search operation triggered based on the reverse search prompt information, displaying, in the search sub-interface of the content details interface, third search information associated with each of the at least one recommended search information and third search information associated with each of the candidate search information;

[0309] The candidate search information is generated based on the content segments marked by this and previous marking operations, and the relevance between the candidate search information and the content segments is lower than the relevance between the recommended search information and the content segments.

[0310] Optionally, the fourth responding unit 2405 is further configured to:

[0311] In response to a search operation triggered based on any one of the third search information, a search result corresponding to any one of the third search information is displayed.

[0312] Optionally, before presenting the reverse search prompt information in the content details interface, the fourth responding unit 2405 is further configured to:

[0313] Determine controversial recommended search information among at least one recommended search information currently presented on the content details interface.

[0314] Optionally, the fourth responding unit 2405 is further configured to determine whether the recommended search information is controversial by at least one of the following methods:

[0315] Determine whether the recommended search information is controversial by verifying the authenticity of the search results corresponding to the recommended search information;

[0316] Identify opposing viewpoints on the search results corresponding to the recommended search information; and determine whether the recommended search information is controversial based on the identification results.

[0317] Optionally, the device further comprises:

[0318] The fifth response unit 2406 is used to update the recommended search information displayed in the content details interface in response to a mark cancellation operation triggered for at least one content segment; the updated recommended search information is generated based on the various content segments currently marked by the mark operation in the target content.

[0319] Based on the same inventive concept, the present embodiment also provides another information search device. As shown in FIG25 , it is a schematic diagram of the structure of the information search device 2500, which may include:

[0320] The determining unit 2501 is configured to obtain content segments marked in the target content; the content segments are determined by the client in response to a marking operation triggered on the target content, and each marking operation corresponds to marking at least one content segment;

[0321] The generating unit 2502 generates at least one recommended search information corresponding to the current marking operation by performing semantic analysis on the segment details information contained in each content segment marked this time and before this time;

[0322] The feedback unit 2503 sends the at least one recommended search information to the client, so that the client displays the mark identifier corresponding to the at least one content segment and the at least one recommended search information corresponding to the current marking operation in the content details interface of the target content.

[0323] Optionally, the content details interface further includes an option viewing control; the generating unit 2502 is further configured to:

[0324] By performing semantic analysis on the fragment details contained in each content fragment marked this time and before, candidate search information corresponding to this marking operation is generated;

[0325] And, the feedback unit 2503 is further used to:

[0326] Sending the candidate search information to the client, so that the client displays the first search information set corresponding to each marked content segment in the content details interface in response to a viewing operation triggered by the option viewing control;

[0327] Among them, the first search information in each first search information set is at least one recommended search information and candidate search information generated based on the various content fragments marked this time and before, and the correlation between at least one recommended search information and the various content fragments meets the preset correlation conditions, and the correlation between the candidate search information and the various content fragments does not meet the preset correlation conditions.

[0328] Optionally, the relevance between the search information and the content segment includes at least one of the following, and the search information includes recommended search information and candidate search information:

[0329] The semantic relevance between search information and content snippets;

[0330] The temporal correlation between the search information and the marked time of the content segment.

[0331] Optionally, the content details interface further includes an option viewing control; the generating unit 2502 is further configured to:

[0332] For each marked content segment in the target content, generating at least one second search information corresponding to the content segment by performing semantic analysis on the segment detail information contained in the content segment;

[0333] The feedback unit 2503 is further configured to:

[0334] The generated respective second search information is sent to the client, so that the client displays at least one second search information corresponding to each content segment marked by the marking operation in the target content in the content details interface in response to the viewing operation triggered by the option viewing control.

[0335] Optionally, the generating unit 2502 is further configured to:

[0336] generating, for each piece of at least one piece of recommended search information, third search information having an opposite viewpoint to the search result corresponding to the recommended search information;

[0337] The feedback unit 2503 is further configured to:

[0338] The generated third search information is sent to the client so that the client responds to the reverse search operation triggered by the reverse search prompt information and displays the third search information associated with at least one recommended search information in the content details interface; the reverse search prompt information is used to prompt the search for content that is opposite to the search result corresponding to the at least one recommended search information.

[0339] Optionally, the generating unit 2502 is further configured to:

[0340] Determining controversial recommended search information among the at least one recommended search information displayed in the content details interface;

[0341] Determine whether the recommended search information is controversial by at least one of the following methods:

[0342] Determine whether the recommended search information is controversial by verifying the authenticity of the search results corresponding to the recommended search information;

[0343] Through text classification technology, opposing viewpoints are identified for the search results corresponding to the recommended search information; based on the identification results, it is determined whether the recommended search information is controversial.

[0344] Based on the above implementation, in an embodiment of the present application, when browsing the content details of the target content, the subject can directly mark the content fragments in the content details interface. For example, the subject can mark the content fragments of interest or the content fragments of doubt according to their own needs. Furthermore, the present application can perform semantic analysis on the fragment details information contained in each content fragment marked by the subject in this and previous times according to the marking operation of the subject during the browsing process, intelligently generate recommended search information, and display it to the subject. Since these recommended search information are generated based on the analysis of the content fragments marked by the subject, they can effectively reflect the search needs of the subject. In addition, the subject can further search for information based on the target recommended search information in the displayed recommended search information. In this search process, the subject only needs to mark the content fragments according to its own search needs, and does not need to spend time thinking and manually entering search keywords, which can effectively improve the efficiency of the search. In addition, in this way, on the basis of better meeting the search needs of the subject, it can also help the subject discover and obtain more relevant information, improve the search experience of the subject, and increase the richness of the information obtained by the subject.

[0345] For the convenience of description, the above parts are divided into modules (or units) according to their functions and described separately. Of course, when implementing this application, the functions of each module (or unit) can be implemented in the same or multiple software or hardware.

[0346] After introducing the information search method and apparatus according to an exemplary embodiment of the present application, an electronic device according to another exemplary embodiment of the present application will be introduced next.

[0347] Those skilled in the art will appreciate that various aspects of the present application can be implemented as systems, methods, or program products. Therefore, various aspects of the present application can be specifically implemented in the following forms: a complete hardware implementation, a complete software implementation (including firmware, microcode, etc.), or an implementation that combines hardware and software aspects, which may be collectively referred to herein as a "circuit," "module," or "system."

[0348] Based on the same inventive concept as the above-mentioned method embodiment, an electronic device is also provided in an embodiment of the present application. In one embodiment, the electronic device may be a server, such as server 120 shown in FIG1 . In this embodiment, the structure of the electronic device may be as shown in FIG26 , including a memory 2601, a communication module 2603, and one or more processors 2602.

[0349] Memory 2601 is used to store computer programs executed by processor 2602. Memory 2601 may primarily include a program storage area and a data storage area. The program storage area may store an operating system and programs required for running instant messaging functions, while the data storage area may store various instant messaging messages and operating instruction sets.

[0350] Memory 2601 may be a volatile memory, such as random-access memory (RAM); a non-volatile memory, such as read-only memory, flash memory, a hard disk drive (HDD), or a solid-state drive (SSD); or any other medium capable of carrying or storing a desired computer program in the form of instructions or data structures and accessible by a computer, but is not limited thereto. Memory 2601 may be a combination of the aforementioned memories.

[0351] The processor 2602 may include one or more central processing units (CPUs) or digital processing units, etc. The processor 2602 is configured to implement the above-mentioned method for generating recommended search information when calling the computer program stored in the memory 2601 .

[0352] The communication module 2603 is used to communicate with terminal devices and other servers.

[0353] The specific connection medium between the memory 2601, communication module 2603, and processor 2602 is not limited in the embodiments of the present application. In Figure 26, the embodiment of the present application shows that the memory 2601 and the processor 2602 are connected via a bus 2604. Bus 2604 is depicted in bold in Figure 26. The connection methods between other components are merely schematic and are not intended to be limiting. Bus 2604 can be divided into an address bus, a data bus, a control bus, etc. For ease of description, Figure 26 only uses a single bold line, but does not represent a single bus or a single type of bus.

[0354] The memory 2601 stores a computer storage medium, which stores computer executable instructions for implementing the information search method of the embodiment of the present application. The processor 2602 is used to execute the above information search method, as shown in Figure 2.

[0355] In another embodiment, the electronic device may also be other electronic devices, such as the terminal device 110 shown in FIG1 . In this embodiment, the structure of the electronic device may be as shown in FIG27 , including: a communication component 2710, a memory 2720, a display unit 2730, a camera 2740, a sensor 2750, an audio circuit 2760, a Bluetooth module 2770, a processor 2780, and other components.

[0356] The communication component 2710 is used to communicate with the server. In some embodiments, it may include a wireless fidelity (WiFi) module. The WiFi module is a short-range wireless transmission technology. Electronic devices can help users send and receive information through the WiFi module.

[0357] The memory 2720 can be used to store software programs and data. The processor 2780 executes various functions and data processing of the terminal device 110 by running the software programs or data stored in the memory 2720. The memory 2720 may include high-speed random access memory and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. The memory 2720 stores the operating system that enables the terminal device 110 to run. In the present application, the memory 2720 can store the operating system and various application programs, and may also store the computer program that executes the information search method of the embodiment of the present application.

[0358] The display unit 2730 can also be used to display information input by the user or provided to the user, as well as a graphical user interface (GUI) for various menus of the terminal device 110. Specifically, the display unit 2730 may include a display screen 2732 disposed on the front of the terminal device 110. The display screen 2732 may be configured in the form of a liquid crystal display, a light-emitting diode, etc. The display unit 2730 can be used to display the content details interface, search sub-interface, etc. in the embodiments of the present application.

[0359] The display unit 2730 can also be used to receive input digital or character information and generate signal input related to the user settings and function control of the terminal device 110. Specifically, the display unit 2730 may include a touch screen 2731 arranged on the front of the terminal device 110, which can collect user touch operations on or near it, such as clicking a button, dragging a scroll box, etc.

[0360] The touch screen 2731 can be covered on the display screen 2732, or the touch screen 2731 and the display screen 2732 can be integrated to realize the input and output functions of the terminal device 110. The integrated display screen can be simply called a touch screen. In this application, the display unit 2730 can display applications and corresponding operation steps.

[0361] Camera 2740 can be used to capture still images, and users can publish images captured by camera 2740 through an application. There can be one or more cameras 2740. The object generates an optical image through the lens and projects it onto the photosensitive element. The photosensitive element can be a charge coupled device (CCD) or a complementary metal-oxide-semiconductor (CMOS) phototransistor. The photosensitive element converts the optical signal into an electrical signal, which is then transmitted to processor 2780 for conversion into a digital image signal.

[0362] The terminal device may further include at least one sensor 2750, such as an acceleration sensor 2751, a distance sensor 2752, a fingerprint sensor 2753, and a temperature sensor 2754. The terminal device may also be configured with other sensors such as a gyroscope, a barometer, a hygrometer, a thermometer, an infrared sensor, a light sensor, and a motion sensor.

[0363] The audio circuit 2760, speaker 2761, and microphone 2762 provide an audio interface between the user and the terminal device 110. The audio circuit 2760 can convert the received audio data into an electrical signal and transmit it to the speaker 2761, which converts it into a sound signal for output. The terminal device 110 may also be equipped with a volume button for adjusting the volume of the sound signal. On the other hand, the microphone 2762 converts the collected sound signal into an electrical signal, which is received by the audio circuit 2760 and converted into audio data. The audio data is then output to the communication component 2710 for transmission to, for example, another terminal device 110, or the audio data is output to the memory 2720 for further processing.

[0364] The Bluetooth module 2770 is used to exchange information with other Bluetooth devices having a Bluetooth module through the Bluetooth protocol. For example, the terminal device can establish a Bluetooth connection with a wearable electronic device (such as a smart watch) that also has a Bluetooth module through the Bluetooth module 2770 to exchange data.

[0365] The processor 2780 is the control center of the terminal device. It uses various interfaces and lines to connect various parts of the entire terminal. By running or executing software programs stored in the memory 2720 and calling data stored in the memory 2720, it performs various functions of the terminal device and processes data. In some embodiments, the processor 2780 may include one or more processing units; the processor 2780 may also integrate an application processor and a baseband processor, wherein the application processor mainly processes the operating system, user interface, and application programs, and the baseband processor mainly processes wireless communications. It is understandable that the above-mentioned baseband processor may not be integrated into the processor 2780. In the present application, the processor 2780 can run the operating system, application programs, user interface display and touch response, as well as the information search method of the embodiment of the present application. In addition, the processor 2780 is coupled to the display unit 2730.

[0366] In some possible implementations, various aspects of the information search method provided in the present application can also be implemented in the form of a program product, which includes a computer program. When the program product is run on an electronic device, the computer program is used to enable the electronic device to execute the steps of the information search method according to various exemplary implementations of the present application described above in this specification. For example, the electronic device can execute the steps shown in Figure 2 or Figure 22.

[0367] The program product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or component, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0368] The program product of the embodiment of the present application may be a portable compact disc read-only memory (CD-ROM) and include a computer program, and can be run on an electronic device. However, the program product of the present application is not limited thereto. In this document, a readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with a command execution system, apparatus, or device.

[0369] A readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries a readable computer program. Such a propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable signal medium may also be any readable medium other than a readable storage medium that can transmit, propagate, or transfer a program for use by or in conjunction with a command execution system, apparatus, or device.

[0370] The computer program embodied on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0371] The computer program for performing the operations of the present application can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The computer program can be executed entirely on the user electronic device, partially on the user electronic device, as a separate software package, partially on the user electronic device and partially on a remote electronic device, or entirely on a remote electronic device or server. In cases involving remote electronic devices, the remote electronic device can be connected to the user electronic device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external electronic device (for example, using an Internet service provider to connect through the Internet).

[0372] It should be noted that although several units or subunits of the device are mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, depending on the embodiment of the application, the features and functions of two or more units described above can be embodied in a single unit. Conversely, the features and functions of a single unit described above can be further divided and embodied by multiple units.

[0373] Furthermore, although the operations of the method of the present application are described in a particular order in the accompanying drawings, this does not require or imply that the operations must be performed in this particular order, or that all illustrated operations must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps.

[0374] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain a computer-usable computer program.

[0375] The present application is described with reference to the flow chart and / or block diagram of the method, device (system) and computer program product according to the embodiment of the present application. It should be understood that each flow process and / or box in the flow chart and / or block diagram and the combination of the flow process and / or box in the flow chart and / or block diagram can be realized by computer program commands. These computer program commands can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processing machine or other programmable data processing device to produce a machine, so that the command executed by the processor of the computer or other programmable data processing device produces a device for realizing the function specified in one flow chart flow or multiple flows and / or one box or multiple boxes of the block diagram.

[0376] These computer program commands may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to operate in a specific manner, so that the commands stored in the computer-readable memory produce a manufactured product including a command device that implements the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0377] These computer program commands can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the commands executed on the computer or other programmable device provide steps for implementing the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0378] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present application.

[0379] Obviously, those skilled in the art may make various changes and modifications to this application without departing from the spirit and scope of this application. Thus, if these modifications and variations of this application fall within the scope of the claims of this application and their equivalents, this application is intended to include these modifications and variations.

Claims

1. An information search method, performed by an electronic device, comprising: Display the content details interface of the target content; In response to a marking operation triggered on at least one content segment in the target content, displaying a marking identifier corresponding to each of the at least one content segment and at least one recommended search information corresponding to the marking operation in the content details interface; The recommended search information is generated based on the segment detail information included in each content segment marked by the marking operation this time and before this time; In response to a search operation triggered based on target recommended search information among the at least one recommended search information, search results corresponding to the target recommended search information are displayed.

2. The method according to claim 1, wherein displaying at least one recommended search information corresponding to the marking operation in the content details interface comprises: In response to the marking operation triggered on the at least one content segment, the at least one recommended search information corresponding to the marking operation is displayed in the search sub-interface of the content details interface.

3. The method according to claim 1 or 2, wherein the content details interface further comprises a switch control for intelligent search control; Then, before displaying, in response to a marking operation triggered on at least one content segment in the target content, a marking identifier corresponding to each of the at least one content segment and at least one recommended search information corresponding to the marking operation in the content details interface, the method further includes: In response to an on operation triggered by the switch control, displaying a search sub-interface in the content details interface; And, displaying at least one recommended search information corresponding to the marking operation in the content details interface includes: In the search sub-interface, the at least one recommended search information corresponding to the marking operation is displayed.

4. The method of claim 3, further comprising: In response to a closing operation triggered by the switch control, the search sub-interface is cancelled, and the marking identifiers corresponding to the respective content segments marked by the marking operation and displayed in the content details interface are cleared.

5. The method according to any one of claims 1 to 4, wherein the content details interface further comprises an option viewing control; the method further comprises: In response to a viewing operation triggered by the option viewing control, displaying, in the content details interface, first search information sets corresponding to respective content segments marked by the marking operation; Among them, the first search information in the first search information set is the at least one recommended search information and candidate search information generated based on the various content fragments marked by the marking operation this time and before this time, and the association degree between the at least one recommended search information and the various content fragments meets the preset association condition, and the association degree between the candidate search information and the various content fragments does not meet the preset association condition.

6. The method of claim 5, further comprising: In response to a search operation triggered based on any one of the first search information, a search result corresponding to the any one of the first search information is displayed.

7. The method according to any one of claims 1 to 6, further comprising: At least one second search information corresponding to each content segment marked by the marking operation is displayed in the content details interface, and the second search information is recommended search information generated based on the content segment marked by the marking operation.

8. The method of claim 7, further comprising: In response to a search operation triggered based on at least one second search information, search results corresponding to the spliced ​​search information are displayed, where the spliced ​​search information is obtained by splicing the at least one second search information.

9. The method according to any one of claims 1 to 8, further comprising: Displaying reverse search prompt information in the content details interface, the reverse search prompt information being used to prompt a search for content that is semantically opposite to a search result corresponding to the at least one recommended search information; In response to a reverse search operation triggered based on the reverse search prompt information, third search information associated with each of the at least one recommended search information is displayed in the content details interface, and the search results corresponding to the third search information are semantically opposite to the search results corresponding to the associated recommended search information.

10. The method of claim 9, wherein in response to the reverse search operation triggered by the reverse search prompt information, displaying the third search information associated with each of the at least one recommended search information in the content details interface comprises: In response to a reverse search operation triggered based on the reverse search prompt information, displaying, in the search sub-interface of the content details interface, the third search information associated with each of the at least one recommended search information and the third search information associated with each of the candidate search information; The candidate search information is generated based on the content segments marked by the marking operation this time and before this time, and the relevance between the candidate search information and the content segments is lower than the relevance between the recommended search information and the content segments.

11. The method according to claim 9 or 10, further comprising: In response to a search operation triggered based on any one of the third search information, a search result corresponding to the any one of the third search information is displayed.

12. The method according to any one of claims 1 to 11, further comprising: Among the at least one recommended search information displayed in the content details interface, controversial recommended search information is determined.

13. The method of claim 12, wherein determining whether the recommended search information is controversial is performed by at least one of the following methods: Determining whether the recommended search information is controversial by verifying the authenticity of the search results corresponding to the recommended search information; Identifying opposing viewpoints on the search results corresponding to the recommended search information; According to the identification result, it is determined whether the recommended search information is controversial.

14. The method according to any one of claims 1 to 13, further comprising: In response to a mark cancellation operation triggered for the at least one content segment, updating the recommended search information displayed in the content details interface; The updated recommended search information is generated based on each content segment in the target content currently marked by the marking operation.

15. An information search method, performed by an electronic device, the method comprising: Get the content fragments marked in the target content; The content segment is determined by the client in response to a marking operation triggered on the target content, and each marking operation corresponds to marking at least one content segment; By performing semantic analysis on the fragment detail information contained in each content fragment marked this time and before this time, at least one recommended search information corresponding to the marking operation this time is generated; The at least one recommended search information is sent to the client, so that the client displays the mark identifier corresponding to each of the at least one content segment and the at least one recommended search information corresponding to the current marking operation in the content details interface of the target content.

16. The method according to claim 15, wherein the content details interface further comprises an option viewing control; the method further comprises: By performing semantic analysis on the fragment details information contained in each content fragment marked this time and before this time, candidate search information corresponding to this marking operation is generated; Sending the candidate search information to the client, so that the client displays the first search information set corresponding to each content segment marked by the marking operation in the content details interface in response to the viewing operation triggered by the option viewing control; Among them, the first search information in each first search information set is the at least one recommended search information and the candidate search information generated based on the various content fragments marked this time and before this time, and the correlation between the at least one recommended search information and the various content fragments meets the preset correlation condition, and the correlation between the candidate search information and the various content fragments does not meet the preset correlation condition.

17. The method according to claim 15 or 16, wherein the relevance between the search information and the content segment comprises at least one of the following, wherein the search information comprises the recommended search information and the candidate search information: The semantic relevance between the search information and the content segment; The temporal correlation between the search information and the marked time of the content segment.

18. The method according to any one of claims 15 to 17, wherein the content details interface further comprises an option viewing control; the method further comprises: For each marked content segment in the target content, generating at least one second search information corresponding to the content segment by performing semantic analysis on segment detail information contained in the content segment; The generated respective second search information are sent to the client, so that the client, in response to the viewing operation triggered based on the option viewing control, displays at least one second search information corresponding to each content segment marked by the marking operation in the target content in the content details interface.

19. The method according to any one of claims 15 to 18, further comprising: For each of the at least one recommended search information, generating third search information having an opposite viewpoint to the search result corresponding to the recommended search information; The generated respective third search information are sent to the client, so that the client responds to a reverse search operation triggered based on the reverse search prompt information, and displays the third search information associated with each of the at least one recommended search information in the content details interface; the reverse search prompt information is used to prompt a search for content that is opposite to the search result corresponding to the at least one recommended search information.

20. The method according to any one of claims 15 to 19, further comprising: Determining controversial recommended search information among the at least one recommended search information displayed in the content details interface; Determine whether the recommended search information is controversial by at least one of the following methods: Determining whether the recommended search information is controversial by verifying the authenticity of the search results corresponding to the recommended search information; Using text classification technology, identifying opposing viewpoints on the search results corresponding to the recommended search information; According to the identification result, it is determined whether the recommended search information is controversial.

21. An information search device, comprising: A display unit, used to display a content details interface of the target content; A first response unit, configured to, in response to a mark operation triggered on at least one content segment in the target content, display a mark identifier corresponding to each of the at least one content segment and at least one recommended search information corresponding to the mark operation in the content details interface; The recommended search information is generated based on the segment detail information included in each content segment marked by the marking operation this time and before this time; The second response unit is used to respond to a search operation triggered by target recommended search information in the at least one recommended search information, and display search results corresponding to the target recommended search information.

22. An information search device, comprising: A determination unit, used for obtaining a content segment marked in the target content; The content segment is determined by the client in response to a marking operation triggered on the target content, and each marking operation corresponds to marking at least one content segment; A generating unit, which generates at least one piece of recommended search information corresponding to the marking operation this time by performing semantic analysis on the segment detail information contained in each content segment marked this time and before this time; The feedback unit sends the at least one recommended search information to the client, so that the client displays the mark identifier corresponding to each of the at least one content segment and the at least one recommended search information corresponding to the marking operation in the content details interface of the target content.

23. An electronic device comprising a processor and a memory, wherein: The memory stores a computer program, and when the computer program is executed by the processor, the processor is caused to perform the steps of any one of the methods of claims 1 to 20.

24. A computer-readable storage medium comprising a computer program, wherein when the computer program is run on an electronic device, the computer program is used to cause the electronic device to execute the steps of the method according to any one of claims 1 to 20.

25. A computer program product, comprising a computer program, wherein the computer program is stored in a computer-readable storage medium; when a processor of an electronic device reads the computer program from the computer-readable storage medium, the processor executes the computer program, so that the electronic device performs the steps of any method described in claims 1 to 20.

Images