Methods, devices, equipment, media, and computer programs for expanding content

Abstract

Embodiments of this disclosure relate to methods, apparatus, devices, media, and program products for expanding content. The method includes the step of determining a set of second search terms associated with a first search term in a set of search terms, wherein the number of original content items associated with the first search term is less than a predetermined number threshold. The method further includes the step of determining a first target content item associated with the first search term, which is used to expand the content associated with the first search term, based on a set of second target content items associated with the second search term. This method makes it possible to expand the recall content associated with other search terms by one search term, and to recall more content as the search terms are searched. In this way, the effectiveness of recommending content for search terms can be increased, and the user experience can be enhanced.

Description

Technical Field

[0001] The embodiments of the present disclosure generally relate to the field of data processing technology, and more specifically to a method, apparatus, device, medium, and program product for expanding content.

Background Art

[0002] With the popularization of the Internet and the increasing dependence of people on the Internet, the amount of data existing on the Internet is becoming increasingly large. This increasing data gradually forms a large amount of data resources, and it is becoming increasingly important to process and classify these data. Along with the continuous increase in Internet data, various search and recommendation systems have begun to be applied. These search and recommendation systems can recommend content to users based on the search terms selected or input by the users, so that users can quickly obtain the required content from a large amount of data.

[0003] Currently, search and recommendation systems usually adopt a funnel structure, which includes a three-stage structure of recall, ranking, and re-ranking. Recall refers to selecting as many results related to the search terms as possible from a database containing a large amount of information and data. Ranking is the preliminary screening and ranking of the recalled results. Re-ranking is to perform accurate screening and ranking on the ranking results and select the optimal small number of results.

Summary of the Invention

[0004] The embodiments of the present disclosure provide a method, apparatus, device, medium, and program product for expanding content.

[0005] A first aspect of this disclosure provides a method for expanding content. This method includes the step of determining a set of second search terms associated with a first search term in a set of search terms, wherein the number of original content items associated with the first search term is less than a predetermined number threshold. This method further includes the step of determining a first target content item associated with the first search term, which is used to expand the content associated with the first search term, based on a set of second target content items associated with the second search term.

[0006] A second aspect of this disclosure provides a device for expanding content. The device includes an association module configured to determine a set of search terms associated with a first search term in a set of search terms, wherein the number of original content items associated with the first search term is less than a predetermined number threshold, and an expansion module used to expand the content associated with the first search term and configured to determine a first target content item associated with the first search term, based on a set of target content items associated with the set of search terms.

[0007] In a third aspect of the present disclosure, the present disclosure provides an electronic device comprising at least one processor and a storage device for storing at least one program which, when executed by the at least one processor, causes the at least one processor to implement a method according to a first aspect of the present disclosure.

[0008] In a fourth aspect of the present disclosure, a computer-readable storage medium is provided, which stores a computer program that implements the method according to the first aspect of the present disclosure when the computer program is executed by a processor.

[0009] A fifth aspect of this disclosure provides a computer program product which includes computer-executable instructions that are physically stored on a non-volatile computer-readable medium and, when executed by a computer, cause the computer to perform a method provided in the first aspect of this disclosure.

[0010] It should be understood that the content described in this section is not intended to limit any essential or important features of the embodiments of the Disclosure, nor to limit the scope of the Disclosure. Other features of the Disclosure will be readily apparent from the following description. [Brief explanation of the drawing]

[0011] The above and other objects, features and advantages of this disclosure will become clearer by describing exemplary embodiments of this disclosure in more detail with reference to the drawings. Herein, in exemplary embodiments of this disclosure, the same reference numerals generally represent the same components.

[0012] [Figure 1] A schematic diagram of an exemplary environment in which the methods of the embodiments of this disclosure may be carried out is shown. [Figure 2] The following are schematic diagrams illustrating scenarios based on search term recommendation content in several embodiments of this disclosure. [Figure 3] A schematic diagram of a method for expanding content in some embodiments of this disclosure is shown. [Figure 4] A schematic diagram shows a method for determining a second search term associated with a first search term in some embodiments of this disclosure. [Figure 5] A schematic diagram of a method for determining the first target content in some embodiments of this disclosure is shown. [Figure 6] A schematic diagram of a method for expanding content in some embodiments of this disclosure is shown. [Figure 7] A schematic block diagram of a device for expanding content according to several embodiments of this disclosure is shown. [Figure 8]A schematic block diagram of an exemplary apparatus suitable for carrying out the contents of this disclosure is shown.

[0013] In each drawing, the same or corresponding reference numerals represent the same or corresponding parts. [Modes for carrying out the invention]

[0014] To ensure understanding, data obtained or used in accordance with this technical solution (including, but not limited to, the data itself, or the acquisition or use of the data) should comply with the requirements of applicable laws, regulations, and related provisions. In response to receiving a voluntary request from the user, prompt information will be sent to the user to clearly prompt the user that the operation requested to be performed requires the acquisition and use of the user's personal information. Thereafter, the user can choose whether or not to provide personal information to software or hardware such as electronic devices, application programs, servers, or storage media that perform the operation of the technical solution of this disclosure based on the prompt information. User interaction operations or interactions between the user and content relating to this disclosure, and data associated with user operations (including, but not limited to, data for analysis, stored data, displayed data, etc.) are all recorded, collected, or stored with the authorization of the user or with full permission of the parties, and the collection, use, and processing of the related data must comply with applicable national and local laws, regulations, and standards, and a corresponding operation entry will be provided for the user to choose to authorize or deny. In the technical solutions of the embodiments of this disclosure, the collection, storage, use, processing, transmission, provision, and disclosure of user-related information all comply with the provisions of relevant laws and regulations and do not violate public order and morals.

[0015] The embodiments of this disclosure will be described in further detail below with reference to the drawings. While the drawings show several embodiments of this disclosure, it should be understood that this disclosure is achievable in various ways and should not be construed as being limited to the embodiments described herein. Rather, these embodiments are provided for the purpose of providing a more thorough and complete understanding of this disclosure. It should be understood that the drawings and embodiments of this disclosure are for illustrative purposes only and do not limit the scope of protection of this disclosure.

[0016] In the descriptions of the embodiments of this disclosure, the term “including” and its synonyms should be understood as open inclusion, i.e., “including, but not limited to.” The term “based on” should be understood as “based at least in part.” The term “one embodiment” or “the embodiment” should be understood as “at least one embodiment.” The terms “first,” “second,” etc., may refer to different or the same object. Other explicit and implicit definitions may also be included below.

[0017] As explained above, in the recall phase of the search recommendation process, the search recommendation system needs to select content relevant to the search terms from a large amount of content from various information sources, and the subsequent ranking and re-ranking are both based on the recalled content. In some cases, when there is a small amount of recalled content for several search terms entered by the user, the final content obtained after ranking and re-ranking may be too little to recommend to the user, thus degrading the user experience.

[0018] To address at least the above and other potential problems, embodiments of the present disclosure propose a method for expanding content. This method can determine a set of second search terms associated with a first search term in a set of search terms for a first search term for which the number of recalled content items is less than a predetermined threshold, and determine a first target content associated with the first search term based on a set of second target content items associated with the second search terms.

[0019] This method allows for the merging of recalled content from other related search terms with limited recalled content, thereby expanding the recall results for those search terms. In this way, the recall results for search terms with limited original recalled content in the search recommendation system are expandable, allowing for a sufficiently large number of recalls for these terms, improving recommendation effectiveness and enhancing the user experience.

[0020] Embodiments of the present disclosure will now be described in detail with further reference to the drawings. Figure 1 illustrates an exemplary environment 100 in which a method according to an embodiment of the present disclosure can be implemented. As shown in Figure 1, the environment 100 may include a recommendation device 101 and a user device 102. The recommendation device 101 may include, but is not limited to, a server, minicomputer, mainframe computer, personal computer, multiprocessor system, handheld or laptop device, or mobile device (e.g., a mobile phone, personal digital assistant (PDA), media player, etc.). The recommendation device 101 may obtain search terms entered by the user from the user device 102, generate content to recommend to the user based on the search terms, and transmit such content to the user device 102.

[0021] The user device 102 may be any device capable of information input, such as a mobile phone, a tablet computer (pad), a notebook computer, a handheld computer, a smart TV, a PDA, a smart printer, smart home appliances, an in-vehicle terminal, wearable devices (such as smart watches, smart bracelets, smart glasses, etc.), virtual reality (VR) devices, augmented reality (AR) devices, etc., and the embodiments of the present application are not limited thereto. The user device 102 may receive the search term input by the user and send the search term to the recommendation device 101. The user device 102 may further receive the content corresponding to the search term from the recommendation device 101 and present it to the user.

[0022] As an example, the user device 102 may have a display interface, and the display interface may include a search term input box 103 and a content display area 104. The user device 102 may receive the search term input by the user through the search term input box 103 and display the content recommended to the user through the content display area 104. In the embodiments of the present disclosure, the content recommended to the user may include, but is not limited to, text, audio, video, etc.

[0023] In some embodiments, the recommendation device 101 may determine the content associated with the search term from a large amount of content by recall based on the search term obtained from the user device 102, and then determine the content recommended to the user from the content recalled by ranking and re-ranking. In some embodiments, the recommendation device 101 may merge the content recalled for one search term with the content recalled for another search term.

[0024] In some embodiments, the recommendation device 101 can locally store content associated with a search term. In some embodiments, the recommendation device 101 can merge content associated with one locally stored search term with content associated with another search term. It should be understood that environment 100 is merely one example of the embodiments of the disclosure and does not limit the disclosure. In some embodiments, the recommendation device 101 and the user device 102 may be the same device, that is, the user device 102 may perform the functions of the recommendation device 101.

[0025] Figure 2 illustrates a schematic diagram of a scenario 200 based on search term recommendation content in an embodiment of the present disclosure. Referring to Figure 2, scenario 200 includes a recommendation device 210, which may be, for example, a recommendation device 101 in environment 100. The recommendation device 210 can acquire a search term 220 and generate recommendation content 230 corresponding to the search term 220. The recommendation device 210 may include a recall module 211, a sorting module 212, and a memory 213. The recall module 211 can determine multiple pieces of content related to the search term 220 from all content from the cloud or local memory based on the search term 220, and the sorting module 212 can perform rough and fine sorting on the content determined by the recall module 211 to finally determine the recommendation content 230.

[0026] In some embodiments, the recall module 211 can determine thousands of content items from millions of content items via a predefined natural language processing (NLP) model, and the selection module 212 can determine the correlation between the content items determined by the recall module 211 and the search terms 220, and can determine the content items with a correlation greater than a predefined correlation threshold or the number of predefined content items with the highest correlation as recommended content items 230.

[0027] In some embodiments, the recall module 211 can merge the content associated with one recalled search term with the content associated with another recalled search term, and the merged content can be the final recall result for the other search term. In some embodiments, the recommendation device 210 can associate the recommended content 230 determined by the selection module 212 with the search term 220 and store it in memory 213, so that thereafter, when retrieving the search term 220, the recommendation device 210 can perform a nearline recall, that is, directly compare the retrieved search term with the search term stored in memory 213 to determine the recommended content 230. In some embodiments, after retrieving the search term 220, the recommendation device 210 can compare the search term 220 with the search term stored in memory 213 to determine multiple candidate contents, and the selection module 212 can further select the multiple candidate contents to determine new recommended content.

[0028] The above has described schematic diagrams of exemplary environments and scenarios in which the methods of the embodiments of this disclosure can be implemented, with reference to Figures 1 and 2. Below, methods for extending the content provided by this disclosure will be described with reference to Figures 3 to 6. Figure 3 illustrates schematic diagrams of methods for extending the content in several embodiments of this disclosure. The methods shown in Figure 3 can be implemented by the recommended equipment 101 shown in Figure 1, the recommended device 200 shown in Figure 2, or any other suitable equipment. Next, using the recommended device as the implementing body as an example, Method 300 provided by the embodiments of this disclosure will be described exemplary. Referring to Figure 3, Method 300 may include Blocks 302 and 304.

[0029] In block 302, the recommendation device determines a set of second search terms associated with a first search term in the search term set. The number of original content items associated with the first search term is less than a predetermined numerical threshold. In embodiments of the present disclosure, the search terms (including the first search term and the search terms in the search term set) may be text or labels entered by a user to search for content of interest to that user, and in some embodiments, the search terms may be labels obtained based on content classifications entered by the user or identified by an algorithmic model.

[0030] A set of search terms may contain one or more search terms, and the first search term may also contain one or more search terms. One or more pieces of content can be associated with either the first search term or any of the search terms in the set of search terms. In some embodiments, the content associated with a search term may be content obtained by the recommendation device through a recall process. In some embodiments, the content associated with a search term can be determined during a process in which the search term has been previously searched, and these search terms and content may be stored associated in the memory of the recommendation device, and may be stored exemplary, for example, in the following format. {query1:[doc1, doc2, doc3…]}; {query2:[doc2, doc4, doc5…]}; Here, "doc1", "doc2", and "doc3" may be content associated with the search term "query1", and "doc2", "doc4", and "doc5" may be content associated with the search term "query2". In some embodiments, the search term set and the first search term, and the content associated with each, may be, for example, search terms and content obtained from one or more user devices, or search terms and content obtained by a recommendation device from other devices. In some embodiments, the number of content items associated with a search term in the search term set is greater than or equal to a predetermined threshold. In some embodiments, the number of content items associated with the first search term may be zero, that is, there may be no content items associated with the first search term.

[0031] In some embodiments, the recommendation device can determine a set of second search terms from a set of search terms based on the similarity between the first search term and the search terms in the set of search terms, and in some embodiments, the recommendation device can determine a set of second search terms from a set of search terms based on a predefined classification model. The set of second search terms may consist of one or more search terms.

[0032] In block 304, the recommendation device determines the first target content associated with the first search term based on a set of second target content associated with the second search term. The first target content is used to expand the content associated with the first search term. As described in block 302 above, the second search term is a search term in the search term set, and one or more pieces of content are associated with the second search term, and for convenience of explanation, these are referred to as the second target content. The recommendation device can determine the first target content from the second target content and make that first target content the content associated with the new first search term. In other words, the recommendation device can supplement the content associated with the first search term by using the content associated with the second search term.

[0033] For example, the second search term associated with the first search term "query1" may be "query2," and the second target content associated with the second search term may be "doc2," "doc4," and "doc5," and the recommendation device can determine the first target content "doc2" and "doc5" from among the second target content. In some embodiments, the recommendation device can determine all content associated with the second search term as the first target content. In some embodiments, the recommendation device can determine the first target content based on the similarity between the second target content and the first search term. In some embodiments, the recommendation device can filter the second target content based on predefined filtering rules and determine the filtered second target content as the first target content.

[0034] By using the method 300 described above, the recommendation device can expand the content associated with the first search term by adding a second search term associated with the first search term, thereby allowing more content to be associated with the first search term, which originally had a small number of associated original contents. In this way, even if there are few recall results for the first search term during the search recommendation process, the number of contents ultimately recommended to the user can be increased by expanding the recall results with the second search term. This enhances the user experience.

[0035] In some embodiments, the recommendation device can store a first target content and a first search term in association. For example, the recommendation device can store the first target content and the first search term in association in local memory. Exemplaryly, the first target content may include "doc2" and "doc5", and the first search term may be "query1", and the recommendation device can store the first target content and the first search term in association as {query1:[doc2,doc5]}. In some embodiments, the first search term has an associated original content such as "doc1", and the recommendation device can store the original content, the first target content, and the first search term in association as {query1:[doc1,doc2,doc5]}. In this way, when the recommendation device subsequently receives the search term "query1", it can perform a nearline recall and directly determine the content associated with the search term using the data stored in local memory. Furthermore, as the content related to search terms stored in the local cache expands, nearline recall can determine more content associated with the search terms, improving the effectiveness of recommendations and enhancing the user experience.

[0036] In some embodiments, in block 302 described above, the recommended device can divide multiple search terms into a first search term and search terms in a set of search terms, and based on this, determine a second search term associated with the first search term from the set of search terms. Exemplarily, Figure 4 illustrates a schematic flowchart of Method 400 for determining a second search term associated with a first search term according to some embodiments of the present disclosure. Referring to Figure 4, Method 400 may include blocks 402 to 408.

[0037] In block 402, the recommendation device retrieves multiple search terms. These multiple search terms may be determined based on the historical search records of one or more users, and the recommendation device can retrieve user search records from one or more user devices and determine one or more search terms. In some embodiments, the recommendation device can retrieve multiple search terms from other devices located in the cloud and providing search recommendation services to users.

[0038] In block 404, the recommendation device determines a first search term and a set of search terms based on the search frequency of the search term within a predetermined time period. The predetermined time period may be, for example, a day or a week, and the search frequency of the search term may be how often the search term is searched by all users on the Internet. In some embodiments, the recommendation device is used to provide a search recommendation service to a particular user, and the search frequency of the search term may be how often that particular user searches for the term.

[0039] In block 406, the recommendation device determines the features of the first search term and the search terms in the search term set. Exemplarily, the recommendation device can input the first search term and the search terms in the search term set into a predefined recall model and, based on the predefined recall model, obtain the features corresponding to the first search term and the features corresponding to the search terms in the search term set. In some embodiments, the features may be, for example, embeddings. In some embodiments, the recommendation device can determine the features corresponding to the first search term and the features corresponding to the search terms in the search term set based on other types of embedding models.

[0040] In block 408, the recommendation device can determine a second search term associated with a first search term in a set of search terms, based on a predefined classification model. The classification model may include, but is not limited to, a k-nearest neighbor (KNN) algorithm model, a self-attention model, or a bidirectional encoder representation from transformers (BERT) model. The recommendation device inputs the characteristics of the first search term and the characteristics of the search terms in the set of search terms into the classification model and obtains a second search term associated with the first search term. In this way, even if the first search term contains multiple search terms, the second search term associated with each of the multiple search terms can be determined quickly.

[0041] Method 400 shown in Figure 4 is only one example of the embodiments of the present disclosure and is not limiting to the embodiments of the present disclosure. It should be understood that in some embodiments of the present disclosure, the recommendation device may distinguish between a first search term and a set of search terms by other means, and may determine a second set of search terms by other means. Exemplaryly, in some embodiments, in block 404, the recommendation device may divide multiple search terms into a first search term and search terms in a set of search terms based on the number of content associated with the search term. For example, search terms with a number of associated content items less than a predetermined threshold may be determined as the first search term, and search terms with a number of associated content items greater than or equal to the predetermined threshold may be determined as search terms in a set of search terms.

[0042] In some embodiments, in block 408, the recommendation device determines the similarity between a first search term and the search terms in the search term set based on the characteristics of the first search term and the characteristics of the search terms in the search term set, which can be determined, for example, by the twin-tower model. For a particular first search term, the recommendation device can check the similarity between the first search term and all the search terms in the search term set, and can rank the search terms in the search term set in descending order of similarity, and the recommendation device can determine a second search term from among them based on this ranking. Exemplarily, the recommendation device can determine the search term ranked 15th earlier as the second search term associated with the first search term.

[0043] In some embodiments, in block 302 described above, the second search term associated with the first search term determined by the recommendation device may include multiple search terms, each of which can be associated with multiple second target contents, and the recommendation device can merge these second target contents and determine the first target content based on this. In some embodiments, the recommendation device can filter and select the second target content; for example, the recommendation device can filter out content that is the same as the original content in the second target content. Alternatively, for example, the recommendation device can determine the similarity between the second target content and the first search term and select the first target content from the second target content based on the similarity.

[0044] Exemplary, Figure 5 illustrates a schematic flowchart of a method 500 for determining a first target content according to some embodiments of the present disclosure. Referring to Figure 5, the method 500 may include blocks 502 to 508. In block 502, the recommendation device can obtain original content associated with a first search term and second target content associated with a second search term. In some embodiments, the original content associated with the first search term may be content determined by the recommendation device from a large amount of content obtained from multiple data sources by a recall model, and the second target content associated with the second search term may be content determined by the recommendation device from a large amount of content by a recall model.

[0045] In some embodiments, the original content associated with the first search term may be content previously determined by a recall model and a selection model during the search recommendation process for the first search term and stored in the memory of the recommendation device; the second search term may be content determined by block 302 described above or by method 400 described above; and the second target content associated with the second search term may be content previously determined by the recommendation device during the search recommendation process for the second search term and stored in the memory of the recommendation device. The recommendation device can retrieve the original content and the second target content from memory. In some embodiments, the original content and the second target content may be content retrieved by the recommendation device from other recommendation devices or systems.

[0046] In block 504, the recommendation device filters out content that is identical to the original content in the second target content. The recommendation device's memory already contains the original content associated with the first search term, or in response to the recommendation device having already determined or obtained the original content associated with the first search term, it can remove content that is identical to the original content in the second target content and process only content that is not identical to the original content, thereby avoiding the appearance of duplicate content in the content associated with the first search term that is finally determined. In some embodiments, the second target content is content that merges the content associated with multiple second search terms, and there is duplicate content in it, and the recommendation device can remove the duplicate content in the second target content.

[0047] In block 506, the recommendation device can determine the similarity between a second target content and a first search term based on a predefined recall model. The recommendation device can input both the second target content and the first search term into a predefined recall model, which can determine the characteristics of the second target content and the first search term, and calculate the cosine similarity between the characteristics of the second target content and the characteristics of the first search term.

[0048] In block 508, the recommendation device determines the second target content as the first target content based on the similarity between the second target content and the first search term. Multiple second target content can be associated with a single first search term, and the recommendation device can rank these multiple second target content in descending order based on the cosine similarity between the features of the second target content and the first search term. In some embodiments, the recommendation device can determine the first target content as second target content whose cosine similarity is greater than a predetermined similarity threshold. In some embodiments, the recommendation device can determine the first target content as a top-ranking, predefined number of second target content.

[0049] Method 500 removes duplicate content in the second target content and further selects the second target content based on the recall model, thereby creating a higher relevance between the determined first target content and the first search term, providing more matched results in subsequent selection or nearline recall processes, and enhancing the recommendation effect.

[0050] In some embodiments, the recommendation device determines the similarity between the second target content and the first search term by block 506 in method 500, and a similarity also exists between the first target content, which is determined from the second target content in block 508, and the first search term. The recommendation device can store the first target content, the first search term, and the similarity between the first target content and the first search term in association. Subsequently, when the recommendation device receives a user search request for the first search term, it can directly provide the user with the first target content in descending order of similarity based on the first target content and similarity stored in memory. In this way, the efficiency of recommending content to the user can be increased.

[0051] In some embodiments, the recommendation device can store the first target content and the first search term in association after determining the first target content by method 300. Subsequently, when a search request for the first search term is received, the recommendation device can retrieve the first target content from memory, input the first target content into a predefined re-ranking model, perform further accurate selection of the first target content, and determine recommended content that is better matched in relation to the first search term.

[0052] In some embodiments, the recommendation device can use the stored first target content and first search term for subsequent search recommendations within a predetermined time period. For example, the predetermined time period may be one day. After the recommendation device determines and stores the first target content associated with the first search term, if the recommendation device receives a user search request for the first search term within the next day, it can recommend content to the user based directly on the content associated with the first search term stored in memory. After more than one day, the recommendation device can again perform method 300, or, using predefined recall and selection models, determine content to recommend to the user from multiple contents residing in the internet or cloud memory. Thus, even if the content associated with the first search term stored in the recommendation device's memory does not meet the user's needs due to the time being too long, the recommendation device can immediately adjust and always recommend content that matches the user's needs.

[0053] Figure 6 illustrates a schematic diagram of a method 600 for expanding content according to some embodiments of the present disclosure. The method shown in Figure 6 can be carried out by a device for expanding content, which may be, for example, a recommendation device 101 in scenario 100 or a recommendation device 210 in scenario 200. Method 600 can expand content associated with long-tail search terms based on high-frequency search terms. Next, method 600 will be described exemplary, with a recommendation device as the implementing entity. Referring to Figure 6, method 600 may include steps (1) to (9).

[0054] In step (1), the recommendation device inputs a high-frequency search term 601 and a long-tail search term 602 into the recall model 603. The high-frequency search term 601 may be a search term whose frequency of being searched within a predetermined time is above a predetermined search threshold, and the long-tail search term 602 may be a search term whose frequency of being searched within a predetermined time is below a predetermined search threshold. In step (2), the recommendation device obtains the features 604 of the high-frequency search term 601 and the features 605 of the long-tail search term 602 from the recall model 603. In step (3), the recommendation device inputs the features 605 of the long-tail search term 605 and the features 604 of the high-frequency search term into the classification model 606. In step (4), the classification model 606 determines a second search term 607 from the high-frequency search term 601. In step (5), the recommendation device determines a second target content 608 associated with the second search term 607. In step (6), the recommendation device determines the original content 610 associated with the long-tail search term 602, removes any content that overlaps with the original content 610 in the second target content 608, and obtains the selected second target content. In step (7), the recommendation device inputs the selected second target content 609 into the recall model 611, and inputs the long-tail search term features 605 into the recall model 611. The recall model 611 may be the same recall model as the recall model 603, or it may be a different model from the recall model 603.

[0055] In step (8), the recall model 611 generates features for the selected second target content 609 and determines the similarity between the selected second target content 609 and the long-tail search term 602 based on the features 605 of the long-tail search term. Furthermore, based on this similarity, it determines the first target content 612 from the selected second target content 609. In step (9), the recommendation device associates and stores the first target content 612, the original content 610, and the long-tail search term 602 in a locally located cache device. This method allows the recommendation device to expand the content associated with the long-tail search term and provide more content associated with the long-tail search term 602 in the subsequent nearline recall process. In this way, the effectiveness of search recommendations can be enhanced.

[0056] The methods provided by embodiments of this disclosure have been described above with reference to Figures 3 to 6, and below, the apparatus provided by embodiments of this disclosure will be described with reference to Figures 7 and 8. Figure 7 is a schematic diagram showing an apparatus 700 for expanding content according to some embodiments of this disclosure. Exemplarily, as shown in Figure 7, the apparatus 700 may include an association module 702 configured to determine a set of search terms associated with a first search term in a set of search terms, wherein the number of original content items associated with the first search term is less than a predetermined number threshold, and an expansion module 704 used to expand the content associated with the first search term and configured to determine a first target content item associated with the first search term, based on a set of target content items associated with the set of search terms.

[0057] In some embodiments, the first search term is a search term whose frequency of being searched within a predetermined time is less than a predetermined search threshold, and the search terms in the search term set are search terms whose frequency of being searched within a predetermined time is greater than or equal to a predetermined search threshold.

[0058] In some embodiments, the association module 702 includes a classification unit configured to determine multiple sets of second search terms associated with each of multiple first search terms from a set of search terms, based on a predefined classification model.

[0059] In some embodiments, the classification unit includes a first embedding determination unit configured to determine a first embedding for each of a plurality of first search terms; a second embedding determination unit configured to determine a second embedding for each of the search terms in a set of search terms; and a first association unit configured to determine a plurality of sets of second search terms associated with each of the plurality of first search terms based on the first embedding, the second embedding, and a predefined classification model.

[0060] In some embodiments, the association module 702 includes a first ranking unit configured to determine the rank of a search term in a set of search terms based on the similarity between a first search term and the search terms in the set of search terms, and a second association unit configured to determine a second search term based on the rank of the search terms in the set of search terms.

[0061] In some embodiments, the extension module 704 includes an acquisition unit configured to acquire a second target content and original content; a filtering unit configured to filter the second target content for content identical to the original content; and a first determination unit configured to determine the first target content based on the filtered second target content.

[0062] In some embodiments, the first decision unit includes a similarity determination unit configured to determine the similarity between filtered second target content and a first search term based on a predefined recall model, and a second decision unit configured to determine the first target content based on the similarity between filtered second target content and a first search term.

[0063] In some embodiments, the device 700 further includes a first storage module configured to store a first target content, a first search term, and the similarity between the first search term and the first target content in association with each other, and a first recommendation module configured to determine content to recommend to the user based on the similarity between the first target content and the first search term, in response to receiving a user's search request for the first search term.

[0064] In some embodiments, the device 700 further includes: a second storage module configured to store a first target content and a first search term in association; a second ranking module configured to determine the ranking of the first target content and the original content based on the first target content, the original content, and a predefined re-ranking model in response to receiving a user's search request for the first search term; and a second recommendation module configured to determine content to recommend to the user based on the ranking of the first target content and the original content.

[0065] Figure 8 illustrates a schematic block diagram of an exemplary device 800 that can be used to carry out embodiments of the present disclosure. Recommended device 101 in Figure 1 or recommended device 210 in Figure 2 can be realized by device 800. As shown in Figure 8, device 800 includes a central processing unit (CPU) 801, which can perform various appropriate operations and processes based on computer program instructions stored in read-only memory (ROM) 802 or computer program instructions loaded from storage unit 808 into random access memory (RAM) 803. RAM 803 may further store various programs and data necessary for the operation of device 800. The CPU 801, ROM 802, and RAM 803 are connected to each other via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0066] Multiple components in the device 800 are connected to the I / O interface 805 and include, for example, an input unit 806 such as a keyboard or mouse; an output unit 807 such as various types of displays or speakers; a storage unit 808 such as a magnetic disk or optical disk; and a communication unit 809 such as a network card, modem, or wireless communication transceiver. The communication unit 809 enables the device 800 to exchange information / data with other devices via computer networks such as the Internet and / or various telecommunication networks.

[0067] Each of the processes and operations described above, for example, methods 300, 400, 500, and / or 600, is performed by the processing unit 801. For example, in some embodiments, process 200 may be implemented as a computer software program tangibly contained in a machine-readable medium, for example, a storage unit 808. In some embodiments, part or all of the computer program may be loaded and / or installed into the device 800 via ROM 802 and / or communication unit 809. Once the computer program is loaded into RAM 803 and executed by CPU 801, one or more operations in methods 300, 400, 500, and / or 600 described above can be performed.

[0068] This disclosure may include methods, apparatus, systems, and / or computer program products. A computer program product may include a computer-readable storage medium on which computer-readable program instructions for performing each aspect of this disclosure are contained.

[0069] A computer-readable storage medium can be a tangible device capable of holding and storing instructions used by an instruction execution device. Computer-readable storage media include, but are not limited to, electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disk read-only memory (CD-ROM), digital multipurpose disks (DVDs), memory sticks, floppy disks, mechanical encoding devices such as punch cards or grooved structures on which instructions are stored, and any suitable combination of the above. As used herein, computer-readable storage media are not to be interpreted as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., optical pulses through fiber optic cables), or electrical signals transmitted through wires.

[0070] The computer-readable program instructions described herein may be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as a local area network (LAN), a wide area network (WAN), and / or a wireless network. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface of each computing / processing device receives computer-readable program instructions from the network and transfers the computer-readable program instructions for storage on the computer-readable storage medium of each computing / processing device.

[0071] Computer program instructions for performing the operations of the Disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source code or target code composed of any combination of one or more programming languages, wherein the programming languages ​​include object-oriented programming languages ​​such as Smalltalk and C++, and general procedural programming languages ​​such as the "C" language or similar programming languages. Computer-readable program instructions may be fully executed on a user computer, partially executed on a user computer, executed as a standalone software package, partially executed on a user computer, partially executed on a remote computer, or fully executed on a remote computer or server. In the case of a remote computer, it may be connected to the user computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or it may be connected to an external computer (for example, connected via the Internet by an Internet service provider). In some embodiments, each aspect of the present disclosure is achieved by personalizing computer-readable programmable electronic circuits, such as programmable logic circuits, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), by utilizing computer-readable programmable instruction state information.

[0072] The embodiments of this disclosure have been described above with reference to the flowcharts and / or block diagrams of the methods, apparatus (systems) and computer program products according to the embodiments of this disclosure. It should be understood that each block in the flowcharts and / or block diagrams, and each combination of blocks in the flowcharts and / or block diagrams, may be implemented by computer-readable program instructions.

[0073] These computer-readable program instructions are provided to a processing unit of a general-purpose computer, a dedicated computer, or other programmable data processing device, and when the instructions are executed through the processing unit of the computer or other programmable data processing device, it is possible to generate a device that produces a device that performs the functions / operations specified in one or more blocks of a flowchart and / or block diagram. These computer-readable program instructions may be stored in a computer-readable storage medium, and when these instructions operate a computer, a programmable data processing device, and / or other device in a particular manner, the computer-readable medium in which the instructions are stored includes a product which includes instructions that perform each mode of the functions / operations defined in one or more blocks of a flowchart and / or block diagram.

[0074] A computer-readable program instruction is loaded into a computer, another programmable data processing device, or other device, thereby generating a process implemented by the computer by executing a series of operational steps in the computer, another programmable data processing device, or other device, and thus the instruction executed by the computer, another programmable data processing device, or other device implements a function / operation defined in one or more boxes of a flowchart and / or block diagram.

[0075] The flowcharts and block diagrams in the drawings illustrate the feasible architectures, functions, and operations of devices, methods, and computer program products according to various embodiments of this disclosure. In this case, each block in the flowchart or block diagram may represent a module, program segment, or part of an instruction, and such module, program segment, or part of an instruction may contain one or more executable instructions for implementing a given logical function. In some alternative implementations, the functions assigned within a block may occur in a different order than those assigned in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, or in reverse order depending on the related functions. It should also be noted that each block in the block diagram and / or flowchart, and combinations of blocks in the block diagram and / or flowchart, may be implemented in a dedicated hardware-based system for performing a given function or operation, or in a combination of dedicated hardware and computer instructions.

[0076] While various embodiments of this disclosure have been described above, the above description is illustrative, not exhaustive, and not limited to the disclosed embodiments. Many modifications and changes will be apparent to those skilled in the art without departing from the scope and spirit of the illustrated embodiments. The terms used herein have been selected to best describe the principle, practical or market-based improvement of each embodiment, or to enable other those skilled in the art to understand the embodiments disclosed herein.

Claims

1. A method for expanding content, the method is A step of determining a set of second search terms associated with a first search term in a set of search terms, wherein the number of original content items associated with the first search term is less than a predetermined threshold; The process includes the step of determining a first target content associated with the first search term, which is used to expand the content associated with the first search term and is associated with the first search term, based on a set of second target content associated with the second search term. method.

2. The method according to claim 1, wherein the first search term is a search term whose frequency of being searched within a predetermined time is less than a predetermined search threshold, and the search terms in the set of search terms are search terms whose frequency of being searched within the predetermined time is equal to or greater than the predetermined search threshold.

3. The step of determining a set of second search terms associated with a first search term in a set of search terms is: The method according to claim 1 or 2, further comprising the step of determining a plurality of sets of second search terms associated with each of a plurality of first search terms from the set of search terms based on a predefined classification model.

4. The step of determining multiple sets of second search terms associated with each of multiple first search terms from the set of search terms based on a predefined classification model is: The steps include determining a first embedding for each of the multiple first search terms, The steps include determining a second embedding for each search term in the aforementioned set of search terms, The method according to claim 3, comprising the step of determining a plurality of sets of second search terms associated with each of a plurality of first search terms based on the first embedding, the second embedding, and a predefined classification model.

5. The step of determining a set of second search terms associated with a first search term in a set of search terms is: A step of determining the ranking of the search terms in the search term set based on the similarity between the first search term and the search terms in the search term set, The method according to claim 1 or 2, comprising the step of determining a second search term based on the ranking of the search terms in the set of search terms.

6. The step of determining the first target content associated with the first search term based on a set of second target content associated with the second search term is: The steps include acquiring the second target content and the original content, The steps include filtering out the same content from the second target content as the original content, The method according to claim 1 or 2, comprising the step of determining the first target content based on the second target content after filtering.

7. The step of determining the first target content based on the filtered second target content is: A step of determining the similarity between the filtered second target content and the first search term based on a predefined recall model, The method according to claim 6, comprising the step of determining the first target content based on the similarity between the filtered second target content and the first search term.

8. The steps include storing the first target content, the first search term, and the similarity between the first search term and the first target content in association with each other, The method according to claim 7, further comprising the step of determining content to recommend to the user based on the similarity between the first target content and the first search term, in response to receiving a user's search request for the first search term.

9. This method is The steps include storing the first target content and the first search term in association with each other, In response to receiving a user's search request for the first search term, the steps include determining the ranking of the first target content and the original content based on the first target content, the original content, and a predefined re-ranking model, The method according to claim 1, further comprising the step of determining content to recommend to the user based on the ranking of the first target content and the original content.

10. A device for expanding content, said device is An association module configured to determine a set of search terms associated with a first search term in a set of search terms, wherein the number of original content items associated with the first search term is less than a predetermined threshold; Includes an extension module used to extend the content associated with a first search term and configured to determine a first target content associated with the first search term, based on a set of target content associated with the set of search terms, Device.

11. Electronic equipment, said electronic equipment, A processor and a memory connected to the processor, The memory has stored instructions, and when the instructions are executed by the processor, it causes the electronic device to perform the method described in any one of claims 1 to 9. electronic equipment.

12. A computer-readable storage medium for storing computer-executable instructions that, when executed by a processor, implement the method described in any one of claims 1 to 9.

13. A computer program product that is physically stored on a non-volatile computer-readable medium and includes a computer-executable instruction that, when executed by a computer, causes the computer to perform the method described in any one of claims 1 to 9.

Images