Video recommendation method and device, electronic equipment and storage medium

By matching word segmentation and index library for English query text, calculate the correlation between video and query text, and recommend videos with high correlation, the problem of insufficient search capabilities in English text in the existing technology is solved, and the effect of video recommendation is improved.

CN118897903APending Publication Date: 2024-11-05BEIJING IQIYI TECH CO LTD
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202411159831.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-22
Publication Date
2024-11-05

AI Technical Summary

Technical Problem

Most existing search engines are searching for Chinese text, and their search capabilities for English text are weak, resulting in poor video recommendation effects.

Method used

Provide a video recommendation method, by receiving the user's English query text, performing word segmentation, using a pre-constructed English index library to determine candidate videos, and calculate the correlation between the query text and the video based on the forward index information, with a high recommended correlation. Video at preset threshold.

Benefits of technology

Improves the search ability of English text and improves the accuracy and effectiveness of video recommendations.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118897903A_ABST
    Figure CN118897903A_ABST
Patent Text Reader

Abstract

The invention relates to a video recommendation method and device, electronic equipment and a storage medium, and the method comprises the steps: receiving a search request triggered by a user, and enabling the search request to carry an English query text; in response to the search request, performing word segmentation on the English query text to obtain a plurality of text units in the English query text; determining at least one candidate video matched with the plurality of text units from a pre-constructed English index database; obtaining forward index information corresponding to the at least one candidate video, and calculating a correlation degree between the English query text and each candidate video in the at least one candidate video based on the forward index information corresponding to the at least one candidate video; according to the correlation degree, target videos are recommended to the user, and the target videos are videos, of which the correlation degree with the English query text is higher than a preset threshold value, in the at least one candidate video. Therefore, the search capability of the English text can be improved, and the video recommendation effect is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of multimedia technology, and in particular to a video recommendation method, device, electronic device and storage medium. Background Art

[0002] In recent years, with the rapid development of multimedia technology, the amount of video data has shown explosive growth. However, when users use search engines to search for videos they want to watch, most existing search engines are oriented to Chinese texts, and their search capabilities for English texts are weak, resulting in poor video recommendation results. Therefore, how to improve the search capabilities of English texts to improve the video recommendation effect has become a technical problem that needs to be solved urgently. Summary of the invention

[0003] The present application provides a video recommendation method, device, electronic device and storage medium to solve the problem that most existing search engines search for Chinese texts and have weak search capabilities for English texts, resulting in poor video recommendation effects.

[0004] In a first aspect, an embodiment of the present application provides a video recommendation method, the method comprising:

[0005] Receiving a search request triggered by a user, wherein the search request carries an English query text;

[0006] In response to the search request, segmenting the English query text to obtain a plurality of text units in the English query text;

[0007] Determine at least one candidate video matching the plurality of text units from a pre-built English index library;

[0008] Obtaining forward index information corresponding to the at least one candidate video, and calculating the degree of relevance between the English query text and each candidate video in the at least one candidate video based on the forward index information corresponding to the at least one candidate video;

[0009] According to the relevance, a target video is recommended to the user, wherein the target video is a video among the at least one candidate video whose relevance to the English query text is higher than a preset threshold.

[0010] Optionally, the forward index information corresponding to the at least one candidate video includes an English title and meta information corresponding to the at least one candidate video;

[0011] The calculating, based on the positive index information corresponding to the at least one candidate video, the relevance between the English query text and each candidate video in the at least one candidate video comprises:

[0012] Determine whether each English text unit in the multiple text units is included in the English title corresponding to the target candidate video, and calculate a first score according to the determination result, wherein the target candidate video is any candidate video in the at least one candidate video;

[0013] Determine whether a special text unit among the multiple text units matches the cloud information corresponding to the target candidate video, and calculate a second score according to the matching result, wherein the special text unit includes a number, a text unit related to the number of video episodes, seasons, and parts;

[0014] The relevance between the English query text and the target candidate video is determined according to the first score and the second score.

[0015] Optionally, obtaining the forward index information corresponding to the at least one candidate video includes:

[0016] Obtaining identification information corresponding to the at least one candidate video;

[0017] Match the identification information corresponding to the at least one candidate video with the identification information in the preset mapping relationship to obtain the positive index information corresponding to the at least one candidate video, wherein the preset mapping relationship is used to characterize the mapping relationship between each identification information and each positive index information.

[0018] Optionally, determining at least one candidate video matching the multiple text units from a pre-built English index library includes:

[0019] Matching each English text unit in the multiple text units with an English text unit in the English index library to obtain a set of video identifiers corresponding to each English text unit in the multiple text units, wherein the English index library includes multiple English indexes, each English index is used to represent a corresponding relationship between an English text unit and a set of video identifiers;

[0020] The at least one candidate video is determined based on a set of video identifiers corresponding to each English text unit in the multiple text units.

[0021] Optionally, before determining at least one candidate video matching the multiple text units from the pre-built English index library, the method further includes:

[0022] Segment the English titles of the videos in the video library, and determine the video identifier set corresponding to each English text unit in the segmentation results of the videos according to the segmentation results of the videos;

[0023] The English index library is constructed based on the video identification set corresponding to each English text unit in the word segmentation results of each video.

[0024] Optionally, the segmenting of the English query text to obtain a plurality of text units in the English query text includes:

[0025] Segmenting the English query text to obtain a plurality of initial text units;

[0026] The multiple initial text units are rewritten and information expanded to obtain the multiple text units.

[0027] In a second aspect, an embodiment of the present application further provides a video recommendation device, the device comprising:

[0028] A receiving module, configured to receive a search request triggered by a user, wherein the search request carries an English query text;

[0029] A word segmentation module, configured to segment the English query text in response to the search request to obtain a plurality of text units in the English query text;

[0030] A first determination module, configured to determine at least one candidate video matching the plurality of text units from a pre-built English index library;

[0031] A calculation module, used to obtain the forward index information corresponding to the at least one candidate video, and calculate the correlation between the English query text and each candidate video in the at least one candidate video based on the forward index information corresponding to the at least one candidate video;

[0032] The recommendation module is used to recommend a target video to the user according to the relevance, wherein the target video is a video among the at least one candidate video whose relevance to the English query text is higher than a preset threshold.

[0033] In a third aspect, an embodiment of the present application further provides an electronic device, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus;

[0034] Memory, used to store computer programs;

[0035] The processor is used to implement the video recommendation method described in any one of the first aspects when executing the program stored in the memory.

[0036] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the video recommendation method described in any one of the first aspects is implemented.

[0037] The above technical solution provided by the embodiment of the present application has the following advantages over the prior art: the method provided by the embodiment of the present application receives a search request triggered by a user, wherein the search request carries an English query text; in response to the search request, the English query text is segmented to obtain multiple text units in the English query text; at least one candidate video matching the multiple text units is determined from a pre-built English index library; the positive index information corresponding to the at least one candidate video is obtained, and based on the positive index information corresponding to the at least one candidate video, the correlation between the English query text and each candidate video in the at least one candidate video is calculated; according to the correlation, a target video is recommended to the user, wherein the target video is a video in the at least one candidate video whose correlation with the English query text is higher than a preset threshold. In the above manner, when a search request carrying an English query text is received, at least one candidate video matching multiple text units in the English query text is obtained, and a target video whose correlation with the English query text is higher than a preset threshold is selected from the at least one candidate video for recommendation, thereby realizing the search function of the English query text, improving the search capability of the English text, and improving the video recommendation effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0039] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0040] One or more embodiments are exemplarily described by pictures in the corresponding drawings, and these exemplified descriptions do not constitute limitations on the embodiments. Elements with the same reference numerals in the drawings represent similar elements, and unless otherwise stated, the figures in the drawings do not constitute proportional limitations.

[0041] Figure 1 A flowchart of a video recommendation method provided in an embodiment of the present application;

[0042] Figure 2 A flowchart of another video recommendation method provided in an embodiment of the present application;

[0043] Figure 3A schematic diagram of the structure of a video recommendation device provided in an embodiment of the present application;

[0044] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0046] The disclosure below provides many different embodiments or examples to implement different structures of the present invention. In order to simplify the disclosure of the present invention, the parts and settings of specific examples are described below. Of course, they are only examples, and the purpose is not to limit the present invention. In addition, the present invention can repeat reference numbers and / or letters in different examples. This repetition is for the purpose of simplification and clarity, and does not itself indicate the relationship between the various embodiments and / or settings discussed.

[0047] In order to solve the problem that most existing search engines search for Chinese texts and have weak search capabilities for English texts, resulting in poor video recommendation effects, the present application provides a video recommendation method, device, electronic device and storage medium, which can improve the search capabilities of English texts, thereby improving the video recommendation effects.

[0048] See also Figure 1 , Figure 1 The following is a flow chart of a video recommendation method provided in an embodiment of the present application. Figure 1 As shown, the video recommendation method may include the following steps:

[0049] Step 101: Receive a search request triggered by a user, wherein the search request carries an English query text.

[0050] It should be noted that the video recommendation method provided in the embodiment of the present application can be implemented by a client with a search engine, or can be implemented by a client and a server together, and the embodiment of the present application is not specifically limited. When the video recommendation method is implemented by a client with a search engine, a user can trigger a search request through a client, and parse and process the search request to obtain a target video, and then recommend the target video to the user. When the video recommendation method is implemented by a client and a server together, a user can trigger a search request through a client and send it to a server, and after receiving the search request, the server parses and processes the search request to obtain a target video, and then recommends the target video to the client to show it to the user.

[0051] Specifically, the search request may carry an English query text, wherein the English query text is a query text containing English text, such as harrypotter 5, etc.

[0052] Step 102: In response to the search request, segment the English query text to obtain multiple text units in the English query text.

[0053] In this step, the English query text may be segmented, that is, the English query text may be segmented into multiple text units. Specifically, the English query text may be segmented according to spaces, or may be segmented according to semantic divisions, which is not specifically limited in the embodiments of the present application. For example, the English query text Query "harrypotter 5" may be segmented into two terms "harrypotter" and "5".

[0054] When performing word segmentation on English query text, you can directly call open source word segmentation tools such as HanLP, PyLTP, LAC, etc. for word segmentation.

[0055] Step 103: Determine at least one candidate video matching the multiple text units from the pre-built English index library.

[0056] Specifically, the pre-built English index library refers to the pre-built index library using the English text units obtained after the English title segmentation of each video and the identification information of each video. For example, assuming that the English text units obtained after the English title segmentation of the video with ID 1 are term1, term2, and term3; the English text units obtained after the English title segmentation of the video with ID 2 are term2 and term4; the English text units obtained after the English title segmentation of the video with ID 3 are term1 and term5; then the English index library built based on these three videos is as follows:

[0057]

[0058] In this step, when determining the candidate videos, a video identifier set corresponding to each text unit can be determined from the pre-built English index library, and then at least one candidate video can be determined based on the video identifier set corresponding to each English text unit. For example, assuming that the English index library is as shown in the above table, and the English search text to be searched this time includes term3 and term4, the candidate videos are the videos corresponding to ID1 and ID2.

[0059] Step 104: Obtain the forward index information corresponding to at least one candidate video, and calculate the relevance between the English query text and each candidate video in the at least one candidate video based on the forward index information corresponding to the at least one candidate video.

[0060] In this step, after determining at least one candidate video, the forward index information corresponding to the at least one candidate video can be obtained. The forward index information here can include the English title and meta information corresponding to the candidate video, such as the number of episodes, seasons and parts of the candidate video. Then, based on the forward index information corresponding to the at least one candidate video, the degree of correlation between the English query text and each candidate video in the at least one candidate video can be calculated. It should be noted that the higher the degree of correlation, the higher the correlation between the candidate video and the English search text, and the greater the possibility of being recommended; the lower the degree of correlation, the lower the correlation between the candidate video and the English search text, and the smaller the possibility of being recommended.

[0061] Step 105: recommend a target video to the user based on the relevance, wherein the target video is a video in at least one candidate video whose relevance to the English query text is higher than a preset threshold.

[0062] Specifically, the target video may be one or more videos in at least one candidate video whose relevance to the English query text is higher than a preset threshold. The preset threshold may be set according to actual needs and is not limited here.

[0063] In this embodiment, when a search request carrying an English query text is received, at least one candidate video matching multiple text units in the English query text can be obtained, and a target video having a relevance to the English query text higher than a preset threshold can be selected from the at least one candidate video for recommendation, thereby realizing the search function of the English query text, improving the search capability of the English text and improving the video recommendation effect.

[0064] Further, the forward index information corresponding to at least one candidate video includes an English title and meta information corresponding to at least one candidate video;

[0065] The above step 104, based on the positive index information corresponding to the at least one candidate video, calculates the relevance between the English query text and each candidate video in the at least one candidate video, including:

[0066] Determine whether each English text unit in the plurality of text units is included in the English title corresponding to the target candidate video, and calculate a first score according to the determination result, wherein the target candidate video is any candidate video in the at least one candidate video;

[0067] Determine whether a special text unit among the multiple text units matches the cloud information corresponding to the target candidate video, and calculate a second score according to the matching result, wherein the special text unit includes a number, a text unit related to the number of video episodes, seasons, and parts;

[0068] The relevance between the English query text and the target candidate video is determined according to the first score and the second score.

[0069] In one embodiment, when calculating the relevance between the target candidate video and the English query text, it is possible to first determine whether each of the multiple text units is included in the English title corresponding to the target candidate video, and calculate the first score based on the determination result. For example, assuming that the English query text includes 5 English text units, when the English title corresponding to the target candidate video includes 1 of the English text units, the first score is 20%; when the English title corresponding to the target candidate video includes 2 of the English text units, the first score is 40%; when the English title corresponding to the target candidate video includes 3 of the English text units, the first score is 60%; when the English title corresponding to the target candidate video includes 4 of the English text units, the first score is 80%; when the English title corresponding to the target candidate video includes 5 of the English text units, the first score is 100%. In addition, it is possible to determine whether a special text unit among the multiple text units matches the cloud information corresponding to the target candidate video, and calculate the second score based on the matching result. For example, assuming that the special text unit "5" in multiple text units means the 5th season, when the cloud information corresponding to the target candidate video is also the 5th season, the second score is 100%; when the cloud information corresponding to the target candidate video is not the 5th season, the second score is 0%. Then, based on the first score and the second score, the degree of relevance between the English query text and the target candidate video is determined. Specifically, the average of the first score and the second score can be calculated, or the weighted average of the first score and the second score can be calculated based on the weights of the first score and the second score. In the above manner, the degree of relevance between the English query text and each candidate video in at least one candidate video can be accurately calculated, so that accurate recommendations can be made to users based on the calculated degree of relevance.

[0070] Furthermore, the above step 104, obtaining the forward index information corresponding to at least one candidate video, includes:

[0071] Obtain identification information corresponding to at least one candidate video;

[0072] Match the identification information corresponding to at least one candidate video with the identification information in a preset mapping relationship to obtain the forward index information corresponding to at least one candidate video, wherein the preset mapping relationship is used to characterize the mapping relationship between each identification information and each forward index information.

[0073] It should be noted that the above preset mapping relationship is used to represent the mapping relationship between each identification information and each positive index information, which can be shown in the following table:

[0074]

[0075] In one embodiment, when obtaining the forward index information corresponding to each candidate video, the identification information corresponding to each candidate video in at least one candidate video may be obtained first, and then the identification information corresponding to each candidate video may be matched with the identification information in the preset mapping relationship to obtain the forward index information corresponding to each candidate video. In this way, it is convenient to subsequently calculate the relevance between the English query text and each candidate video based on the forward index information corresponding to each candidate video.

[0076] Furthermore, the above step 103, determining at least one candidate video matching the multiple text units from the pre-built English index library, includes:

[0077] Matching each English text unit in the multiple text units with an English text unit in an English index library to obtain a video identification set corresponding to each English text unit in the multiple text units, wherein the English index library includes multiple English indexes, each English index is used to represent a corresponding relationship between an English text unit and a video identification set;

[0078] At least one candidate video is determined based on a set of video identifiers corresponding to each English text unit in the multiple text units.

[0079] Specifically, the pre-built English index library refers to the pre-built index library using the English text units obtained after the English title segmentation of each video and the identification information of each video. For example, assuming that the English text units obtained after the English title segmentation of the video with ID 1 are term1, term2, and term3; the English text units obtained after the English title segmentation of the video with ID 2 are term2 and term4; the English text units obtained after the English title segmentation of the video with ID 3 are term1 and term5; then the English index library built based on these three videos is as follows:

[0080]

[0081] In one embodiment, when determining the candidate video, a video identification set corresponding to each text unit can be determined from a pre-built English index library, and then the union of the video identification sets corresponding to each English text unit is determined as the identification of the candidate video, thereby obtaining the candidate video. For example, assuming that the English index library is as shown in the above table, and the English search text to be searched this time includes term3 and term4, the candidate video is the video corresponding to ID1 and ID2. In this way, the candidate video corresponding to the English query text can be accurately obtained.

[0082] Furthermore, before the above step 103 of determining at least one candidate video matching the multiple text units from the pre-built English index library, the method further includes:

[0083] Segment the English titles of the videos in the video library, and determine the video identifier set corresponding to each English text unit in the segmentation results of the videos according to the segmentation results of the videos;

[0084] An English index library is constructed based on a set of video identifiers corresponding to each English text unit in the word segmentation results of each video.

[0085] In one embodiment, when constructing an English index library, the English titles of each video in the video library may be first segmented, and based on the segmentation results of each video, a video identifier set corresponding to each English text unit in the segmentation results of each video may be determined, and then the English index library may be constructed based on the video identifier set corresponding to each English text unit in the segmentation results of each video. In this way, it is convenient to subsequently determine candidate videos for the English query text based on the English index library.

[0086] Furthermore, the English query text is segmented to obtain multiple text units in the English query text, including:

[0087] Segment the English query text to obtain multiple initial text units;

[0088] A plurality of initial text units are rewritten and information expanded to obtain a plurality of text units.

[0089] In one embodiment, when the English query text is segmented, the English query text can be segmented to obtain multiple initial text units, and then the multiple initial text units are rewritten (such as correcting English spelling errors, missing words, and overlapping words) and information expansion (such as when the user only remembers the name of the leading actor, the name of the leading actor's related video can be expanded based on the leading actor's name) to obtain multiple text units. In this way, the segmentation results of the English query text can be obtained more accurately, thereby improving the accuracy of the search.

[0090] In one embodiment, the video recommendation process provided by the embodiment of the present application is as follows: Figure 2 As shown, the following steps may be specifically included:

[0091] 1. English index library construction: Based on the word segmentation results of the English video title (i.e., title) in the meta information library, an inverted index information from the English video title to the video content is constructed.

[0092] 2. Recall candidate video content. First, segment the English query text query to obtain the entity terms in the English query text query. Then, based on the entity terms, recall the candidate video content corresponding to the English query text query from the English index library in step 1.

[0093] 3. English query text query analysis: Determine whether the terms in the English query text query after word segmentation are all in the English title of the candidate video content, and determine whether the English query text query, when containing text such as numbers, season, and part, matches the meta information in the forward index information of the candidate video content.

[0094] 4. Relevance calculation. If the terms after word segmentation of the English query text query are all in the English title of the candidate video content, and the English query text query contains text such as numbers, season, part, etc., and matches the meta information in the forward index information of the candidate video content, it is an exact match; if the terms after word segmentation of the English query text query are partially in the English title of the candidate video content, it is a highly relevant match; if the terms after word segmentation of the English query text query are not in the English title of the candidate video content, it is irrelevant.

[0095] 5. Search result display: Display accurate matching and highly relevant video content.

[0096] In this embodiment, by segmenting and rewriting the English query text query and expanding the information, and then performing matching calculation with the English query text query, the full matching and partial matching functions of English search are realized. For example, when a user searches for "harrypotter 5", "harrypotter" can be partially matched through the English title, and 5 (the fifth part) can be matched through information expansion, and "Harry Potter and the Order of the Phoenix" is recalled as a result, which improves the user's search accuracy and content distribution efficiency.

[0097] See also Figure 3 , Figure 3 This is a schematic diagram of the structure of a video recommendation device provided in an embodiment of the present application. Figure 3 As shown, the video recommendation device 300 includes:

[0098] The receiving module 301 is used to receive a search request triggered by a user, wherein the search request carries an English query text;

[0099] A word segmentation module 302 is used to segment the English query text in response to the search request to obtain multiple text units in the English query text;

[0100] A first determination module 303 is used to determine at least one candidate video matching the multiple text units from a pre-built English index library;

[0101] A calculation module 304 is used to obtain the forward index information corresponding to at least one candidate video, and calculate the relevance between the English query text and each candidate video in the at least one candidate video based on the forward index information corresponding to the at least one candidate video;

[0102] The recommendation module 305 is used to recommend a target video to the user according to the relevance, wherein the target video is a video whose relevance to the English query text is higher than a preset threshold among at least one candidate video.

[0103] Further, the forward index information corresponding to at least one candidate video includes an English title and meta information corresponding to at least one candidate video; the calculation module 304 includes:

[0104] a first calculation submodule, configured to determine whether each English text unit in the plurality of text units is included in the English title corresponding to the target candidate video, and to calculate a first score according to the determination result, wherein the target candidate video is any candidate video in the at least one candidate video;

[0105] A second calculation submodule is used to determine whether a special text unit among the multiple text units matches the cloud information corresponding to the target candidate video, and calculate a second score according to the matching result, wherein the special text unit includes a number, a text unit related to the number of video episodes, seasons, and parts;

[0106] The first determination submodule is used to determine the relevance between the English query text and the target candidate video according to the first score and the second score.

[0107] Furthermore, the calculation module 304 also includes:

[0108] A first acquisition submodule is used to acquire identification information corresponding to at least one candidate video;

[0109] The first matching submodule is used to match the identification information corresponding to at least one candidate video with the identification information in a preset mapping relationship to obtain the positive index information corresponding to the at least one candidate video, wherein the preset mapping relationship is used to characterize the mapping relationship between each identification information and each positive index information.

[0110] Further, the first determining module 303 includes:

[0111] A second matching submodule is used to match each English text unit in the multiple text units with an English text unit in an English index library to obtain a video identification set corresponding to each English text unit in the multiple text units, wherein the English index library includes multiple English indexes, each English index is used to represent the corresponding relationship between the English text unit and the video identification set;

[0112] The second determination submodule is used to determine at least one candidate video based on a set of video identifiers corresponding to each English text unit in the plurality of text units.

[0113] Furthermore, the device 300 also includes:

[0114] The second determination module is used to segment the English titles of each video in the video library, and determine the video identification set corresponding to each English text unit in the segmentation results of each video according to the segmentation results of each video;

[0115] The construction module is used to construct an English index library based on a set of video identifiers corresponding to each English text unit in the word segmentation results of each video.

[0116] Furthermore, the word segmentation module 302 includes:

[0117] The word segmentation submodule is used to segment the English query text to obtain multiple initial text units;

[0118] The rewriting and information expansion submodule is used to rewrite and expand the information of multiple initial text units to obtain multiple text units.

[0119] It should be noted that the device 300 can implement the video recommendation method provided by the aforementioned method embodiment and can achieve the same technical effect, which will not be described in detail here.

[0120] like Figure 4 As shown, the embodiment of the present application further provides an electronic device, including a processor 411, a communication interface 412, a memory 413 and a communication bus 414, wherein the processor 411, the communication interface 412, and the memory 413 communicate with each other through the communication bus 414.

[0121] Memory 413, used for storing computer programs;

[0122] In one embodiment of the present application, the processor 411 is used to implement the video recommendation method provided by any one of the aforementioned method embodiments when executing the program stored in the memory 413.

[0123] An embodiment of the present application further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the video recommendation method provided by any one of the aforementioned method embodiments is implemented.

[0124] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, i.e., they may be located in one place, or may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0125] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a general hardware platform, and of course, by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the relevant technology can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0126] It should be understood that the terms used herein are only for the purpose of describing specific example embodiments and are not intended to be limiting. Unless the context clearly indicates otherwise, the singular forms "one", "an" and "said" as used herein may also be meant to include plural forms. The terms "include", "comprise", "contain", and "have" are inclusive, and therefore specify the existence of stated features, steps, operations, elements and / or parts, but do not exclude the existence or addition of one or more other features, steps, operations, elements, parts, and / or combinations thereof. The method steps, processes, and operations described herein are not interpreted as necessarily requiring them to be performed in the specific order described or illustrated, unless the execution order is clearly indicated. It should also be understood that additional or alternative steps may be used.

[0127] The foregoing is merely a specific embodiment of the present invention, which enables those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A video recommendation method, characterized in that: The method comprises: Receiving a search request triggered by a user, wherein the search request carries an English query text; In response to the search request, segmenting the English query text to obtain a plurality of text units in the English query text; Determine at least one candidate video matching the plurality of text units from a pre-built English index library; Obtaining forward index information corresponding to the at least one candidate video, and calculating the degree of relevance between the English query text and each candidate video in the at least one candidate video based on the forward index information corresponding to the at least one candidate video; According to the relevance, a target video is recommended to the user, wherein the target video is a video among the at least one candidate video whose relevance to the English query text is higher than a preset threshold.

2. The method according to claim 1, characterized in that: The forward index information corresponding to the at least one candidate video includes an English title and meta information corresponding to the at least one candidate video; The calculating, based on the positive index information corresponding to the at least one candidate video, the relevance between the English query text and each candidate video in the at least one candidate video comprises: Determine whether each English text unit in the multiple text units is included in the English title corresponding to the target candidate video, and calculate a first score according to the determination result, wherein the target candidate video is any candidate video in the at least one candidate video; Determine whether a special text unit among the multiple text units matches the cloud information corresponding to the target candidate video, and calculate a second score according to the matching result, wherein the special text unit includes a number, a text unit related to the number of video episodes, seasons, and parts; The relevance between the English query text and the target candidate video is determined according to the first score and the second score.

3. The method according to claim 1, characterized in that The obtaining the forward index information corresponding to the at least one candidate video includes: Obtaining identification information corresponding to the at least one candidate video; Match the identification information corresponding to the at least one candidate video with the identification information in the preset mapping relationship to obtain the positive index information corresponding to the at least one candidate video, wherein the preset mapping relationship is used to characterize the mapping relationship between each identification information and each positive index information.

4. The method according to claim 1, characterized in that The step of determining at least one candidate video matching the plurality of text units from a pre-built English index library comprises: Matching each English text unit in the multiple text units with an English text unit in the English index library to obtain a set of video identifiers corresponding to each English text unit in the multiple text units, wherein the English index library includes multiple English indexes, each English index is used to represent a corresponding relationship between an English text unit and a set of video identifiers; The at least one candidate video is determined based on a set of video identifiers corresponding to each English text unit in the multiple text units.

5. The method according to claim 4, characterized in that Before determining at least one candidate video matching the multiple text units from the pre-built English index library, the method further includes: Segment the English titles of the videos in the video library, and determine the video identifier set corresponding to each English text unit in the segmentation results of the videos according to the segmentation results of the videos; The English index library is constructed based on the video identification set corresponding to each English text unit in the word segmentation results of each video.

6. The method according to claim 1, characterized in that The word segmentation of the English query text to obtain multiple text units in the English query text includes: Segmenting the English query text to obtain a plurality of initial text units; The multiple initial text units are rewritten and information expanded to obtain the multiple text units.

7. A video recommendation device, characterized in that: The device comprises: A receiving module, configured to receive a search request triggered by a user, wherein the search request carries an English query text; A word segmentation module, configured to segment the English query text in response to the search request to obtain a plurality of text units in the English query text; A first determination module, configured to determine at least one candidate video matching the plurality of text units from a pre-built English index library; A calculation module, used to obtain the forward index information corresponding to the at least one candidate video, and calculate the correlation between the English query text and each candidate video in the at least one candidate video based on the forward index information corresponding to the at least one candidate video; The recommendation module is used to recommend a target video to the user according to the relevance, wherein the target video is a video among the at least one candidate video whose relevance to the English query text is higher than a preset threshold.

8. The device according to claim 7, characterized in that The forward index information corresponding to the at least one candidate video includes an English title and meta information corresponding to the at least one candidate video; The computing module includes: a first calculation submodule, configured to determine whether each English text unit in the plurality of text units is included in the English title corresponding to the target candidate video, and calculate a first score according to the determination result, wherein the target candidate video is any candidate video in the at least one candidate video; A second calculation submodule is used to determine whether a special text unit among the multiple text units matches the cloud information corresponding to the target candidate video, and calculate a second score according to the matching result, wherein the special text unit includes a number, a text unit related to the number of video episodes, seasons, and parts; The first determination submodule is used to determine the relevance between the English query text and the target candidate video according to the first score and the second score.

9. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; Memory, used to store computer programs; The processor is used to implement the video recommendation method described in any one of claims 1 to 6 when executing the program stored in the memory.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the video recommendation method according to any one of claims 1 to 6 is implemented.

Citation Information

Patent Citations

  • Generation method for relevant video data bank, recommendation method and recommendation system for relevant videos

    CN103164405A

  • Video searching method and device

    CN104731938A

  • Video searching method and device and terminal

    CN106874502A

  • Mass file retrieval method and device, computer equipment and storage medium

    CN108874956A

  • System and method for providing recommended information

    KR101487540B1