Hotspot content recommendation method and device, computer device and readable storage medium

Abstract

The application provides a hotspot content recommendation method and device, computer equipment and a readable storage medium, which comprises: obtaining a search text and a plurality of albums corresponding to the search text in a sorted order; determining a hotspot album in a content platform from the plurality of albums; when the number of the hotspot album is greater than or equal to a preset threshold, merging keywords corresponding to the hotspot album; inputting the search text and the merged keyword set into a pre-trained classification model to determine whether the search text is a hotspot text; when it is determined that the search text is a hotspot text, reordering the remaining albums except the hotspot album to make the recommendation order of the hotspot album located in front of the recommendation order of the remaining albums; and sending the hotspot album and the remaining albums to a user according to the recommendation order. The application can effectively provide hotspot content to the user and improve the play quantity of the hotspot content in the content platform.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and more specifically, to a method, apparatus, computer device, and readable storage medium for recommending trending content. Background Technology

[0002] Currently, trending topics and related hot events have become popular among users. However, due to the lag in content platforms, these trending topics often have low click-through rates. Users often find outdated content when searching for trending topics, resulting in many trending topics failing to gain effective exposure. Therefore, how to effectively provide trending topics to users and increase their viewership on content platforms has become a key research focus. Summary of the Invention

[0003] One objective of this invention is to provide a method, apparatus, computer device, and readable storage medium for recommending trending content, in order to effectively extract trending content within a content platform and increase the click-through rate of trending content on the content platform. This invention can be achieved as follows:

[0004] In a first aspect, the present invention provides a method for recommending trending content, the method comprising:

[0005] Obtain the search text, and the sorted albums corresponding to the search text;

[0006] From the multiple albums mentioned, popular albums on the content platform are identified; wherein, the popular albums are matched with keywords of pre-acquired popular content;

[0007] When the number of popular albums is greater than or equal to a preset threshold, the keywords corresponding to the popular albums are merged.

[0008] The search text and the merged keyword set are input into a pre-trained classification model to determine whether the search text is a hot topic text.

[0009] When the search text is determined to be a hot topic, the remaining albums excluding the hot topic albums are reordered so that the hot topic albums are recommended before the remaining albums.

[0010] The popular albums and the remaining albums are sent to the user in the recommended order.

[0011] In a second aspect, the present invention provides a hot topic content recommendation device, including an acquisition module, a determination module, a merging module, a classification module, a sorting module, and a recommendation module;

[0012] The acquisition module is used to acquire the search text and multiple albums corresponding to the search text in sorted order;

[0013] The determining module is used to determine the hot albums in the content platform from the multiple albums; wherein the hot albums are matched with keywords of pre-acquired hot content;

[0014] The merging module is used to merge the keywords corresponding to the hot albums when the number of hot albums is greater than or equal to a preset threshold.

[0015] The classification module is used to input the search text and the merged keyword set into a pre-trained classification model to determine whether the search text is a hot topic text.

[0016] The sorting module is used to reorder the remaining albums other than the hot albums when the search text is determined to be hot text, so that the recommendation order of the hot albums is placed before the recommendation order of the remaining albums.

[0017] The recommendation module is used to send the popular albums and the remaining albums to the user according to the recommendation order.

[0018] Thirdly, the present invention provides a computer device including a processor and a memory, the memory storing a computer program executable by the processor, the processor being able to execute the computer program to implement the hot topic content recommendation method as described in the first aspect.

[0019] Fourthly, the present invention provides a readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the hot topic content recommendation method as described in the first aspect.

[0020] This invention provides a method, apparatus, computer device, and readable storage medium for recommending trending content. After obtaining the search text, multiple sorted albums corresponding to the search text can be retrieved. Then, based on a pre-stored set of keywords for trending content, trending albums matching the keywords of the trending content are determined. When the number of these trending albums exceeds a preset threshold, the search text is identified as a trending text, and the corresponding trending album is prioritized and sent to the user. This achieves the goal of uncovering trending content on the content platform. Furthermore, by setting the recommendation order of the trending albums before the recommendation order of the remaining albums, the trending albums can be sent to the user, solving the problem of low click-through rates and insufficient exposure on the content platform, and effectively improving the recommendation rate of trending content. Attached Figure Description

[0021] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 A schematic flowchart illustrating the hot topic content recommendation method provided in an embodiment of the present invention;

[0023] Figure 2 Another illustrative flowchart of the hot topic content recommendation method provided in the embodiments of the present invention;

[0024] Figure 3 A functional block diagram of the hot topic content recommendation device provided in the embodiments of the present invention;

[0025] Figure 4 A structural block diagram of a computer device provided in an embodiment of the present invention. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0027] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0028] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0029] In the description of this invention, it should be noted that if terms such as "upper," "lower," "inner," or "outer" are used to indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship in which the product of this invention is usually placed, they are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this invention.

[0030] Furthermore, the terms "first" and "second" are used only to distinguish descriptions and should not be interpreted as indicating or implying relative importance.

[0031] It should be noted that, where there is no conflict, the features in the embodiments of the present invention can be combined with each other.

[0032] Trending content refers to popular topics or content that receives widespread attention and discussion within a certain period. Most users obtain trending content through news, microblogs, and other channels. However, due to the lag in content platforms, the overall quality of this trending content is not as good as professionally produced thematic collections, resulting in lower click-through rates on content platforms. If users search for trending news, they are likely to be returned older, higher-quality content. Therefore, much trending content on content platforms fails to achieve effective exposure. Identifying this trending content and effectively recommending it to users can significantly increase click-through rates on content platforms. It can also cultivate users' habits of using content platforms to obtain timely news and other timely content, thereby improving user retention and enhancing user engagement.

[0033] Therefore, to effectively provide trending content to users of content platforms and increase the viewership of trending content on these platforms, this invention provides a trending content recommendation method. This method can be executed by a server, a terminal, or both. Specifically, the terminal can execute the trending content recommendation method of this invention via a client installed on it.

[0034] Please see Figure 1 , Figure 1 A schematic flowchart of the hot content recommendation method provided in the embodiments of this application may include the following steps:

[0035] S105: Obtain the search text, and the sorted albums corresponding to the search text.

[0036] In this embodiment of the invention, the search text comes from manual input or voice input by the user. The user can be a user of the content platform. For example, after the content platform receives the search text input by the user, it can match the albums related to the search text from the album library, sort these albums, and then display each album in front of the user according to the sorting results.

[0037] S106: Identify the most popular albums on the content platform from multiple albums; among them, match the keywords of the most popular albums with those of the pre-acquired popular content.

[0038] In this embodiment of the invention, trending content can be events that have received widespread attention and discussion over a period of time, and it can have a certain degree of timeliness and regionality. This embodiment of the invention can pre-store multiple sets of keywords corresponding to trending content. This trending content is mined from platforms or portals outside the content platform. The keyword sets can be used to associate trending albums within the content platform; that is, the trending albums are pre-determined by matching the keyword sets with various trending searches within the content platform, and then it is determined whether a trending album exists among the albums corresponding to the search text.

[0039] S107: When the number of popular albums is greater than or equal to a preset threshold, merge the keywords corresponding to the popular albums.

[0040] In this embodiment of the invention, a preset threshold for the number of trending albums is set. That is, only when the number of trending albums is greater than or equal to the preset threshold will the search text be considered to be a trending text. Otherwise, the search text is not a trending text, and there is no need to recommend trending albums to it.

[0041] S108: Input the search text and the merged keyword set into the pre-trained classification model to determine whether the search text is a hot topic.

[0042] For example, if the search text is "Chengdu weather is sunny", and the combined keyword set is "Chengdu, weather, sunny, good, bright sunshine", and the search text and keyword set are input into the BERT model, if the model outputs a result of 1, it means that "Chengdu weather is sunny" is a hot text, and a result of 0 means that it is not a hot text.

[0043] S109: When the search text is determined to be hot text, the remaining albums other than the hot albums are reordered so that the hot albums are recommended before the remaining albums.

[0044] In this embodiment of the application, if the search text is marked as hot text, it needs to be reordered. The candidate list marked as hot albums is extracted separately, and the remaining albums are arranged after the hot albums. The sorting order within the hot album list is still the initial sorting order.

[0045] S110: Send popular albums and remaining albums to users in the recommended order.

[0046] After obtaining the search text through steps S105 to S110, the server or terminal can search for multiple sorted albums corresponding to the search text. Then, based on the pre-stored set of keywords for hot content, the server or terminal will determine the hot albums that match the keywords of the hot content. When the number of these hot albums is greater than a preset threshold, the server or terminal will determine that the search text is a hot text and prioritize sending the hot albums corresponding to this hot text to the user. This achieves the goal of mining hot content in the content platform. Furthermore, by setting the recommendation order of hot albums before the recommendation order of the remaining albums, hot albums can be sent to users, solving the problem of low click-through rates and failure to obtain effective exposure within the content platform. This achieves the technical effect of effectively improving the recommendation rate of hot content.

[0047] In an optional implementation, in order to predetermine the keyword set corresponding to trending content and mine trending albums within the content platform, this embodiment of the invention provides an implementation method, please refer to [link to implementation details]. Figure 2 , Figure 2 Another illustrative flowchart of the hot topic content recommendation method provided in the embodiments of the present invention, namely, before step S105, may further include the following steps:

[0048] S101: Obtain a set of timely content texts.

[0049] In this embodiment of the invention, considering the timeliness of trending content, the trending content text refers to trending text within a target time period. The target time period can be within one week or two days from the current time. These trending content texts can be, but are not limited to, trending content from Weibo hot searches, Baidu hot search list, 360 News, and Sogou News. That is, the trending content texts are obtained by crawling from these portal websites using web crawling technology.

[0050] To ensure the timeliness of trending content, it is possible to periodically crawl from these portal websites, that is, to update the text set of timely content and keyword set according to the preset acquisition frequency.

[0051] S102: Perform keyword detection on each timely content text in sequence to obtain multiple sets of keywords corresponding to the hot content.

[0052] In this embodiment of the invention, in order to comprehensively obtain the keywords corresponding to the hot content and prevent omissions or incomplete results, this embodiment of the invention extracts the keywords in each time-sensitive content text through three methods: named entity extraction, event extraction, and dependency parsing. These will be described in detail later.

[0053] S103: Match the title information of each album within the target content platform with each set of trending keywords to determine whether the album is associated with trending content.

[0054] In this application embodiment, the method to determine whether an album is a hot album can be: inputting the album's title information and keyword set into a pre-trained classification model, such as the BERT model. If the model outputs 1, it indicates that the album is a hot album; if the model outputs 0, it indicates that the album is not a hot album.

[0055] For example, if the keyword corresponding to trending content is "Journey to the West," and the title of a certain album is "Journey to the West: Detailed Explanation Videos," then inputting "Journey to the West" and "Journey to the West: Detailed Explanation Videos" into a classification model will determine whether the album matches the trending content. If it does, the album is identified as a trending album associated with the trending content; otherwise, it is considered a non-trending album.

[0056] In an alternative implementation, the classification model can be pre-trained using corpus within the content platform. For example, popular albums and non-popular albums can be labeled first, and these albums can be used as training corpus to train the BERT model.

[0057] S104: When an album is associated with trending content, the album will be marked as a trending album, and the set of keywords corresponding to the album will be written into the album information.

[0058] In this embodiment of the application, when an album is determined to be a popular album, it can be marked. Then, when executing step S106, the popular album corresponding to the search text can be determined based on whether the mark exists. At the same time, in order to determine whether the search text is popular text in the future, the set of keywords matching the popular album can be associated. Specifically, the keywords can be stored in the album information of the popular album so that they can be matched with the search text when the user searches.

[0059] Through steps S101 to S104 above, the server or terminal first obtains a set of timely content texts, performs keyword detection on each timely content text, and obtains a keyword set. Then, the server or terminal combines the obtained keyword set and uses a classification model to determine whether each album in the content platform matches the keyword set, and whether the album is also hot content. If the judgment result is yes, the server or terminal will mark the album as a hot album and associate the album with the keyword set, thus achieving the effect of comprehensively and accurately mining hot content.

[0060] In the technical solution of step S103, considering that traditional webpage retrieval for keyword extraction is relatively easy to extract summary keywords due to the length of webpages, current trending content has diverse text formats and lacks a unified writing format. Furthermore, most of it is short text, making it impossible to extract keywords based solely on textual statistical information. It requires semantic analysis of the core content of the short text. For example, if the trending Weibo topic is "Chengdu's weather was sunny this morning, but it will rain this afternoon," extracting only "Chengdu will rain this morning" would be quite different from the original meaning. Moreover, after extracting the core content, it needs to be linked to resources within the website. Incorrect extraction could lead to incorrect association with trending topics, impacting user experience. Therefore, the core of this application's embodiment in extracting keywords from trending content is how to parse the core content from the diverse short texts of trending topics, comprehensively and accurately extracting keywords for the trending content.

[0061] Therefore, for step S103 above, the implementation of this embodiment of the invention can be as follows: input the timely content text into the classification model, extract the named entities in the timely content text as keywords; group all keywords according to the content type of the timely content text to obtain multiple sets of keywords.

[0062] In the above implementation, the classification model can be a pre-trained BERT model, which extracts named entities such as names of people, locations, times, and organizations from trending content through named entity recognition. Before using the classification model, corpora labeled with names, locations, times, and organizations can be obtained first, then the BERT model can be trained, the trained model can be used for extraction, and the results can be stored.

[0063] Considering that a single detection method may not be effective in extracting keywords from trending content, this application also provides another implementation method based on the above implementation method. The specific embodiment is as follows: extract events from time-sensitive content and use the extracted event components as keywords; wherein, the event components include event trigger words, event participants, and the role information of the event participants in the event.

[0064] Event extraction employs a sequence-to-structure architecture to uniformly model the entire event extraction process. All trigger words, arguments, and their role category labels are uniformly generated as natural language terms, extracting events from the text in an end-to-end manner. The input is text, and the output is the structured event components.

[0065] Event trigger: The core word indicating the occurrence of an event, usually a verb or noun. Event participants, also called event arguments, mainly consist of entities, values, and timestamps. The role of an event participant in the event, also called an argument role.

[0066] For example, a timely content text is: "I bought a computer at the mall today". By extracting the event, the result is: "{Event: Buy a computer, Event location: Mall, Event time: Today}". Here, "Buying a computer" is the event trigger word, the event location and event time are the roles of the event participants in the event, and their respective content is the event participants.

[0067] Furthermore, considering that some trending content may be a descriptive sentence, such as "How good-looking are school uniforms now?", steps 1 and 2 cannot extract the corresponding entity and event components. In this case, it is desirable to detect keywords such as "school uniforms are good-looking". Therefore, dependency parsing can be used to extract keywords for this type of trending content. Therefore, based on the above two implementation methods, this embodiment of the invention also provides an implementation method, which is as follows: perform dependency parsing on the timely content text to obtain the main content in the timely content text as keywords.

[0068] Dependency parsing refers to the asymmetrical relationship between two words in a sentence that have a syntactic relationship. One word is central, called the head, and the other is subordinate, called the dependent. The dependent depends on the head according to a certain relationship, usually serving to restrict and modify the head. Describing the head-dependent relationships between all words in a sentence forms the dependency structure representation of the sentence. If we consider the words in a sentence as nodes and the dependency relationships between words as directed edges, the dependency structure of the sentence can be described by a directed graph, called the sentence dependency graph.

[0069] Dependency parsing employs a graph decoding-based analysis method, specifically: finding a maximum dependency tree for the sentence to obtain the globally optimal solution for the sentence's dependency structure (i.e., transforming the construction process of the optimal dependency structure into the process of finding the maximum spanning tree). Let x represent a sentence x = w0w1w2...wn composed of n words, where w0 represents the manually added root word ROOT. If D(x) represents the set of all possible dependency trees for sentence x, then the search problem can be described as: z = argmaxScore(x,y), y in D(x), which means finding the dependency tree y with the highest score among all candidate dependency trees. The score of dependency tree y is calculated using the scoring function Score(x,y). By combining the extracted results, larger-granular phrases are obtained as the core content of the sentence, and the extracted results are stored. The main content used includes: S_V_O:(subject, predicate, object), for example: I bought a computer (I <- buy, buy->computer); ATT_N (modifier, core word), for example: I have a beautiful dress (beautiful->dress); DOB (subject-predicate, predicate, indirect object, direct object), for example: He gave me a computer (given->me, given->computer).

[0070] By using the three methods described above to detect keywords in timely content text, and then merging and deduplicating the detected keywords, a keyword set for trending content is obtained.

[0071] In an optional implementation, to improve association efficiency, keywords can be clustered according to the type of hot content to obtain multiple sets of keywords.

[0072] Optionally, when the number of popular albums is less than a preset threshold, or when the classification model determines that the search text is non-popular text, each album is sent to the user according to the sorting results of each album.

[0073] Compared with existing technologies, the hot content recommendation method provided in this application trackes timely information sources outside the site and mines hot topics within the site through text processing. When users search for timely content, hot topics are prioritized for distribution to users, which more accurately meets users' needs for hot content, improves user retention, and enhances user stickiness.

[0074] Based on the same inventive concept, embodiments of the present invention also provide a hot topic content recommendation device for performing the corresponding steps in the above method embodiments and various possible implementations. An implementation of the hot topic content recommendation device is given below. Please refer to... Figure 3 , Figure 3The diagram shows a block illustration of a hot topic content recommendation device provided in an embodiment of the present invention. The hot topic content recommendation device 300 includes: an acquisition module 310, a determination module 320, a merging module 330, a classification module 340, a sorting module 350, and a recommendation module 360;

[0075] The acquisition module 310 is used to acquire the search text and the sorted albums corresponding to the search text;

[0076] The determination module 320 is used to identify the most popular albums on the content platform from multiple albums; wherein, the most popular albums are matched with keywords of pre-acquired popular content;

[0077] The merging module 330 is used to merge the keywords corresponding to the hot albums when the number of hot albums is greater than or equal to a preset threshold.

[0078] The classification module 340 is used to input the search text and the merged keyword set into the pre-trained classification model to determine whether the search text is a hot text.

[0079] The sorting module 350 is used to reorder the remaining albums other than the hot albums when the search text is determined to be hot text, so that the recommendation order of the hot albums is placed before the recommendation order of the remaining albums.

[0080] The recommendation module 360 ​​is used to send popular albums and remaining albums to users in the order of recommendation.

[0081] In an optional implementation, the determining module 320 is further configured to obtain a set of timely content texts; perform keyword detection on each timely content text in sequence to obtain multiple sets of keywords corresponding to the hot content; match the title information of each album in the content platform with each set of hot keywords to determine whether the album is associated with the hot content; when the album is associated with the hot content, the album is marked as a hot album and the keyword set corresponding to the album is written into the album information of the album.

[0082] In an optional implementation, the determining module 320 is specifically used to input the timely content text into the classification model, extract the named entities in the timely content text as keywords, and group all keywords according to the content type of the timely content text to obtain multiple sets of keywords.

[0083] In an optional implementation, the determining module 320 is specifically used to extract events from time-sensitive content and use the extracted event components as keywords; wherein, the event components include event trigger words, event participants, and the role information of the event participants in the event.

[0084] In an optional implementation, the determining module 320 is specifically used to perform dependency sentence analysis on the time-sensitive content text to obtain the main content in the time-sensitive content text as keywords.

[0085] In an optional implementation, the recommendation module 360 ​​is further configured to send each album to the user according to the sorting results of each album when the number of popular albums is less than a preset threshold, or when the classification model determines that the search text is non-popular text.

[0086] In an optional implementation, the hot topic content recommendation device 300 further includes an update module for updating the time-sensitive content text set and keyword set according to a preset acquisition frequency.

[0087] It should be noted that the module division in the above embodiments of the present invention is illustrative and only represents one logical functional division. In actual implementation, there may be other division methods. Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, exist as separate physical units, or have two or more units integrated into one unit. The integrated units described above can be implemented in hardware or as software functional units.

[0088] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a readable storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned readable storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0089] This invention also provides a computer device for executing the hot topic content recommendation method described in this invention. Please refer to... Figure 4 , Figure 4 A block diagram of a computer device provided in an embodiment of the present invention is shown. The computer device 400 may be a server, a personal computer, an edge gateway, etc., and includes a processor 402, a memory 401, a bus 404, and a communication interface 403. The processor 402 is connected to the memory 401 through the bus 404.

[0090] Optionally, bus 404 can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, Figure 4 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0091] In this embodiment, the processor 402 may be a general-purpose processor, digital signal processor, application-specific integrated circuit, field-programmable gate array, or other programmable logic device, discrete gate or transistor logic device, or discrete hardware component, capable of implementing or executing the methods, steps, and logic block diagrams disclosed in this embodiment. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the methods disclosed in this embodiment can be directly implemented by the hardware processor, or implemented by a combination of hardware and software modules within the processor. The software modules may reside in the memory 401, and the processor 402 reads the program instructions from the memory 401 and, in conjunction with its hardware, completes the steps of the aforementioned methods.

[0092] In this embodiment, memory 401 can be non-volatile memory, such as a hard disk drive (HDD) or a solid-state drive (SSD), or it can be volatile memory, such as RAM. Memory can also be any other medium capable of carrying or storing desired program code in the form of instructions or data structures, and accessible by a computer, but is not limited thereto. The memory in this embodiment can also be a circuit or any other device capable of implementing storage functions, used to store instructions and / or data.

[0093] The memory 401 can be used to store software programs and modules, such as the instructions / modules of the hot content recommendation device 300 provided in this embodiment of the invention. These can be stored in the memory 401 in the form of software or firmware, or embedded in the operating system (OS) of the computer device 400. The processor 402 executes various functional applications and data processing by executing the software programs and modules stored in the memory 401. The communication interface 403 can be used for signaling or data communication with other node devices.

[0094] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described apparatus and unit can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0095] Understandable. Figure 4 The structure shown is for illustrative purposes only; the computer device 400 may also include more than [other components]. Figure 4 The more or fewer components shown, or having the same Figure 4 The different configurations shown. Figure 4 The components shown can be implemented using hardware, software, or a combination thereof.

[0096] Based on the above embodiments, the present invention also provides a readable storage medium storing a computer program, which, when executed by a computer, causes the computer to perform the hot content recommendation method provided in the above embodiments.

[0097] Based on the above embodiments, this invention also provides a computer program that, when run on a computer, causes the computer to execute the hot topic content recommendation method provided in the above embodiments.

[0098] Based on the above embodiments, this invention also provides a chip for reading computer programs stored in a memory and executing the hot topic content recommendation method provided in the above embodiments.

[0099] This invention also provides a computer program product, including instructions that, when run on a computer, cause the computer to execute the hot topic content recommendation method provided in the above embodiments.

[0100] Embodiments of the present invention are described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by instructions. These instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0101] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0102] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0103] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method of recommending hotspot content, characterized by, The method comprises: obtaining search text and a plurality of albums corresponding to the search text; determining a hot album in a content platform from the plurality of albums, wherein the hot album matches a keyword of a pre-obtained hot content; when a quantity of the hot albums is greater than or equal to a preset threshold, merging the keywords corresponding to the hot albums; inputting the search text and the merged keyword set into a pre-trained classification model to determine whether the search text is a hot text; when it is determined that the search text is a hot text, reordering the hot albums and remaining albums other than the hot albums so that a recommendation order of the hot albums is in front of a recommendation order of the remaining albums; sending the hot albums and the remaining albums to a user in the recommendation order; before obtaining the search text and the plurality of albums corresponding to the search text, the method further comprises: obtaining a set of time-sensitive content texts; performing keyword detection on each time-sensitive content text to obtain a plurality of keyword sets corresponding to hot content; matching title information of each album in the content platform with each keyword set to determine whether the album is associated with the hot content; when the album is associated with the hot content, marking the album as the hot album, and writing a keyword set corresponding to the album into album information of the album.

2. The method of claim 1, wherein, The keyword detection on each time-sensitive content text to obtain a plurality of keyword sets corresponding to hot content comprises: inputting the time-sensitive content text into the classification model to extract a named entity in the time-sensitive content text as the keyword; grouping all the keywords according to a content type of the time-sensitive content text to obtain a plurality of keyword sets.

3. The method of claim 2, wherein, The keyword detection on each time-sensitive content text to obtain a plurality of keyword sets corresponding to hot content further comprises: performing event extraction on the time-sensitive content to extract an event component as the keyword, wherein the event component comprises an event trigger word, an event participant, and role information of the event participant in the event.

4. The method of claim 3, wherein, The keyword detection on each time-sensitive content text to obtain a plurality of keyword sets corresponding to hot content further comprises: performing dependency sentence analysis on the time-sensitive content text to obtain a main content in the time-sensitive content text as the keyword.

5. The method of claim 1, wherein, The method further comprises: when the quantity of the hot albums is less than the preset threshold, or when the classification model determines that the search text is a non-hot text, sending each album to a user according to a sorting result of each album.

6. The method of claim 1, wherein, The method further comprises: updating the set of time-sensitive content texts and the keyword sets according to a preset acquisition frequency.

7. A hot spot content recommendation apparatus characterized by comprising: The method comprises an obtaining module, a determining module, a merging module, a classification module, a sorting module, and a recommendation module; the obtaining module is configured to obtain search text and a plurality of albums corresponding to the search text; the obtaining module is configured to obtain search text and a plurality of albums corresponding to the search text; The determining module is configured to determine a hot album in the content platform from the multiple albums, wherein the hot album matches a keyword of the pre-acquired hot content; The merging module is configured to merge the keywords corresponding to the hot album when a quantity of the hot albums is greater than or equal to a preset threshold value; The classification module is configured to input the search text and the merged keyword set into a pre-trained classification model to determine whether the search text is a hot text; The sorting module is configured to reorder the hot album and remaining albums other than the hot album when the search text is determined to be a hot text, so that a recommendation order of the hot album is located in front of a recommendation order of the remaining albums; The recommendation module is configured to send the hot album and the remaining albums to a user according to the recommendation order; The determining module is further configured to acquire a set of time-sensitive content texts, perform keyword detection on each time-sensitive content text in sequence to obtain multiple sets of keywords corresponding to hot content, match title information of each album in the content platform with each set of keywords to determine whether the album is associated with the hot content, and mark the album as the hot album when the album is associated with the hot content, and write a keyword set corresponding to the album into album information of the album.

8. A computer device, comprising: A processor and a memory are included, the memory stores a computer program capable of being executed by the processor, and the processor can execute the computer program to implement the hot content recommendation method of any one of claims 1 to 6.

9. A readable storage medium, having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the hot content recommendation method of any one of claims 1 to 6.

Images