Resource recommendation method and device, electronic equipment and computer program product

Abstract

The embodiment of the invention provides a resource recommendation method and device, electronic equipment and a computer program product, and relates to the technical field of computers. The method comprises the steps that a target hot search word is acquired, an intention song matched with the target hot search word is determined based on song playing index data, and the target hot search word is an initial search word meeting a hot search word screening condition; when the playable intention song does not exist in the music platform, determining associated service resources related to the intention song from the published service resources; and based on the intention song identifier of the intention song, determining a target matching resource corresponding to the target hot search word from the associated business resources as a target search result of the target hot search word. According to the invention, a full-link automatic implementation scheme from hot search word statistics to resource matching is provided, a lot of time and labor cost are saved, and the timeliness of hot search resource undertaking is improved.

Description

Background Art

[0002] This section aims to provide background or context for the embodiments of the present disclosure described in the claims. The description herein is not admitted to be prior art merely by including it in this section.

[0003] Internally, a music platform usually includes multiple sub - services, such as singles, podcasts, singers, dynamics, etc. Under this premise, there may be a scenario where, after a user searches for a corresponding search term (query term) on the platform, the platform will launch the best - matching resources based on the input query term and the calculated intended song. However, when there are copyright restrictions or temporary copyright locks on a song, it is necessary to find the most suitable resources from other related sub - services and distribute them.

[0004] For sub - service providers, how to quickly know which hot query terms are in the platform, which query - term - corresponding singles currently have no copyright issues, and whether the sub - service currently has resources or which resources can undertake relevant searches is a problem that the hot - search - term resource - undertaking scheme needs to solve. Summary of the Invention

[0005] When undertaking hot - search terms, only when the sub - service provider knows the above - mentioned information in real time can the sub - service provider further timely find the high - quality resources that match them and undertake the resources in a timely manner. However, due to factors such as a large number of hot - search terms, the related hot - search - term resource - undertaking scheme has problems of high implementation cost and poor timeliness.

[0006] Therefore, the present disclosure proposes an improved resource recommendation method to enable the hot - search - term resource - undertaking scheme to save a large amount of human and time costs and improve the timeliness of resource undertaking.

[0007] In this context, embodiments of the present disclosure are expected to provide a resource recommendation method, a resource recommendation device, a computer - readable storage medium, an electronic device, and a computer program product.

[0008] In the first aspect of the embodiments of the present disclosure, a resource recommendation method is provided, including: obtaining a target hot - search term, determining an intended song that matches the target hot - search term based on song play metric data, where the target hot - search term is an initial search term that meets the hot - search - term screening condition; when there is no playable intended song on the music platform, determining associated service resources related to the intended song from the published service resources; and determining a target matching resource corresponding to the target hot - search term from the associated service resources based on the intended - song identifier of the intended song as the target search result of the target hot - search term.

[0009] In one embodiment of the present disclosure, obtaining a target hot search term and determining an intent song that matches the target hot search term based on song play metric data includes: obtaining an initial search term and a pre-configured hot search term screening condition; determining candidate hot search terms from the initial search term based on a preset time interval and the hot search term screening condition, where the candidate hot search terms include the target hot search term; determining song play metric data corresponding to candidate songs in the song library, where the song play metric data includes one or more of song click-through rate, click event data, and play event data; and determining an intent song and an intent song identifier corresponding to the intent song from the candidate songs based on the song play metric data and the target hot search term.

[0010] In one embodiment of the present disclosure, determining candidate hot search terms from the initial search term based on a preset time interval and the hot search term screening condition includes: obtaining a search term data table corresponding to the initial search term; obtaining the hot search term screening condition, where the hot search term screening condition includes one or more of search access volume being greater than a search volume threshold and the month-on-month growth rate of search access volume being greater than a growth rate threshold; and performing search term screening processing on the search term data table based on the preset time interval and the hot search term screening condition to obtain candidate hot search terms.

[0011] In one embodiment of the present disclosure, the above method further includes: determining basic song information of the intent song, where the basic song information includes one or more of song copyright information, song content information, and song format quality information; and determining a sound source play state corresponding to the intent song based on the basic song information, where the sound source play state includes a playable state and an unplayable state.

[0012] In one embodiment of the present disclosure, determining associated business resources related to the intent song from the published business resources includes: determining initial business resources, where the initial business resources are business resources related to the music category; performing audio-video recognition processing on the initial business resources to determine associated song information related to the initial business resources, where the associated song information includes an associated song identifier and an associated song metric; establishing a business resource index corresponding to the initial business resources; storing the initial business resources and the associated song information in a resource database based on the business resource index to generate published business resources; and determining associated business resources based on the published business resources.

[0013] In one embodiment of the present disclosure, determining associated business resources based on the published business resources includes: performing an association recognition process on the published business resources based on the intent song to obtain initial associated resources, where the audio similarity between the initial associated resources and the intent song is greater than a similarity threshold, and / or the song duration ratio of the initial associated resources is greater than a duration ratio threshold; determining associated business resources from the initial associated resources based on a pre-configured resource screening condition; and determining associated resource information corresponding to the associated business resources.

[0014] In one embodiment of the present disclosure, determining the associated resource information corresponding to the associated service resource includes: determining the resource matching metrics corresponding to the associated service resource, where the resource matching metrics include one or more of a resource quality score, a resource popularity score, the songs included in the resource, the number of included songs, the audio similarity of the resource, the audio matching duration of the resource, the audio proportion duration of the resource, and the resource content type; generating the associated resource information based on the resource matching metrics, and storing the associated resource information in the resource database.

[0015] In one embodiment of the present disclosure, based on the intent song identifier of the intent song, determining the target matching resource corresponding to the target hot search term from the associated service resources includes: performing an idempotent filtering process on the associated service resources based on the intent song identifier to obtain the initial matching resources corresponding to the target hot search term; obtaining the pre-configured resource matching rules, and determining the candidate matching resources from the initial matching resources according to the resource matching rules and the intent song identifier; determining the resource search level corresponding to the candidate matching resources, and performing an associated tagging process on the candidate matching resources, the target hot search term, and the resource search level to obtain the search level weight corresponding to the candidate matching resources; and determining the target matching resource from the candidate matching resources according to the search level weight, and displaying the target matching resource.

[0016] In one embodiment of the present disclosure, determining the target matching resource from the candidate matching resources according to the search level weight includes: determining the associated matching metrics corresponding to the candidate matching resources, where the associated matching metrics include one or more of the similarity between the resource name and the target hot search term, the resource play volume, and the resource timeliness; determining the associated matching weight corresponding to the candidate matching resources according to the associated matching metrics; and determining the target matching resource from the candidate matching resources according to the search level weight and the associated matching weight.

[0017] In the second aspect of the embodiments of the present disclosure, a resource recommendation device is provided, including: an intent song determination module, configured to obtain a target hot search term, and determine an intent song that matches the target hot search term based on the song play metric data, where the target hot search term is an initial search term that meets the hot search term screening conditions; an associated resource determination module, configured to determine associated service resources related to the intent song from the published service resources when there is no playable intent song on the music platform; and a search result determination module, configured to determine a target matching resource corresponding to the target hot search term from the associated service resources based on the intent song identifier of the intent song, as the target search result of the target hot search term.

[0018] In an embodiment of the present disclosure, the intended song determination module includes an intended song determination unit, configured to: obtain an initial search term and a pre-configured hot search term screening condition; determine candidate hot search terms from the initial search term based on a preset time interval and the hot search term screening condition, where the candidate hot search terms include target hot search terms; determine the song play metric data corresponding to the candidate songs in the song library, where the song play metric data includes one or more of song click-through rate, click event data, and play event data; and determine the intended song and the corresponding intended song identifier from the candidate songs based on the song play metric data and the target hot search term.

[0019] In an embodiment of the present disclosure, the intended song determination unit includes a hot search term determination subunit, configured to: obtain the search term data table corresponding to the initial search term; obtain the hot search term screening condition, where the hot search term screening condition includes one or more of the search access volume being greater than the search volume threshold, and the month-on-month growth rate of the search access volume being greater than the growth rate threshold; and perform search term screening processing on the search term data table based on the preset time interval and the hot search term screening condition to obtain candidate hot search terms.

[0020] In an embodiment of the present disclosure, the resource recommendation device further includes a play status determination module, configured to: determine the basic song information of the intended song, where the basic song information includes one or more of song copyright information, song content information, and song format quality information; and determine the sound source play status corresponding to the intended song according to the basic song information, where the sound source play status includes a playable status and an unplayable status.

[0021] In an embodiment of the present disclosure, the associated resource determination module includes an associated resource determination unit, configured to: determine an initial business resource, where the initial business resource is a business resource related to the music category; perform audio-visual recognition processing on the initial business resource to determine the associated song information related to the initial business resource, where the associated song information includes an associated song identifier and an associated song metric; establish a business resource index corresponding to the initial business resource; store the initial business resource and the associated song information in the resource database based on the business resource index to generate a published business resource; and determine the associated business resource according to the published business resource.

[0022] In an embodiment of the present disclosure, the associated resource determination unit includes an associated resource determination subunit, configured to: perform an association recognition process on the published business resource based on the intended song to obtain an initial associated resource, where the audio similarity between the initial associated resource and the intended song is greater than the similarity threshold, and / or the song duration ratio of the initial associated resource is greater than the duration ratio threshold; determine the associated business resource from the initial associated resource based on the pre-configured resource screening condition; and determine the associated resource information corresponding to the associated business resource.

[0023] In one embodiment of the present disclosure, the associated resource determination unit includes a resource information determination subunit, which is configured to: determine a resource matching metric corresponding to the associated service resource, where the resource matching metric includes one or more of a resource quality score, a resource popularity score, the songs included in the resource, the number of included songs, the audio similarity of the resource, the audio matching duration of the resource, the audio occupancy duration of the resource, and the resource content type; generate associated resource information based on the resource matching metric, and store the associated resource information in the resource database.

[0024] In one embodiment of the present disclosure, the search result determination module includes a target resource determination unit, which is configured to: perform an idempotent filtering process on the associated service resource based on the intended song identifier to obtain an initial matching resource corresponding to the target hot search term; obtain a pre-configured resource matching rule, and determine a candidate matching resource from the initial matching resources according to the resource matching rule and the intended song identifier; determine the resource search level corresponding to the candidate matching resource, and perform an associated tagging process on the candidate matching resource, the target hot search term, and the resource search level to obtain the search level weight corresponding to the candidate matching resource; and determine a target matching resource from the candidate matching resources according to the search level weight, and display the target matching resource.

[0025] In one embodiment of the present disclosure, the target resource determination unit includes a target resource determination subunit, which is configured to: determine an associated matching metric corresponding to the candidate matching resource, where the associated matching metric includes one or more of the similarity between the resource name and the target hot search term, the resource play volume, and the resource timeliness; determine the associated matching weight corresponding to the candidate matching resource according to the associated matching metric; and determine a target matching resource from the candidate matching resources according to the search level weight and the associated matching weight.

[0026] In the third aspect of the embodiments of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the resource recommendation method as described above is implemented.

[0027] In the fourth aspect of the embodiments of the present disclosure, an electronic device is provided, including: a processor; and a memory, on which computer-readable instructions are stored, and when the computer-readable instructions are executed by the processor, the resource recommendation method as described above is implemented.

[0028] According to the fifth aspect of the embodiments of the present disclosure, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the resource recommendation method as described in any one of the above is implemented.

[0029] According to the technical solution of the embodiments of the present disclosure, on the one hand, a full-link automated implementation solution from hot search term statistics to resource matching is provided, which can save a large amount of time and labor costs. On the other hand, by combining hot search term statistics with a matching mechanism, the target matching resources corresponding to the target hot search terms can be quickly matched, improving the timeliness of hot search term resource reception. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present disclosure will become readily understood. In the drawings, several embodiments of the present disclosure are shown by way of illustration and not limitation, wherein:

[0031] Figure 1 A schematic block diagram of a system architecture of an exemplary application scenario according to some embodiments of the present disclosure is schematically shown;

[0032] Figure 2 A flowchart of a resource recommendation method according to some embodiments of the present disclosure is schematically shown;

[0033] Figure 3 A general flowchart of realizing hot search term resource mining according to some embodiments of the present disclosure is schematically shown;

[0034] Figure 4 A flowchart of automatic matching based on target hot search terms according to some embodiments of the present disclosure is schematically shown;

[0035] Figure 5 A flowchart of resource review and automatic marking according to some embodiments of the present disclosure is schematically shown;

[0036] Figure 6 A schematic block diagram of a resource recommendation device according to some embodiments of the present disclosure is schematically shown;

[0037] Figure 7 A schematic diagram of a storage medium according to an example embodiment of the present disclosure is schematically shown;

[0038] Figure 8 A block diagram of an electronic device according to an example embodiment of the invention is schematically shown.

[0039] In the drawings, the same or corresponding reference numerals indicate the same or corresponding parts. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0040] The principles and spirit of the present disclosure will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided only to enable those skilled in the art to better understand and implement the present disclosure, and are not intended to limit the scope of the present disclosure in any way. On the contrary, these embodiments are provided to make the present disclosure more thorough and complete, and to fully convey the scope of the present disclosure to those skilled in the art.

[0041] Those skilled in the art will appreciate that the embodiments of the present disclosure may be implemented as a system, device, apparatus, method or computer program product. Therefore, the present disclosure may be specifically implemented in the following forms, namely: complete hardware, complete software (including firmware, resident software, microcode, etc.), or a combination of hardware and software.

[0042] According to an embodiment of the present disclosure, a resource recommendation method, a resource recommendation device, a computer-readable storage medium, an electronic device, and a computer program product are proposed.

[0043] In this article, it is important to understand that the terms involved, such as search terms (query terms), can be keywords that platform users enter and search for in the platform. Intended songs can be the intended songs with the highest matching degree associated with different query terms calculated by the system through big data model analysis. Song copyright can be whether a song has the copyright to be played in the platform at a certain moment.

[0044] In addition, any number of elements in the drawings is for illustration and not limitation, and any naming is only for distinction and does not have any limiting meaning.

[0045] The principle and spirit of the present disclosure are explained in detail below with reference to several representative embodiments of the present disclosure. Summary of the Invention

[0047] In a hot search word resource undertaking solution, the real-time output of query word page views (PageView, PV) reports is combined with hot search word message reminders, and the operation is carried out by operators who are specifically responsible for hot search operations. The specific process can be described as: the operation of the sub-business or receiving hot search word message reminders, or regularly traversing the hot search word report; searching the song carrying status of the main site one by one in the site; for those that cannot be carried by songs, searching for related resources in their own business resources; through labeling, establishing the relationship between search terms and resources and search levels to improve the exposure of resources.

[0048] The advantages of the above solution are that it is simple and easy to implement and has a relatively high accuracy. However, there are many deficiencies: 1) High cost. The number of query words with high PV per day is very large, and the operation needs to invest a lot of manpower and energy to continuously conduct investigations and maintenance updates. 2) Poor timeliness. Since the timeliness of most hot topics is usually short, coupled with the problem of high cost, the operation usually cannot maintain and update in real time, resulting in missing a lot of real-time hot topic traffic.

[0049] Based on the above content, the basic idea of the present disclosure is to obtain a target hot search term, determine an intended song that matches the target hot search term based on song play index data, where the target hot search term is an initial search term that meets the hot search term screening conditions; when there is no playable intended song on the music platform, determine an associated business resource related to the intended song from the published business resources; based on the intended song identifier of the intended song, determine a target matching resource corresponding to the target hot search term from the associated business resources as the target search result of the target hot search term. On the one hand, a full-link automated implementation solution from hot search term statistics to resource matching is provided, which can save a large amount of time cost and labor cost. On the other hand, by combining hot search term statistics with a matching mechanism, the target matching resource corresponding to the target hot search term can be quickly matched, improving the timeliness of hot search term resource reception.

[0050] After introducing the basic principle of the present disclosure, the following specifically introduces various non-limiting implementation manners of the present disclosure.

[0051] Overview of Application Scenarios

[0052] First, refer to Figure 1 , Figure 1 which shows a schematic block diagram of the system architecture of an exemplary application scenario of a resource recommendation method and apparatus to which the embodiments of the present disclosure can be applied.

[0053] As Figure 1 shown, the system architecture 100 may include one or more of the terminal devices 101, 102, 103, a network 104, and a server 105. The network 104 is used to provide a medium for a communication link between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, etc. The terminal devices 101, 102, 103 may be various electronic devices with a display screen, including but not limited to desktop computers, portable computers, smart phones, and tablet computers, etc. It should be understood that Figure 1 the numbers of terminal devices, networks, and servers in

[0054] The resource recommendation method provided by the embodiments of the present disclosure is generally executed by the server 105. Correspondingly, the resource recommendation device is generally set in the server 105. However, those skilled in the art can easily understand that the resource recommendation method provided by the embodiments of the present disclosure can also be executed by the terminal devices 101, 102, and 103. Correspondingly, the resource recommendation device can also be set in the terminal devices 101, 102, and 103. No special limitation is made in this exemplary embodiment. For example, in an exemplary embodiment, it can be that a staff member uploads a pre-constructed knowledge graph and a user question input by the user to the server 105 through the terminal devices 101, 102, and 103. The server performs resource matching processing according to the current target hot search term by using the resource recommendation method provided by the embodiments of the present disclosure, determines the target matching resource, and transmits the determined target matching resource to the terminal devices 101, 102, and 103, etc., so that the terminal devices 101, 102, and 103 display the target matching resource corresponding to the target hot search term to the user.

[0055] It should be understood that Figure 1 The application scenario shown is only an example in which the embodiments of the present disclosure can be implemented. The scope of application of the embodiments of the present disclosure is not limited by any aspect of this application scenario.

[0056] Exemplary Method

[0057] Next, in combination with Figure 1 the application scenario of Figure 2 reference is made to

[0058] to describe the resource recommendation method according to the exemplary embodiments of the present disclosure. It should be noted that the above application scenario is only shown for the convenience of understanding the spirit and principle of the present disclosure, and the embodiments of the present disclosure are not limited in this regard. On the contrary, the embodiments of the present disclosure can be applied to any applicable scenario.

[0058] The present disclosure first provides a resource recommendation method. The execution subject of this method can be a terminal device or a server. The present disclosure does not make special limitations in this regard. In this exemplary embodiment, it is described by taking the server as an example for executing this method.

[0059] Referring to Figure 2 shown, the resource recommendation method may include the following steps S210 to step S230:

[0060] Step S210, obtain a target hot search term, and determine an intention song that matches the target hot search term based on song play metric data. The target hot search term is an initial search term that meets the hot search term screening conditions;

[0061] Step S220, when there is no playable intention song on the music platform, determine an associated business resource related to the intention song from the published business resources;

[0062] In step S230, based on the intent song identifier of the intent song, determine the target matching resource corresponding to the target hot search term from the associated service resources as the target search result of the target hot search term.

[0063] In the resource recommendation method provided by this exemplary embodiment, on the one hand, a full-link automated implementation solution from hot search term statistics to resource matching is provided, which can save a large amount of time cost and labor cost. On the other hand, by combining hot search term statistics with the matching mechanism, the target matching resource corresponding to the target hot search term can be quickly matched, improving the timeliness of hot search term resource reception. Next, the above steps of this exemplary embodiment will be described in more detail.

[0064] In step S210, obtain the target hot search term, and determine the intent song that matches the target hot search term based on the song play metric data. The target hot search term is the initial search term that meets the hot search term screening criteria.

[0065] In some exemplary embodiments, the target hot search term can be the hot search term currently input by the user. The target hot search term can be a hot search term, a soaring term, or other hot terms. A hot search term can be a query term whose cumulative search pv reaches a specified threshold or more during a certain period of the day. A soaring term can be a query term whose cumulative search pv reaches a specified threshold or more during a certain period of the day compared with the same period of the previous day. The song play metric data can be the relevant metric data of the operations such as the song being searched or played by the user on the platform. The hot search term screening criteria can be the screening criteria for determining whether a certain search term belongs to a hot search term.

[0066] Obtain the target hot search term. A hot search term is usually a search term that meets the hot search term screening criteria. For example, a hot search term can be a search term that meets the hot search or soaring criteria. The target hot search term can be the hot search term currently input by the user through the platform interface of the music platform, or the hot search term recommended by the music platform for the user. When the user uses the music platform, they can input a hot search term with popularity as the target hot search term by themselves, or directly use the hot search term defaultly displayed by the music platform as the target hot search term to perform resource search.

[0067] After obtaining the target hot search term, the music platform can perform hot search term recognition and song recognition processing, and calculate the song that best matches the target hot search term as the intent song through an algorithm model. The intent song has a unique corresponding intent song identifier.

[0068] In step S220, when there is no playable intent song on the music platform, determine the associated service resources related to the intent song from the published service resources.

[0069] In some example embodiments, the published business resources may be business resources for resource matching published by each business party connected to the music platform in their respective business domains. For example, they may be business resources published by a sub-business party connected to the music platform. The associated business resources may be published business resources related to the intended song. For example, the associated business resources may be works containing partial audio segments of the intended song, works of different renditions of the intended song, works with the same song name as the intended song, etc.

[0070] After determining the intended song, determine the sound source playback status of the intended song. Due to reasons such as song copyright, content review, technical issues, or platform promotion strategies, the sound source playback status of the intended song may be an unplayable state. If there is an intended song in the music platform that matches the target hot search term and the intended song is in a playable state on the music platform, directly use the intended song as the resource search result for the target hot search term.

[0071] When there is no intended song in the music platform that is in a playable state, search the business resources published by the business parties connected to the music platform. The published business resources mainly refer to resources related to the music category. Identify resources through the published business resources and use audio-video algorithms for identification to determine the associated business resources related to the intended song.

[0072] In step S230, based on the intended song identifier of the intended song, determine the target matching resource corresponding to the target hot search term from the associated business resources as the target search result for the target hot search term.

[0073] In some example embodiments, the intended song identifier may be a unique identifier used to distinguish the intended song from other songs. The target matching resource may be an associated business resource with a relatively high degree of matching with the target hot search term. The number of target matching resources may be one or more. When the number of target matching resources is multiple, display the target matching resources in a list form and arrange them in descending order of matching degree in the list.

[0074] After determining the associated business resources related to the intended song from the sub-business party, perform resource matching processing on the associated business resources. For example, combine the corresponding target hot search term and the calculated intended song to perform resource matching within the integrated resource pool that meets the conditions to find the best-matched audio or video resource. Determine the target matching resources with a matching degree of topN with the target hot search term from the associated business resources as the target search result for the target hot search term and display the target search result.

[0075] In one embodiment of the present disclosure, for step S210, obtaining a target hot search term and determining an intent song that matches the target hot search term based on song play metric data includes: obtaining an initial search term and a pre-configured hot search term screening condition; determining candidate hot search terms from the initial search terms based on a preset time interval and the hot search term screening condition, where the candidate hot search terms include the target hot search term; determining the song play metric data corresponding to the candidate songs in the song library, where the song play metric data includes one or more of song click-through rate, click event data, and play event data; and determining the intent song and the corresponding intent song identifier from the candidate songs based on the song play metric data and the target hot search term.

[0076] Among them, the initial search term can be a term used for resource search in a music platform. The preset time interval can be a pre-configured time interval for periodically executing an offline hot search term calculation task. The candidate hot search terms can be the initial search terms that meet the hot search term screening condition. The song library can be a resource library composed of multiple candidate songs. The candidate songs can be songs matched through a resource search operation; in the resource search operation, different initial search terms have their respective matched songs.

[0077] Reference Figure 3 , Figure 3 schematically shows an overall flowchart for realizing hot search term resource mining according to some embodiments of the present disclosure. For the initial search terms existing in a music platform, candidate hot search terms that meet the hot search term screening condition can be determined through an offline calculation timing task, such as taking search terms that reach the hot search or soaring standard as candidate hot search terms. For example, through an offline calculation task executed every 0.5 hours or 1 hour, the initial search terms that meet the requirements are calculated as candidate hot search terms, and at the same time, hot search term recognition and song recognition are performed based on an algorithm model, and the intent song identifier that best matches each candidate hot search term is calculated.

[0078] The algorithm model can perform hot search term recognition on the initial search terms, perform song recognition on the candidate songs, combine the candidate hot search terms with the song play metric data of the candidate songs, screen out the intent songs that match the intent songs from the candidate songs, and record the intent song ID. The song play metric data can include, but is not limited to, song click-through rate, click event data, and play event data, etc.

[0079] The song click-through rate can be the statistics of the number of times a user plays a song by clicking on the play button, cover, song name, etc. of the song on the music platform. The song click-through rate is an important indicator to measure the popularity and attention of a song on a specific platform.

[0080] Click tracking data refers to some code snippets (i.e. tracking points) inserted at specific locations in the code of the music platform, which are used to capture the user's click operation behavior related to the song, and send these behavior data to the data server for recording and storage, thus forming a series of data. Click tracking data can be used to accurately collect and analyze a series of user behaviors such as clicking, collecting, and sharing songs.

[0081] Playback tracking data can be detailed data collected for in-depth analysis of user song playback behavior, which is obtained by setting tracking points in playback-related links. Playback tracking data can include playback behavior data and playback process data. Playback behavior data can include playback click time, playback mode, and playback source, etc.; playback process data can include playback duration, pause times and time, and drag progress, etc.

[0082] After determining the candidate hot search terms and candidate songs, the audio and video algorithm will associate these candidate hot search terms (also called intent terms) with the candidate songs' click-through rates, click tracking points, play tracking points and other related data to obtain the ID of the intended song that matches them. Figure 3 For the matched songs, the identified song information is further cleaned and integrated, such as filtering out candidate songs that are in a playable state and retaining those that are not playable, and filtering out other songs that do not meet the conditions, such as candidate songs in the black library. At this point, the initial cleaning of the data source is completed, and the task can output incremental data to the matching service in the form of asynchronous messages at a specified time frequency. By determining the ID of the intended song that matches the target hot search term, it can be determined whether to perform a business resource matching operation based on the playback status of the intended song.

[0083] In one embodiment of the present disclosure, candidate hot search terms are determined from initial search terms based on a preset time interval and hot search term filtering conditions, including: obtaining a search term data table corresponding to the initial search term; obtaining hot search term filtering conditions, the hot search term filtering conditions including one or more of a search volume greater than a search volume threshold, and a month-on-month growth rate of the search volume greater than a growth rate threshold; based on the preset time interval and the hot search term filtering conditions, performing search term filtering processing on the search term data table to obtain candidate hot search terms.

[0084] The search term data table may be an offline data table corresponding to the search term, and the search term data table may include relevant data for each initial search term. The search traffic may be the sum of page views generated after a user searches for a specific search term using the music platform. The search volume threshold may be a reference threshold for comparison with the value of the search traffic.

[0085] For the initial search terms, offline tasks are required to clean and calculate the offline data table of the initial search terms. Continuing to refer to Figure 3 , Figure 3 In the offline calculation timing task in, the search term data table can be searched at preset time intervals, and all query term data that meets the business expectations can be cleaned out as candidate search terms. Referring to Table 1, Table 1 schematically shows the specific content of the fields included in the search term data table, specifically including field descriptions, field keywords, and specific types, etc.

[0086] Table 1 Search Term Data Table

[0087]

[0088]

[0089] For example, obtain the search access volume corresponding to each initial search term and the month-on-month growth rate of the search access volume. When the search pv of the initial search term is greater than the search volume threshold, the initial search term is used as a candidate hot search term; and for the soaring terms whose search pv month-on-month increase of the initial search term is greater than the growth rate threshold, they are used as candidate hot search terms. Determining candidate hot search terms through the hot search term screening conditions has high flexibility, and both the range of hot search terms to be processed and the degree of resource matching can be flexibly adjusted through parameter configuration.

[0090] In an embodiment of the present disclosure, the basic song information of the intended song is determined, and the basic song information includes one or more of song copyright information, song content information, and song format quality information; the sound source playback state corresponding to the intended song is determined according to the basic song information, and the sound source playback state includes a playable state and an unplayable state.

[0091] Among them, the basic song information can be the basic information corresponding to the intended song. The song copyright information can be the information on whether there is copyright for the intended song on the music platform. The song content information can be the information related to the content of the intended song itself. For example, the song content information can include but is not limited to song name, performing singer, lyricist, composer, arranger, release time, album to which it belongs, song duration, lyric content, language type, etc. The song format quality information can be the information related to the song format and song quality corresponding to the intended song.

[0092] For the matched intended song, the basic song information can be analyzed to determine whether there is a playable intended song on the music platform currently. The basic song information can include but is not limited to song copyright information, song content information, song format quality information, etc. The sound source playback state of the song on the music platform may be affected by possible copyright problems, content problems, or technical problems of the intended song.

[0093] Specifically, with regard to copyright issues, the music platform may not be able to provide playback services for the intended song because it has not obtained the playback authorization for the intended song from the copyright holder; the authorization agreement between the copyright holder and the music platform may limit the playback area of ​​the song, so that the intended song can only be played in a specific area; in addition, the intended song may not be able to be played due to copyright expiration or disputes.

[0094] As for content issues, the song cannot be played because it contains illegal or other inappropriate information, which leads to content crowding. The song cannot be played normally or the playback effect is poor because the content of the song is incomplete or erroneous.

[0095] As for technical issues, some songs cannot be played because the software version of the music platform is too low, there may be compatibility issues or no support for new audio formats, etc. In addition, the music platform cannot play the intended song because the device or player where the music platform is located does not support the audio format of the song.

[0096] If the intended song does not have the above problems and other unplayable problems, the sound source playback status of the intended song will be in a playable state, and the intended song can be used as the target search result of the target hot search term. If the intended song has one or more of the above problems, the sound source playback status of the intended song will be unplayable, and the intended song in a playable state cannot be returned to the user at this time. Therefore, it is necessary to match the target matching resources corresponding to the intended song from the published business resources of the business party of the music platform, so that when there is no playable intended song on the music platform, the business party can promptly realize the resource acceptance of the hot search term, bring in more additional traffic to the business, and have the possibility of more traffic conversion.

[0097] In one embodiment of the present disclosure, for step S220, the associated business resources related to the intended song are determined from the published business resources, including: determining the initial business resources, which are business resources related to the music category; performing audio and video recognition processing on the initial business resources to determine the associated song information related to the initial business resources, the associated song information including the associated song identifier and the associated song index; establishing a business resource index corresponding to the initial business resource; based on the business resource index, storing the initial business resources and the associated song information in a resource database to generate published business resources; and determining the associated business resources based on the published business resources.

[0098] Among them, the initial business resources can be business resources related to music categories released by various business parties connected to the music platform. The associated song information can be the song information associated with the initial business resources. The associated song identifier can be the unique identifier of the associated song. The associated song metrics can be one or more dimensional metrics of the associated song corresponding to the initial business resources. For example, the associated song metrics can include, but are not limited to, metrics in multiple dimensions such as the start timestamp, end timestamp, audio similarity, time ratio, and vocal ratio of the audio segment of the song in the initial business resources. The business resource index can be the index corresponding to the initial business resources in the resource database. The resource database can be a database used to store the initial business resources.

[0099] Continue to refer to Figure 3 , each sub-business party of the music platform can perform resource publishing operations. The server needs to identify the resources published in different business domains, that is, the initial business resources related to music categories. Through the capabilities of audio-visual algorithms, it can identify the song information associated with the resources, such as information like the associated song ID and associated song metrics. For example, the server configures resource identification workflows such as the timestamp identification workflow and the similarity calculation workflow; through the timestamp identification workflow, it can identify the start and end times and playing duration of the intended song included in the initial business resources, and through the similarity calculation workflow, it can determine the similarity between the audio included in the initial business resources and the intended song.

[0100] Through the above workflows of the server, the business resources matching different intended songs are determined through resource selection operations; then, the associated data of each resource after data integration processing is stored in the database. During the process of resource storage in the database, the business resource index corresponding to the initial business resources can be determined first, and then based on the business resource index, the initial business resources and the associated song information are stored in the resource database to generate the published business resources. For the published business resources that have completed storage in the database, the associated business resources related to the intended song can be determined according to the resource screening conditions. Through the resource selection operation, the associated business resources of the intended song can be screened out from the resource database as the data basis for subsequent determination of the target matching resources.

[0101] In an embodiment of the present disclosure, determining the associated business resources according to the published business resources includes: performing an associated identification process on the published business resources based on the intended song to obtain initial associated resources, where the audio similarity between the initial associated resources and the intended song is greater than the similarity threshold, and / or the song duration ratio of the initial associated resources is greater than the duration ratio threshold; determining the associated business resources from the initial associated resources based on the pre-configured resource screening conditions; and determining the associated resource information corresponding to the associated business resources.

[0102] Among them, the initial associated resource can be an initial business resource whose similarity degree with the intended song reaches the specified similarity screening condition. The audio similarity can be the similarity degree between the audio segment in the initial associated resource and the intended song in terms of audio content and features, etc. The similarity threshold can be a reference value for comparison with the audio similarity. The song duration ratio can be the proportional relationship between the duration of the song audio segment of the intended song included in the initial associated resource and the total audio duration of the initial associated resource. The duration ratio threshold can be a reference value for comparison with the song duration ratio. The resource screening condition can be a pre-configured screening condition for screening associated business resources from the initial associated resources.

[0103] For the published business resources in the resource database, resource identification is performed on them to determine the association degree between the published business resources and the intended song, and the published business resources that meet the specified conditions are used as the initial associated resources. Resource screening is performed according to the audio similarity between the published business resources and the intended song, as well as the song duration ratio of the published business resources, to determine the initial associated resources. For example, the determined initial associated resources meet one or both of the following conditions: the audio similarity between the initial associated resource and the intended song is greater than the similarity threshold; the song duration ratio of the initial associated resource is greater than the duration ratio threshold.

[0104] The resource identification process can include but is not limited to song timestamp identification, audio similarity calculation, and song duration ratio, etc. Song timestamp identification refers to analyzing and comparing through feature point sampling to obtain which time interval segments in the sound contain which songs, and obtaining these songs and their start and end timestamps. The audio similarity can be obtained by performing voiceprint matching through sampling at a specified duration (every x-second segment), performing feature point matching counting, and obtaining a similarity between the two audios through weighted calculation, with a value range of 0-1, and the closer to 1, the more similar. Song duration ratio: By performing voiceprint matching through sampling at a specified duration (every x-second segment), performing feature point matching counting, and obtaining the ratio of the duration occupied by the song in the entire sound duration through weighted calculation, with a value range of 0-1, and the closer to 1, the more similar.

[0105] After determining the initial associated resources, it is necessary to further clean the initial associated resources to clean out the associated business resources that match the intended song. Specifically, based on the pre-configured resource screening conditions, the associated business resources are determined from the initial associated resources, and at the same time, the associated resource information corresponding to the associated business resources is determined to facilitate subsequent resource warehousing operations.

[0106] In one embodiment of the present disclosure, determining the associated resource information corresponding to the associated service resource includes: determining the resource matching metrics corresponding to the associated service resource, where the resource matching metrics include one or more of resource quality score, resource popularity score, included songs in the resource, number of included songs, resource audio similarity, resource audio matching duration, resource audio proportion duration, and resource content type; generating the associated resource information based on the resource matching metrics, and storing the associated resource information in the resource database.

[0107] For business data, in order to match the data of the cleaned data source (such as the intended song), the business party needs to match the relevant resource information for the associated service resource as the associated resource information. Taking podcast resources as an example, podcast works can include, but are not limited to, music podcasts, creative covers, and other work categories, which have a very high degree of fit with songs, and at the same time, the audio content contains a large amount of song information. For such associated resources, the following various resource matching metrics can be calculated, including but not limited to resource quality score, resource popularity score, included songs in the resource, number of included songs, resource audio similarity, resource audio matching duration, resource audio proportion duration, and resource content type. The resource matching metrics can be metrics related to the degree of fit between the audio resources included in the associated service resource and the intended song.

[0108] Specifically, the resource quality score can be a quantitative metric value used to evaluate the pros and cons of the associated service resource itself. The resource popularity score can be a quantitative metric value measuring the popularity and attention degree of the associated service resource. The included songs in the resource can be the intended songs included in the associated service resource. The number of included songs can be the number of songs included in the associated service resource, and an associated service resource may include one or more song segments. The resource audio similarity can be the degree of similarity between the audio segments included in the associated service resource and the intended song. The resource audio matching duration can be the length of the mutually matching part between the audio segments in the associated service resource and the intended song. The resource audio proportion duration can be the duration proportion occupied by the vocal part of the intended song in the audio segments included in the associated service resource. The resource content type can be the specific type corresponding to the associated service resource, such as whether the associated service resource belongs to the audio type of pure background music (Background Music, BGM).

[0109] Through the above resource matching metrics, it is possible to determine indicators such as the quality of the work resource itself, its popularity, which songs are included, whether the number of songs is complex or single, the similarity between the included song audio and the original singer, the duration matching degree between the included song audio and the original singer, the proportion of the vocal part in the included song audio, and whether it is pure BGM. All of the above indicators need to be calculated and stored offline by the business party in advance for automated resource matching.

[0110] In some other exemplary embodiments of the present disclosure, the resource matching metrics may also be specifically selected and configured according to the matching requirements, and the present disclosure does not make any special limitations on the specific content of the resource matching metrics.

[0111] In one implementation manner of the present disclosure, for step S230, based on the intent song identifier of the intent song, determining a target matching resource corresponding to the target hot search term from the associated service resources includes: performing an idempotent filtering process on the associated service resources based on the intent song identifier to obtain an initial matching resource corresponding to the target hot search term; obtaining a pre-configured resource matching rule, and determining candidate matching resources from the initial matching resources according to the resource matching rule and the intent song identifier; determining the resource search level corresponding to the candidate matching resources, and performing an associated tagging process on the candidate matching resources, the target hot search term, and the resource search level to obtain the search level weight corresponding to the candidate matching resources; and determining the target matching resource from the candidate matching resources according to the search level weight, and displaying the target matching resource.

[0112] Among them, the idempotent filtering process may be a processing method used to ensure that for the same operation or request, the result remains consistent whether it is executed once or multiple times. The initial matching resource may be a service resource determined by performing an idempotent filtering process on the associated service resources based on the intent song identifier and matching the intent song. The resource matching rule may be a pre-configured rule for automatic resource matching. The candidate matching resource may be a service resource filtered out from the initial matching resources through the resource matching rule. The resource search level may be the priority for the selected matching resource to be searched and matched. The associated tagging process may be a process of associating and tagging the resource content, the target hot search (query term), and the corresponding search resource level for the successfully matched resources. The search level weight may be the calculated weight corresponding to the resource search level.

[0113] Continue to refer to Figure 3 , for the target matching term, automatic resource matching can be achieved through processing such as an idempotent filtering operation, a content retrieval operation, a weight sorting operation, and a matching storage operation. Refer to Figure 4 , Figure 4 Schematically shows a flowchart of automatic matching based on a target hot search term according to some embodiments of the present disclosure. In Figure 4 , when performing automatic matching on the target hot search term, a query idempotent filtering process may be first performed based on the target hot search term. Through the idempotent filtering process, it can be ensured that the same initial matching resource is returned whether it is a single search or multiple searches.

[0114] After obtaining the initial matching resources, a content retrieval method is used for resource matching. Operation personnel or other relevant business personnel can pre-configure certain resource matching rules and screen resources through the resource matching rules. Content retrieval can be performed within the business resource pool (CIO pool) based on the intended song identifier (such as the field SongId); the resource matching rules can include, but are not limited to, the resource classification including music podcasts or creative covers, not including the "not recommended" label, only including one associated intended song, the associated intended song not being pure BGM, the audio similarity being greater than the specified similarity threshold, the duration matching degree being greater than the specified duration matching threshold, etc., and more high-quality business resources are selected from the calculated resource circle. In addition, real-time filtering rules can be configured according to real-time screening requirements, and some business resources that do not meet the requirements are filtered out through the real-time rules, and high-quality resources that can be used for resource matching are selected.

[0115] For the initial matching resources obtained through the above steps, using the intended song identifier (song Id) as the associated medium, resource matching is performed. During the resource matching process, multiple high-quality resources may be simultaneously matched as candidate matching resources. In this case, it is necessary to further sort the candidate matching resources, and the best matching resource obtained from the search is used as the target matching resource, and only one target matching resource is output.

[0116] Determine the resource search level corresponding to the above candidate matching resources, and then perform an associated tagging process on each candidate matching resource, its corresponding target hot search term, and the resource search level. After the tagging process is completed, the search level weight corresponding to the candidate matching resource is obtained. Subsequently, according to the search level weight of the candidate matching resources, the target matching resource that best matches the target hot search term is determined from the above candidate matching resources, and the target matching resource is displayed.

[0117] In another resource matching scenario of the present disclosure, after resource matching processing, the number of target matching resources that match the intended song can be multiple (such as 5). For example, the top 5 candidate matching resources determined according to the search level weight are used as the target matching resources and output in a list form in the order of priority.

[0118] In an embodiment of the present disclosure, determining the target matching resource from the candidate matching resources according to the search level weight includes: determining the associated matching index corresponding to the candidate matching resource, and the associated matching index includes one or more of the similarity between the resource name and the target hot search term, the resource play volume, and the resource timeliness; determining the associated matching weight corresponding to the candidate matching resource according to the associated matching index; and determining the target matching resource from the candidate matching resources according to the search level weight and the associated matching weight.

[0119] Among them, the associated matching index can be an index reflecting the degree of associated matching between the candidate matching resource and the intent song determined based on the target hot search term. The resource name can be the name of the candidate matching resource. When the candidate matching resource is a music podcast work, the resource name can be the name of the music podcast. The similarity with the target hot search term can be the degree of text similarity between the resource name of the candidate matching resource and the target hot search term. The resource playback volume can be the number of times the candidate matching resource is played by users. The resource timeliness can be the value and usability of the candidate matching resource to users within a certain time range. The associated matching weight can be a calculation weight determined according to the associated matching index and used to reflect the degree of associated matching between the candidate matching resource and the target hot search term.

[0120] In a scenario where only one target matching resource can be finally output, at this time, weighted scoring needs to be combined with the characteristic indexes of the candidate matching resource itself. For example, according to the candidate matching resource, its corresponding associated matching index can be determined. The associated matching index can include, but is not limited to, multi-dimensional indexes such as the similarity between the resource name and the search term, the playback volume of the resource itself, and the timeliness of the resource. By normalizing the above-mentioned associated matching indexes in multiple dimensions, the associated matching weight corresponding to the candidate matching resource can be determined.

[0121] Combining the associated matching weight of the candidate matching resource itself with the search level weight, the candidate matching resources can be weighted and sorted, and finally the optimal target matching resource can be determined. Through the matching mechanism in the automatic matching process, the process of quickly matching the optimal resource is realized.

[0122] After the resource matching is successful, the matching result and relevant data can be archived in the database. For example, the target matching resource matched based on the target hot search term and related resource information can be stored in the database. Refer to Figure 5 , Figure 5 Schematically shows a flowchart of resource review and automatic marking according to some embodiments of the present disclosure. In order to prevent inappropriate situations in the matched resources, especially for extremely high hot search query terms, it is necessary to be very cautious about the matched resources. Therefore, manual review can be accessed. For the completed matched resources, it can first be determined whether the resource requires manual review. If it is determined that the resource requires manual review, it will be archived after being marked through manual review. Manual resource review can refer to query terms in the TopX of the pv list of hot search terms and soaring terms. In order to ensure the accuracy of system matching, a round of manual review is required. The resources that pass the review are manually marked to raise the search level.

[0123] However, for non-header query terms, since there are a large number of them, it would be too costly to connect all of them to manual review. Therefore, the system needs to configure an automatic search rank boosting rule. For resources that do not require manual review, in order to distinguish them from resources that undergo manual review, this is achieved through the setting of different search ranks. The search rank of such resources is automatically boosted through the search rank boosting rule to achieve search rank boosting. Search rank boosting can refer to the operation of boosting the search rank of a successfully matched resource and associating it with the corresponding query term to obtain higher resource exposure. To distinguish the operation of the system automatically boosting the search rank from the manual review operation, this can be achieved through the setting of different search ranks. Based on the automated search rank boosting design, the matched resources can be quickly applied to the business, and business benefits can be obtained quickly.

[0124] In addition, due to the different weights of different query terms, the associated tagging strategies are also different. For example, for words with a high search PV, in case of any matching errors for the matched resources, after subsequent confirmation by operation review, the search rank is boosted; for words with a non-high search PV, automated search rank boosting can be directly performed. In the later stage of resource matching, when the operation review passing rate approaches 100% after continuous optimization of the matching algorithm, automated search rank boosting can be fully enabled to achieve the automated connection of the entire link.

[0125] In summary, for the resource recommendation method of the present disclosure, the target hot search terms are obtained, and the intent songs matching the target hot search terms are determined based on the song play metric data. The target hot search terms are the initial search terms that meet the hot search term screening conditions; when there are no playable intent songs on the music platform, the associated business resources related to the intent songs are determined from the already published business resources; based on the intent song identifier of the intent song, the target matching resources corresponding to the target hot search terms are determined from the associated business resources as the target search results of the target hot search terms. On the one hand, a full-link automated implementation solution from hot search term statistics to resource matching is provided, which can save a large amount of time and labor costs. On the other hand, by combining hot search term statistics with the matching mechanism, the target matching resources corresponding to the target hot search terms can be quickly matched, improving the timeliness of hot search term resource connection, especially for short-lived forms such as hot topics. On the other hand, through timely connection of hot resources, it can bring more traffic introduction to the business.

[0126] Exemplary Device

[0127] After introducing the method of the exemplary embodiment of the present disclosure, next, refer to Figure 6 to describe the resource recommendation device of the exemplary embodiment of the present disclosure.

[0128] In Figure 6Among them, the resource recommendation device 600 may include: an intended song determination module 610, an associated resource determination module 620, and a search result determination module 630.

[0129] Among them: the intended song determination module 610 is used to obtain a target hot search term and determine an intended song that matches the target hot search term based on song play metric data, where the target hot search term is an initial search term that meets the hot search term screening conditions; the associated resource determination module 620 is used to determine associated business resources related to the intended song from the published business resources when there is no playable intended song on the music platform; the search result determination module 630 is used to determine a target matching resource corresponding to the target hot search term from the associated business resources based on the intended song identifier of the intended song as the target search result of the target hot search term.

[0130] In an embodiment of the present disclosure, the intended song determination module 610 includes an intended song determination unit, which is used to: obtain an initial search term and a pre-configured hot search term screening condition; determine candidate hot search terms from the initial search terms based on a preset time interval and the hot search term screening condition, where the candidate hot search terms include the target hot search term; determine the song play metric data corresponding to the candidate songs in the song library, and the song play metric data includes one or more of song click-through rate, click event data, and play event data; determine the intended song and the corresponding intended song identifier of the intended song from the candidate songs based on the song play metric data and the target hot search term.

[0131] In an embodiment of the present disclosure, the intended song determination unit includes a hot search term determination subunit, which is used to: obtain a search term data table corresponding to the initial search term; obtain the hot search term screening condition, where the hot search term screening condition includes one or more of the search access volume being greater than the search volume threshold and the month-on-month growth rate of the search access volume being greater than the growth rate threshold; perform search term screening processing on the search term data table based on a preset time interval and the hot search term screening condition to obtain candidate hot search terms.

[0132] In an embodiment of the present disclosure, the resource recommendation device 600 further includes a play status determination module, which is used to: determine the basic song information of the intended song, where the basic song information includes one or more of song copyright information, song content information, and song format quality information; determine the sound source play status corresponding to the intended song according to the basic song information, and the sound source play status includes a playable state and an unplayable state.

[0133] In one embodiment of the present disclosure, the associated resource determination module 620 includes an associated resource determination unit for: determining an initial service resource, where the initial service resource is a service resource related to the music category; performing audio-visual recognition processing on the initial service resource to determine associated song information related to the initial service resource, where the associated song information includes an associated song identifier and an associated song metric; establishing a service resource index corresponding to the initial service resource; storing the initial service resource and the associated song information in a resource database based on the service resource index to generate a published service resource; and determining an associated service resource according to the published service resource.

[0134] In one embodiment of the present disclosure, the associated resource determination unit includes an associated resource determination subunit for: performing an association recognition process on the published service resource based on the intended song to obtain an initial associated resource, where the audio similarity between the initial associated resource and the intended song is greater than a similarity threshold, and / or the song duration ratio of the initial associated resource is greater than a duration ratio threshold; determining an associated service resource from the initial associated resources based on a pre-configured resource screening condition; and determining associated resource information corresponding to the associated service resource.

[0135] In one embodiment of the present disclosure, the associated resource determination unit includes a resource information determination subunit for: determining a resource matching metric corresponding to the associated service resource, where the resource matching metric includes one or more of a resource quality score, a resource popularity score, an included song in the resource, the number of included songs, a resource audio similarity, a resource audio matching duration, a resource audio ratio duration, and a resource content type; generating associated resource information based on the resource matching metric and storing the associated resource information in the resource database.

[0136] In one embodiment of the present disclosure, the search result determination module 630 includes a target resource determination unit for: performing an idempotent filtering process on the associated service resource based on the intended song identifier to obtain an initial matching resource corresponding to the target hot search term; obtaining a pre-configured resource matching rule, and determining a candidate matching resource from the initial matching resources according to the resource matching rule and the intended song identifier; determining a resource search level corresponding to the candidate matching resource, and performing an association marking process on the candidate matching resource, the target hot search term, and the resource search level to obtain a search level weight corresponding to the candidate matching resource; and determining a target matching resource from the candidate matching resources according to the search level weight and displaying the target matching resource.

[0137] In one embodiment of the present disclosure, the target resource determination unit includes a target resource determination subunit for: determining an associated matching metric corresponding to a candidate matching resource, where the associated matching metric includes one or more of the similarity between the resource name and the target hot search term, the resource playback volume, and the resource timeliness; determining an associated matching weight corresponding to the candidate matching resource according to the associated matching metric; and determining a target matching resource from the candidate matching resources according to the search level weight and the associated matching weight.

[0138] Since each functional module of the resource recommendation device in the exemplary embodiment of the present disclosure corresponds to the steps in the exemplary embodiment of the above resource recommendation method, for the details not disclosed in the device embodiment of the present disclosure, please refer to the embodiment of the above resource recommendation method of the present disclosure, which will not be elaborated here.

[0139] It should be noted that although several modules or units of the resource recommendation device are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of the two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0140] Exemplary Medium

[0141] After introducing the device of the exemplary embodiment of the present disclosure, next, refer to Figure 7 to describe the storage medium of the exemplary embodiment of the present disclosure.

[0142] In some embodiments, various aspects of the present disclosure can also be implemented as a medium having program code stored thereon, which, when executed by a processor of a device, is used to implement the steps in the resource recommendation method according to various exemplary embodiments of the present disclosure described in the above "Exemplary Method" section of this specification.

[0143] For example, when the processor of the device executes the program code, it can implement steps such as Figure 2 step S210 described in, obtaining a target hot search term, and determining an intent song that matches the target hot search term based on song playback metric data, where the target hot search term is an initial search term that meets the hot search term screening condition; step S220, when there is no playable intent song on the music platform, determining an associated business resource related to the intent song from the published business resources; step S230, based on the intent song identifier of the intent song, determining a target matching resource corresponding to the target hot search term from the associated business resources as the target search result of the target hot search term.

[0144] Refer to Figure 7As shown, a program product 700 for implementing the above-mentioned resource recommendation method or implementing the above-mentioned resource recommendation method according to an embodiment of the present disclosure is described. It can use a portable compact disc read-only memory (CD-ROM), include program code, and can run on a terminal device, such as a personal computer. However, the program product of the present disclosure is not limited to this.

[0145] The program product can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or component, or any combination of the above. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0146] The computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, which carries the readable program code. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The readable signal medium can also be any readable medium other than the readable storage medium.

[0147] The program code for performing the operations of the present disclosure can be written in any combination of one or more programming languages, including object-oriented programming languages such as Java, C++, etc., and also including conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computing device, partially on the user's device, partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user's computing device through any type of network, including a local area network (LAN) or a wide area network (WAN).

[0148] Exemplary Computing Device

[0149] After introducing the resource recommendation method, resource recommendation device, and storage medium of the exemplary embodiments of the present disclosure, next, reference is made to Figure 8 to describe the electronic device of the exemplary embodiments of the present disclosure.

[0150] Those skilled in the art can understand that various aspects of the present disclosure can be implemented as a system, a method, or a program product. Therefore, various aspects of the present disclosure can be specifically implemented in the following forms, namely: a complete hardware embodiment, a complete software embodiment (including firmware, microcode, etc.), or an embodiment combining hardware and software aspects, which can be collectively referred to as "circuits", "modules", or "systems" here.

[0151] In some possible embodiments, an electronic device according to the present disclosure may include at least one processing unit and at least one storage unit. Among them, the storage unit stores program code, and when the program code is executed by the processing unit, the processing unit executes the steps in the resource recommendation method according to various exemplary embodiments of the present disclosure described in the above "Exemplary Method" section of this specification. For example, the processing unit may execute steps such as Figure 2 shown in S210, obtain a target hot search term, determine an intention song that matches the target hot search term based on song playback metric data, where the target hot search term is an initial search term that meets the hot search term screening conditions; step S220, when there is no playable intention song on the music platform, determine an associated business resource related to the intention song from the published business resources; step S230, based on the intention song identifier of the intention song, determine a target matching resource corresponding to the target hot search term from the associated business resources as the target search result of the target hot search term.

[0152] Next, refer to Figure 8 to describe the electronic device 800 according to the exemplary embodiments of the present disclosure. Figure 8 The electronic device 800 shown is merely an example and should not impose any limitation on the functions and usage scope of the embodiments of the present disclosure.

[0153] As Figure 8 shown, the electronic device 800 is presented in the form of a general-purpose computing device. The components of the electronic device 800 may include, but are not limited to: the above at least one processing unit 801, the above at least one storage unit 802, a bus 803 connecting different system components (including the storage unit 802 and the processing unit 801), and a display unit 807.

[0154] The bus 803 represents one or more of several types of bus structures, including a memory bus or a memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus using any bus structure in a variety of bus structures.

[0155] The storage unit 802 may include a readable medium in the form of a volatile memory, such as a random access memory (RAM) 821 and / or a cache memory 822, and may further include a read-only memory (ROM) 823.

[0156] The storage unit 802 may also include a program / utilities 825 having a set (at least one) of program modules 824. Such program modules 824 include, but are not limited to: an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment.

[0157] The electronic device 800 may also communicate with one or more external devices 804 (such as a keyboard, a pointing device, a Bluetooth device, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 800, and / or may communicate with any device that enables the electronic device 800 to communicate with one or more other computing devices (such as a router, a modem, etc.). Such communication may be carried out through an input / output (I / O) interface 805. Moreover, the electronic device 800 may also communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through a network adapter 806. As shown in the figure, the network adapter 806 communicates with other modules of the electronic device 800 through a bus 803. It should be understood that although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 800, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0158] It should be noted that although several units / modules or sub-units / modules of the resource recommendation device are mentioned in the above detailed description, this division is merely exemplary and not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more of the above-described units / modules may be embodied in one unit / module. Conversely, the features and functions of one unit / module described above may be further divided and embodied by multiple units / modules.

[0159] In addition, although the operations of the method of the present disclosure are described in a specific order in the drawings, this does not require or imply that these operations must be performed in that specific order, or that all of the shown operations must be performed to achieve the desired result. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step for execution, and / or one step may be decomposed into multiple steps for execution.

[0160] Although the spirit and principles of the present disclosure have been described with reference to several specific embodiments, it should be understood that the present disclosure is not limited to the specific embodiments disclosed, and the division of each aspect does not mean that the features in these aspects cannot be combined for benefit. This division is only for the convenience of description. The present disclosure is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.

Claims

1. A resource recommendation method, characterized in that: include: Obtain a target hot search word, and determine an intended song matching the target hot search word based on the song play index data, wherein the target hot search word is an initial search word that satisfies the hot search word screening condition; When there is no playable intended song on the music platform, determine associated business resources related to the intended song from published business resources; Based on the intended song identifier of the intended song, a target matching resource corresponding to the target hot search word is determined from the associated business resources as a target search result of the target hot search word.

2. The method according to claim 1, characterized in that The step of obtaining a target hot search word and determining an intended song matching the target hot search word based on song play index data includes: Get the initial search terms and pre-configured hot search terms filter conditions; Determining candidate hot search words from the initial search words based on a preset time interval and the hot search word screening condition, the candidate hot search words including the target hot search word; Determine song play index data corresponding to candidate songs in the song library, wherein the song play index data includes one or more of song click rate, click buried point data, and play buried point data; Based on the song play index data and the target hot search term, the intended song and the intended song identifier corresponding to the intended song are determined from the candidate songs.

3. The method according to claim 2, characterized in that The determining the candidate hot search words from the initial search words based on the preset time interval and the hot search word screening condition includes: Obtaining a search term data table corresponding to the initial search term; Acquire the hot search word screening condition, wherein the hot search word screening condition includes one or more of the following: the search access volume is greater than a search volume threshold, and the month-on-month growth rate of the search access volume is greater than a growth rate threshold; Based on the preset time interval and the hot search word screening condition, the search word data table is subjected to search word screening processing to obtain the candidate hot search words.

4. The method according to claim 2, characterized in that: The method further comprises: Determine basic song information of the intended song, wherein the basic song information includes one or more of song copyright information, song content information, and song format quality information; The sound source playing state corresponding to the intended song is determined according to the basic information of the song, and the sound source playing state includes a playable state and a non-playable state.

5. The method according to claim 1, characterized in that The determining of associated business resources related to the intended song from the published business resources includes: Determining initial business resources, where the initial business resources are business resources related to the music category; Performing audio and video recognition processing on the initial business resource to determine associated song information related to the initial business resource, wherein the associated song information includes an associated song identifier and an associated song index; Establishing a business resource index corresponding to the initial business resource; Based on the business resource index, the initial business resource and the associated song information are stored in a resource database to generate the published business resource; The associated service resource is determined according to the published service resource.

6. The method according to claim 5, characterized in that The determining the associated service resource according to the published service resource includes: Based on the intended song, the published business resources are subjected to association identification processing to obtain an initial associated resource, wherein the audio similarity between the initial associated resource and the intended song is greater than a similarity threshold, and / or the song duration ratio of the initial associated resource is greater than a duration ratio threshold; Based on a pre-configured resource screening condition, determining the associated service resource from the initial associated resources; Determine the associated resource information corresponding to the associated service resource.

7. The method according to claim 1, characterized in that The determining, based on the intended song identifier of the intended song, a target matching resource corresponding to the target hot search term from the associated business resources includes: Based on the intended song identifier, the associated business resources are subjected to idempotent filtering to obtain initial matching resources corresponding to the target hot search term; Obtaining a pre-configured resource matching rule, and determining a candidate matching resource from the initial matching resources according to the resource matching rule and the intended song identifier; Determine the resource search level corresponding to the candidate matching resource, and perform association marking processing on the candidate matching resource, the target hot search term and the resource search level to obtain the corresponding search level weight of the candidate matching resource; According to the search level weight, the target matching resource is determined from the candidate matching resources, and the target matching resource is displayed.

8. A resource recommendation device, characterized in that: include: An intended song determination module is used to obtain a target hot search word and determine an intended song matching the target hot search word based on song playback index data, wherein the target hot search word is an initial search word that satisfies a hot search word screening condition; An associated resource determination module, used to determine associated business resources related to the intended song from published business resources when there is no intended song that can be played on the music platform; A search result determination module is used to determine a target matching resource corresponding to the target hot search word from the associated business resources based on the intended song identifier of the intended song, as a target search result of the target hot search word.

9. An electronic device, characterized in that: include: processor; as well as A memory having computer-readable instructions stored thereon, wherein the computer-readable instructions, when executed by the processor, implement the resource recommendation method according to any one of claims 1 to 7.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the resource recommendation method according to any one of claims 1 to 7 is implemented.

Images