Search associational word display method, computer equipment, readable storage medium and program product

Abstract

The invention relates to a search associational word display method, computer equipment, a readable storage medium and a program product, and the method comprises the steps: obtaining a plurality of associational words according to a search word input by a user in a music application; obtaining the correlation between each associational word and the search word, and the historical interaction feature, the real-time interaction feature and the soaring interaction feature of each associational word; according to the correlation, the historical interaction features, the real-time interaction features and the soaring interaction features, sorting the associational words; and displaying each associational word according to a sorting result. According to the method, the associational words are sorted by integrating the characteristics and the relevance characteristics of the associational words and the search words, the relevance and real-time characteristics of the associational words can be considered at the same time, and the history and popularity conditions of the associational words are also considered, so that the sorting result can respond to the hotspots in real time, and the balance between the hotspots and tradition is achieved.

Description

Technical Field

[0001] The present application relates to the field of Internet technology, and in particular to a method, apparatus, computer device, computer-readable storage medium, and computer program product for displaying search association words. Background Art

[0002] With the rapid development of digital and mobile music, numerous songs are created every day, creating numerous popular variety shows and songs that are trending online. To ensure that the associated words in user searches reflect these trends, this information needs to be promptly stored and prioritized for display, allowing users to quickly and easily find relevant associated words.

[0003] Related art solutions for responding to hot topics in real time are generally based on the real-time update mechanism of the recommendation system. This is achieved by having the recommendation system generate embeddings based on newly released songs or emerging hot words in real time to respond to hot topics in real time, ensuring their rapid recall by the recommendation system. However, in the context of music search for associated words, hot topic response solutions based on recommendation systems tend to be overly biased towards real-time hot topics, causing the recommended associated words to deviate from the user's actual needs. Therefore, traditional hot topic response methods based on recommendation systems are difficult to apply. Summary of the Invention

[0004] Based on this, it is necessary to provide a search association word display method, device, computer equipment, computer-readable storage medium and computer program product to address the technical problem that the traditional real-time response solution to heat is not suitable for search association words.

[0005] In a first aspect, the present application provides a method for displaying search association words. The method comprises:

[0006] Obtain multiple associated words based on the search words entered by the user in the music application;

[0007] Obtain the relevance of each of the associative words with the search word, the historical interaction characteristics, the real-time interaction characteristics, and the surge interaction characteristics of each of the associative words; the historical interaction characteristics represent the characteristics of the interaction behaviors implemented by all users in the music application on the music resources corresponding to the associative words in a first historical time period; the real-time interaction characteristics represent the characteristics of the interaction behaviors implemented by all users in the music application on the music resources corresponding to the associative words in a current time period; the surge interaction characteristics represent the characteristics of the interaction behaviors implemented by all users in the music application on the music resources corresponding to the associative words in a second historical time period, the number of which the statistical index of the increase is greater than a preset threshold;

[0008] sorting the associated words according to the relevance of each associated word to the search term, the historical interaction feature of each associated word, the real-time interaction feature of each associated word, and the soaring interaction feature of each associated word to obtain a sorting result of the multiple associated words;

[0009] According to the ranking results of the plurality of associated words, each of the associated words is displayed in the music application.

[0010] In a second aspect, the present application also provides a search association word display device. The device includes:

[0011] The associative word acquisition module is used to obtain multiple associative words based on the search words entered by the user in the music application;

[0012] A feature acquisition module is used to obtain the relevance between each of the associated words and the search term, the historical interaction features, real-time interaction features, and surge interaction features of each of the associated words; the historical interaction features represent the features of the interaction behaviors implemented by all users of the music application on the music resource corresponding to the associated word within a first historical time period; the real-time interaction features represent the features of the interaction behaviors implemented by all users of the music application on the music resource corresponding to the associated word within a current time period; the surge interaction features represent the features of the interaction behaviors implemented by all users of the music application on the music resource corresponding to the associated word within a second historical time period, the number of which the statistical index of the increase is greater than a preset threshold;

[0013] an associative word ranking module, configured to rank each of the associative words according to a relevance between each of the associative words and the search term, a historical interaction feature of each of the associative words, a real-time interaction feature of each of the associative words, and a soaring interaction feature of each of the associative words, to obtain a ranking result of the plurality of associative words;

[0014] The associative word display module is used to display each of the associative words in the music application according to the sorting results of the multiple associative words.

[0015] In a third aspect, the present application further provides a computer device. The computer device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are performed:

[0016] Obtain multiple associated words based on the search words entered by the user in the music application;

[0017] Obtain the relevance of each of the associative words with the search word, the historical interaction characteristics, the real-time interaction characteristics, and the surge interaction characteristics of each of the associative words; the historical interaction characteristics represent the characteristics of the interaction behaviors implemented by all users in the music application on the music resources corresponding to the associative words in a first historical time period; the real-time interaction characteristics represent the characteristics of the interaction behaviors implemented by all users in the music application on the music resources corresponding to the associative words in a current time period; the surge interaction characteristics represent the characteristics of the interaction behaviors implemented by all users in the music application on the music resources corresponding to the associative words in a second historical time period, the number of which the statistical index of the increase is greater than a preset threshold;

[0018] sorting the associated words according to the relevance of each associated word to the search term, the historical interaction feature of each associated word, the real-time interaction feature of each associated word, and the soaring interaction feature of each associated word to obtain a sorting result of the multiple associated words;

[0019] According to the ranking results of the plurality of associated words, each of the associated words is displayed in the music application.

[0020] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the following steps:

[0021] Obtain multiple associated words based on the search words entered by the user in the music application;

[0022] Obtain the relevance of each of the associative words with the search word, the historical interaction characteristics, the real-time interaction characteristics, and the surge interaction characteristics of each of the associative words; the historical interaction characteristics represent the characteristics of the interaction behaviors implemented by all users in the music application on the music resources corresponding to the associative words in a first historical time period; the real-time interaction characteristics represent the characteristics of the interaction behaviors implemented by all users in the music application on the music resources corresponding to the associative words in a current time period; the surge interaction characteristics represent the characteristics of the interaction behaviors implemented by all users in the music application on the music resources corresponding to the associative words in a second historical time period, the number of which the statistical index of the increase is greater than a preset threshold;

[0023] sorting the associated words according to the relevance of each associated word to the search term, the historical interaction feature of each associated word, the real-time interaction feature of each associated word, and the soaring interaction feature of each associated word to obtain a sorting result of the multiple associated words;

[0024] According to the ranking results of the plurality of associated words, each of the associated words is displayed in the music application.

[0025] In a fifth aspect, the present application further provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the following steps:

[0026] Obtain multiple associated words based on the search words entered by the user in the music application;

[0027] Obtain the relevance of each of the associative words with the search word, the historical interaction characteristics, the real-time interaction characteristics, and the surge interaction characteristics of each of the associative words; the historical interaction characteristics represent the characteristics of the interaction behaviors implemented by all users in the music application on the music resources corresponding to the associative words in a first historical time period; the real-time interaction characteristics represent the characteristics of the interaction behaviors implemented by all users in the music application on the music resources corresponding to the associative words in a current time period; the surge interaction characteristics represent the characteristics of the interaction behaviors implemented by all users in the music application on the music resources corresponding to the associative words in a second historical time period, the number of which the statistical index of the increase is greater than a preset threshold;

[0028] sorting the associated words according to the relevance of each associated word to the search term, the historical interaction feature of each associated word, the real-time interaction feature of each associated word, and the soaring interaction feature of each associated word to obtain a sorting result of the multiple associated words;

[0029] According to the ranking results of the plurality of associated words, each of the associated words is displayed in the music application.

[0030] The above-mentioned search association word display method, device, computer device, storage medium, and computer program product obtain multiple association words based on the search word entered by the user in the music application, further obtain the relevance of each association word with the search word, the historical interaction characteristics, real-time interaction characteristics, and soaring interaction characteristics of each association word; sort each association word based on the relevance of each association word with the search word, the historical interaction characteristics, real-time interaction characteristics, and soaring interaction characteristics of each association word; and display each association word according to the sorting results. In this method, the historical interaction characteristics can reflect the long-term popularity changes of the association word, the real-time interaction characteristics can reflect the recent situation of the association word, and the soaring interaction characteristics can reflect the popularity of the association word. By combining these characteristics and the relevance characteristics of the association word with the search word to sort the association words, it is possible to simultaneously take into account the relevance and real-time characteristics of the association word, as well as the history and popularity of the association word, so that the sorting results can respond to hot spots in real time while not only reflecting hot spots, achieving a balance between hot spots and tradition. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1A flowchart of a method for displaying search association words in one embodiment is shown;

[0032] Figure 2 Schematic diagram of the process of database update step in one embodiment;

[0033] Figure 3 A schematic flow chart of a step for determining a term quality score in one embodiment;

[0034] Figure 4 A framework diagram of a real-time process for searching for associative words in one embodiment;

[0035] Figure 5 A schematic diagram of a process for updating a database in another embodiment;

[0036] Figure 6 A schematic diagram of a process for sorting search association words in one embodiment;

[0037] Figure 7 A structural block diagram of a search association word display device in one embodiment;

[0038] Figure 8 FIG. 1 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0039] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0040] It should be noted that the terms "first," "second," and the like in the specification and claims of this application and the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate, such that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein.

[0041] In one embodiment, Figure 1As shown, a method for displaying search associated words is provided. In this embodiment, the method is illustrated by taking its application to a terminal as an example. It can be understood that the method can also be applied to a server, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. Among them, the terminal can be, but is not limited to, various personal computers, laptop computers, smart phones, tablet computers, Internet of Things devices, and portable wearable devices. The Internet of Things devices can be smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, etc. The portable wearable devices can be smart watches, smart bracelets, head-mounted devices, etc. The server can be implemented by an independent server or a server cluster composed of multiple servers.

[0042] In this embodiment, the method includes the following steps:

[0043] Step S110, obtain multiple associated words according to the search term input by the user in the music application.

[0044] Among them, the search term can be understood as the keyword input by the user in the search box of the music application. For example, the singer name, song name, album name, etc., or partial words in the singer name, song name, album name.

[0045] Among them, the associated word refers to the entry that may be associated with the search term input by the user. For example, the associated word can be an entry that matches the text of the search term (including prefix matching, fuzzy matching, pinyin matching, etc.). For example, when entering "Zhou", the associated word can be the singer name, song name, album name, etc. that contain the character "Zhou"; the associated word can also be obtained according to the music label (genre, scene, emotion, etc.). For example, when entering "rock", the associated word is some rock singers, rock bands, etc.

[0046] Specifically, the terminal can respond to the input operation of the user in the search box of the music application, obtain the search term input by the user, and further obtain multiple associated words associated with the search term.

[0047] For example, when the user enters the singer name "Wang xx" in the search box, the obtained associated words can be the song names sung by this singer. In another case, the search term may also be partial keywords of an entity input by the user, and the associated word is its complete keyword. For example, when the user enters the partial word "Wang" in the search box, the obtained associated words can be the singer names that contain the character "Wang", or the song names, album names, etc. that contain the character "Wang".

[0048] Step S120, obtain the relevance between each associated word and the search term, the historical interaction characteristics, real-time interaction characteristics, and soaring interaction characteristics of each associated word.

[0049] The interactive features may include the number of clicks, searches, favorites, plays, and shares of users; click-through rate (clicks / number of impressions), favorite rate (favorites / plays), sharing rate (shares / plays), etc.; the highest number of searches, the highest number of clicks, the highest number of plays, the highest number of favorites, etc.; and the average number of searches, the average number of clicks, the average number of plays, the average number of favorites, etc. Those skilled in the art may set these features based on actual needs.

[0050] Historical interaction features refer to the characteristics of all users' interactions with music resources associated with associated words within a historical time period (referred to as the first historical time period to distinguish it from other historical time periods below). Historical interaction features can be used to understand the long-term popularity and seasonal changes of associated words.

[0051] Among them, the real-time interaction feature refers to the feature of the interactive behavior implemented by all users in the music application on the music resources corresponding to the associative word within the current time period, focusing on immediate user activities. The current time period involved here is obviously shorter than the first historical time period and the second historical time period, and can be any time length that can be understood by those skilled in the art for statistical real-time behavior features, such as within 6 hours, within 2 hours, within 1 hour, etc. For example, the real-time interaction feature can be the number of clicks, searches, favorites, plays, shares, etc. of the user on the associative word within the current time period. The real-time interaction feature reflects the user's recent interest or trend in the music resources corresponding to the associative word.

[0052] A surge interaction feature refers to the interaction behavior of all users of a music app with the music resource corresponding to the associated word within a historical time period (specifically, two adjacent historical time periods, which can be referred to as two second historical time periods to distinguish them from the first historical time period). The statistical indicator for this increase can be the increase in the number of times, the growth rate of the number of times, or the multiple of the number of times the number of times increased. For example, the search volume for a particular associated word in the current time period has increased by 90% compared to the search volume in the previous time period. This surge interaction feature can reflect the popularity of the associated word.

[0053] Among them, the length of the first historical time period is generally greater than that of the second historical time period. For example, the first historical time period is the past three months, and the second historical time period is the past week. The first historical time period can include the second historical time period. It can be understood that according to actual needs, the first historical time period may not include the second historical time period, and the first historical time period is exactly the same as the second historical time period. Technical personnel in this field can set it according to actual needs.

[0054] Specifically, historical interaction features, real-time interaction features, and surge interaction features respectively reflect the characteristics of associated words at different stages and dimensions. By combining these features, the relevance and popularity of each associated word can be more accurately evaluated, so that the associated words can be more accurately ranked and displayed to users.

[0055] Step S130 , sorting each associative word according to its relevance to the search term, its historical interaction characteristics, its real-time interaction characteristics, and its soaring interaction characteristics, to obtain sorting results of multiple associative words.

[0056] In a specific implementation, each associated word can be scored based on its relevance to the search term, its historical interaction features, its real-time interaction features, and its spike interaction features. Weights can then be defined for each feature, and the scores for each feature can be weighted using the weights to obtain a comprehensive score for each associated word. The associated words are then sorted based on the comprehensive scores, with higher scores ranking them higher, thus obtaining the ranking result.

[0057] In one embodiment, a heat response model for sorting can also be pre-trained, and the correlation between each associated word and the search word, the historical interaction features, real-time interaction features and surge interaction features of each associated word are input into the heat response model, and the heat response model is used to sort the associated words to obtain a sorting result.

[0058] Step S140 : displaying each associated word in the music application according to the sorting results of the multiple associated words.

[0059] Specifically, after the ranking results are obtained, the associated words can be displayed in the music application according to the order of the associated words in the ranking results, so that the user can see the most relevant and popular associated words.

[0060] In the above-mentioned search association word display method, after obtaining multiple association words based on the search words entered by the user in the music application, the correlation between each association word and the search word, the historical interaction characteristics, real-time interaction characteristics, and surge interaction characteristics of each association word are further obtained; each association word is sorted according to the correlation between each association word and the search word, the historical interaction characteristics, real-time interaction characteristics, and surge interaction characteristics of each association word; and each association word is displayed according to the sorting results. In this method, the historical interaction characteristics can reflect the long-term popularity changes of the association word, the real-time interaction characteristics can reflect the recent situation of the association word, and the surge interaction characteristics can reflect the popularity of the association word. By combining these characteristics and the correlation characteristics between the association word and the search word to sort the association words, it is possible to simultaneously take into account the correlation and real-time characteristics of the association word, as well as the history and popularity of the association word. This allows the sorting results to respond to hot spots in real time while not only reflecting hot spots, achieving a balance between hot spots and tradition.

[0061] In an exemplary embodiment, in the above-mentioned step S120, the real-time interaction features of each associative word are obtained, including: obtaining in real time the relevant data of the interaction behaviors performed by all users in the music application on the music resources corresponding to each associative word within the current time period; and performing statistical processing on the relevant data of the interaction behaviors obtained in real time for each associative word to obtain the real-time interaction features of each associative word.

[0062] Among them, the relevant data of interactive behavior may include the search terms entered by the user, the associated words of the search terms clicked by the user, and the interactive data after clicking the associated words (such as whether to play, the number of plays, whether to collect, whether to share, the number of shares, etc.).

[0063] It's understandable that the data related to user interaction on music apps is constantly growing and represents streaming data. Therefore, obtaining the real-time interaction features of each associated word can be achieved through streaming computing. This data is statistically analyzed to obtain the interaction features of each associated word as real-time interaction features. Streaming computing treats data as streaming data and processes it in real time. It offers real-time, continuous, and flexible features. It can process and analyze continuously generated data in real time, providing immediate feedback and dynamically adjusting computing resources based on demand.

[0064] Specifically, relevant data on the interactive behaviors of all users in the music application can be collected in real time, and the relevant data on these interactive behaviors can be converted into a format suitable for streaming computing. Then, using the stream processing framework, statistical calculations are performed on the processed relevant data on the interactive behaviors to obtain relevant real-time interactive features.

[0065] In this embodiment, by performing streaming calculations on relevant data of users' real-time interactive behaviors in music applications, real-time interactive features of each associative word are obtained, so that the associative words can be subsequently sorted according to the real-time interactive features of each associative word, so that the sorting results can reflect the user's real-time interest in the associative words.

[0066] In an exemplary embodiment, in the above-mentioned step S120, the soaring interaction characteristics of each associative word are obtained, including: obtaining relevant data on the interactive behaviors performed by all users in the music application on the music resources corresponding to each associative word within the second historical time period; for each associative word, statistically processing the relevant data on the interactive behaviors of the associative word within the second historical time period according to different time granularities to obtain statistical data on the interactive behaviors of the associative word at each time granularity; and comparing multiple interactive behavior statistical data of each associative word at the same time granularity to obtain the soaring interaction characteristics of each associative word.

[0067] The time granularity can be hours, days, weeks, months, and other time units, or it can be selected according to actual business needs, such as 5 hours, 10 days, etc.

[0068] Among them, interactive behavior statistics may include the total value of one or more interactive features, such as interactive behavior statistics may include search volume, click volume, collection volume, playback volume, etc.

[0069] Specifically, data related to the interactions performed by all users of a music application on music resources corresponding to each associated word within a second historical time period can be collected and preprocessed, such as removing invalid data, to ensure data accuracy. The second historical time period can then be divided according to different time granularities. For example, if the second historical time period covers the past week, using a daily granularity, the second historical time period can be divided into seven time windows; if a 12-hour granularity is used, the second historical time period can be divided into 14 time windows. At the same time granularity, the data related to the interactions within each time window is statistically analyzed to obtain multiple interaction behavior statistics for each time granularity. For example, using a daily granularity, the interaction behavior statistics obtained include the daily play count and favorite count of a music resource corresponding to a particular associated word within the past week. The interaction behavior statistics for each interaction behavior within the same time granularity are further compared to identify changing trends in behavioral characteristics and calculate statistical indicators for the number of times different behavioral characteristics are played, such as the play growth rate = (play count in the current time window - play count in the previous time window) / play count in the previous time window. The behavioral feature of the number of growth statistical indicators greater than the threshold is used as the surge interaction feature of the associative word.

[0070] It can be understood that when there are multiple time windows at a certain time granularity, for example, with days as the time granularity, the most recent week (the second historical time period) can be divided into 7 time windows. At this time, the surge interaction characteristics at the time granularity "day" can be determined by comparing the two most recent time windows, such as the growth rate of the behavioral characteristics of the day closest to the statistical time relative to the second most recent day; it can also be determined by the growth rate of the behavioral characteristics of the day closest to the statistical time relative to the farthest day; it can also be determined by the growth rate of the behavioral characteristics of any day closer to the statistical time relative to the farther day, which can be determined according to actual needs.

[0071] In this embodiment, by statistically processing the relevant data of the interactive behavior in the second historical time period according to different time granularities, the interactive behavior statistical data at each time granularity is obtained; by comparing the interactive behavior statistical data of different time windows at each time granularity, the soaring interactive features of each associative word are obtained, so that the associative words can be sorted according to the soaring interactive features of each associative word in the future, so that the sorting results can reflect the popularity of the associative words.

[0072] In an exemplary embodiment, in the above-mentioned step S120, the historical interaction characteristics of each associative word are obtained, including: obtaining relevant data on the interaction behaviors performed by all users in the music application on the music resources corresponding to each associative word within the first historical time period; and performing statistical processing on the relevant data on the interaction behaviors within the first historical time period of each associative word to obtain the historical interaction characteristics of each associative word.

[0073] Specifically, data related to the interaction behaviors of all users in a music application within a first historical time period can be collected. For example, over the past three months, statistics can be collected on the interaction behaviors of each associated word within the first historical time period. For example, the behavioral characteristics of each associated word within the first historical time period can be counted. The statistically obtained behavioral characteristics are used as historical interaction features for subsequent ranking of associated words.

[0074] In this embodiment, relevant data on the interactive behaviors of all users in the music application within a first historical time period are obtained; the relevant data on the interactive behaviors within the first historical time period are statistically processed to identify the interaction patterns and trends of each associative word, so that the associative words can be subsequently sorted according to the historical interaction characteristics of each associative word, so that the sorting results can reflect the historical situation of the associative words.

[0075] In an exemplary embodiment, the step S110 of acquiring a plurality of associative words includes searching a database, and acquiring a plurality of associative words associated with the search word according to the relevance between each entry in the database and the search word.

[0076] In a specific implementation, a number of entries may be pre-stored in a database. When a search term input by a user is received, multiple entries associated with the search term are obtained from the database as multiple associative words.

[0077] More specifically, associated terms related to the search term are obtained from the database. A similarity calculation algorithm (such as cosine similarity) is used to calculate the similarity between the search term and each candidate term in the database. Several terms with high similarity are selected from the candidate terms and ranked, and the most relevant terms are selected as associated terms for the search term. Furthermore, each term in the database can be dynamically updated.

[0078] In this embodiment, by analyzing the correlation between entries in the database and search terms, associative words that are more in line with user intentions can be provided, and by dynamically updating entries in the database, the entries can reflect hot spots in a timely manner, allowing users to find related entries conveniently and quickly.

[0079] In an exemplary embodiment, Figure 2 As shown, each entry in the database is updated in the following way:

[0080] Step S210: regularly obtain relevant data on search interaction behaviors performed by all users in the music application.

[0081] Specifically, regular acquisition can be performed according to a preset cycle or in real time. If it is real-time acquisition, after obtaining the relevant data of the interactive behaviors of all users in the music application, the relevant data of these interactive behaviors are first converted into a format suitable for streaming computing, and then the stream processing framework is used to process the processed relevant data of the interactive behaviors in real time.

[0082] Step S220 , extracting a plurality of to-be-processed terms from the regularly acquired search interaction behavior related data; the to-be-processed terms include all search terms input by the user and associated terms of the clicked search terms.

[0083] Specifically, the acquired data related to search interaction behavior may include the search term entered by the user, the associated terms of the search term clicked by the user, and the interaction data after clicking the associated terms (such as whether the video was played, the number of plays, whether it was saved, whether it was shared, the number of shares, etc.). From this data related to search interaction behavior, the search term entered by the user and the associated terms of the search term clicked by the user are extracted as the terms to be processed.

[0084] Step S230 , determining the quality score of each to-be-processed term, extracting target terms whose quality scores meet the requirements from the multiple to-be-processed terms, and updating the database based on the target terms.

[0085] Specifically, after identifying each pending term, a quality assessment is performed on each pending term based on multiple preset indicators, with corresponding weights assigned to each indicator. After scoring each indicator, the scores are weighted according to their weights to obtain a quality score for each pending term. Based on the quality scores of each pending term, terms with quality scores greater than a threshold are selected from the pending terms and designated as target terms. These target terms are then indexed and stored in the database or replace existing terms in the database to support subsequent recommendation algorithms and user queries.

[0086] In this embodiment, by analyzing the relevant data of the user's search interaction behavior, extracting the terms to be processed, and screening out the target terms whose quality scores meet the conditions to update the database, the update of the database is achieved, which can enable the terms in the database to respond to hot spots in a timely manner, so that when users search, they can display the associative words reflecting the hot spots in a timely manner; and the terms used to update the database are high-quality terms whose quality scores meet the conditions, which can ensure the quality of the terms in the database, improve the quality of the associative words displayed subsequently, and make the associative words more likely to meet user needs.

[0087] In an exemplary embodiment, Figure 3 As shown, step S230 determines the quality score of each to-be-processed term, including:

[0088] Step S310, determining the search intent, term accuracy, and interaction behavior data of each term to be processed; term accuracy indicates the degree of matching between the term to be processed and the terms in the database;

[0089] Step S320: Determine the quality score of each to-be-processed term based on the search intent, term accuracy, and interaction behavior data.

[0090] Search intent refers to the user's true motivation or purpose when conducting a search, such as looking for a specific song or artist, or exploring a specific musical style.

[0091] The accuracy of a term refers to the consistency and correctness between the term to be processed and the terms in the database. It can be the degree of match between the term and the content in the database, including spelling, grammar, and semantics.

[0092] Specifically, after identifying multiple pending terms, natural language processing (NLP) technology can be used to determine the search intent for each pending term, such as whether to find a specific song or artist or explore a specific musical style. The extracted pending terms are matched with existing terms in the database, taking into account spelling errors and synonyms to determine their accuracy. Furthermore, interactive behavior data related to each pending term is extracted from the regularly acquired interactive behavior data, and statistical processing is performed on the interactive behavior data for each pending term to obtain the interactive features of each pending term.

[0093] Furthermore, a quality scoring model can be constructed based on the search intent, term accuracy, and interaction features of each pending term. For example, a quality scoring model can be constructed based on a weighted scoring approach. Different features may contribute to different scores, so different weights can be assigned to search intent, term accuracy, and interaction features. After scoring each of these features, the scores are weighted according to their respective weights to determine the quality score of each pending term. Qualified high-quality terms are updated to the database to support subsequent recommendation algorithms and user queries. Each pending term can have multiple interaction features, such as number of plays and number of favorites. Therefore, when determining its score, a preliminary score can be assigned to each interaction feature, followed by a comprehensive (e.g., weighted) combination of the scores to obtain a total score. This total score is then weighted with the search intent and term accuracy scores to determine the quality score of the pending term.

[0094] In some embodiments, the interaction features of each entry to be processed can also be grouped and aggregated according to multiple predefined time granularities (such as hours, days, weeks, months, etc.), and the interaction behavior data at each time granularity can be statistically analyzed separately to extract the interaction features at multiple time granularities. The same interaction feature at each time granularity (such as daily playback volume, weekly playback volume, monthly playback volume, etc.) can be scored separately to obtain the score value of the interaction feature.

[0095] In this embodiment, by analyzing the search intent, the user's need to search for the term can be grasped more accurately; by checking the accuracy of the term, invalid or erroneous information is reduced; by analyzing the interactive behavior data, the user's interest in the term can be more deeply understood; by evaluating the quality of the term through this multi-dimensional consideration method, the recognition accuracy of the term quality can be improved, the recommendation results of subsequent associated words can be optimized, and the click-through rate can be increased.

[0096] In one embodiment, in order to facilitate those skilled in the art to understand the embodiments of the present application, the following will be described with reference to specific examples in conjunction with the accompanying drawings. Figure 4 , shows a framework diagram of a real-time process of searching for associated words, including the following steps:

[0097] (1) Collect relevant data (Datalog) of user click operation interaction behavior, and perform real-time stream processing through a stream processing framework (such as Flink) to obtain Data A .

[0098]

[0099] (2) Statistics are performed on the stream-processed data at different time granularities to obtain aggregated data at different time granularities (i.e., interactive behavior statistics).

[0100]

[0101] (3) Extract the terms (denoted as Word) from the aggregated data in step (2) A ), and compare the extracted terms with the terms in the database to identify their search intent and term accuracy.

[0102]

[0103] (4) If Figure 5 As shown in Figure 2, based on the aggregated data obtained in step (2), the interactive features (such as CTR, playback volume, etc.) are determined.

[0104]

[0105] (5) The search intent and term accuracy obtained in step (3) and the interactive features obtained in step (4) are input into the term quality discrimination model (WordQualityModel) to decide whether to index and recall the term.

[0106]

[0107] (6) Before presenting the associative words, Figure 6 As shown, for each associative word obtained from the database, its relevance to the search word is calculated, and the aggregated data obtained in step (2) is used to calculate its real-time interaction feature (ontime), soaring interaction feature (soarScore), and historical interaction feature.

[0108]

[0109] (7) The correlation, real-time interaction features, surge interaction features, and historical interaction features obtained in step (6) are input into the heat response model for sorting to obtain the sorting results, and each associated word is displayed according to the sorting results.

[0110]

[0111] Based on cold start normalization, this method designs a normalized cold start scheme for music search terms. This approach achieves the following benefits: 1) Through multi-dimensional statistical analysis of user interaction behavior data, new and popular content can be promptly added to the recall pool and indexed and recalled in real time, shortening the response time of the existing system. 2) It simultaneously considers the relevance of the associated terms to the search terms, as well as the historical, real-time, and popularity of the associated terms. 3) It also promptly updates the database entries, ensuring more accurate associated terms.

[0112] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0113] Based on the same inventive concept, the embodiments of the present application also provide a search-association-word display device for implementing the aforementioned search-association-word display method. The implementation solution provided by this device is similar to the implementation solution described in the aforementioned method. Therefore, the specific limitations in one or more embodiments of the search-association-word display device provided below can be found in the above-mentioned limitations on the search-association-word display method and will not be further elaborated here.

[0114] In one embodiment, Figure 7 As shown, a search association word display device is provided, comprising:

[0115] The associative word acquisition module 710 is used to acquire multiple associative words according to the search word input by the user in the music application;

[0116] Feature acquisition module 720 is used to obtain the relevance between each associative word and the search term, the historical interaction characteristics, real-time interaction characteristics, and surge interaction characteristics of each associative word; the historical interaction characteristics represent the characteristics of the interactive behaviors implemented by all users of the music application on the music resource corresponding to the associative word in a first historical time period; the real-time interaction characteristics represent the characteristics of the interactive behaviors implemented by all users of the music application on the music resource corresponding to the associative word in a current time period; the surge interaction characteristics represent the characteristics of the interactive behaviors implemented by all users of the music application on the music resource corresponding to the associative word in a second historical time period, the number of which the statistical index of the increase is greater than a preset threshold;

[0117] The associative word ranking module 730 is configured to rank each associative word according to its relevance to the search term, its historical interaction characteristics, its real-time interaction characteristics, and its soaring interaction characteristics, to obtain ranking results for the multiple associative words.

[0118] The associative word display module 740 is configured to display each associative word in the music application according to the sorting results of the multiple associative words.

[0119] In one embodiment, the feature acquisition module 720 is also used to obtain in real time relevant data on the interactive behaviors of all users in the music application on the music resources corresponding to each associative word within the current time period; and perform statistical processing on the relevant data on the interactive behaviors obtained in real time for each associative word to obtain the real-time interactive features of each associative word.

[0120] In one embodiment, the feature acquisition module 720 is also used to obtain relevant data on the interactive behaviors of all users in the music application on the music resources corresponding to each associative word within the second historical time period; for each associative word, the relevant data on the interactive behaviors of the associative word within the second historical time period are statistically processed according to different time granularities to obtain statistical data on the interactive behaviors of the associative word at each time granularity; and the statistical data on the interactive behaviors of each associative word at the same time granularity are compared to obtain the soaring interactive features of each associative word.

[0121] In one embodiment, the feature acquisition module 720 is also used to obtain relevant data on the interactive behaviors of all users in the music application on the music resources corresponding to each associative word within the first historical time period; and statistically process the relevant data on the interactive behaviors of each associative word within the first historical time period to obtain the historical interaction features of each associative word.

[0122] In one embodiment, the associative word acquisition module 710 is used to search a database and acquire multiple associative words associated with the search word based on the relevance between each word in the database and the search word.

[0123] In one embodiment, the device also includes a database update module for regularly obtaining relevant data on search interaction behaviors performed by all users in a music application; extracting multiple entries to be processed from the relevant data on search interaction behaviors obtained regularly; the entries to be processed include search terms entered by all users and associative words of search terms clicked; determining the quality score of each entry to be processed, extracting target entries whose quality scores meet the conditions from the multiple entries to be processed, and updating the database based on the target entries.

[0124] In one embodiment, the database update module is used to determine the search intent, term accuracy and interactive behavior data of each term to be processed respectively; the term accuracy represents the matching degree between the term to be processed and the terms in the database; and the quality score of each term to be processed is determined based on the search intent, term accuracy and interactive behavior data.

[0125] Each module in the above-mentioned search association word display device can be implemented in whole or in part through software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in the computer device in the form of software, so that the processor can call and execute the corresponding operations of each module.

[0126] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 8 As shown. The computer device includes a processor, memory, a communication interface, a display screen, and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal via wired or wireless communication. The wireless communication can be achieved via Wi-Fi, a mobile cellular network, NFC (near-field communication), or other technologies. When executed by the processor, the computer program implements a method for displaying search association words. The display screen of the computer device can be a liquid crystal display or an electronic ink display. The input device of the computer device can be a touch layer covering the display screen, or keys, a trackball, or a touchpad provided on the computer device housing, or an external keyboard, touchpad, or mouse.

[0127] Those skilled in the art will understand that Figure 8 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0128] In one embodiment, a computer device is further provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps in the above method embodiments when executing the computer program.

[0129] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0130] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0131] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions.

[0132] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the above-mentioned embodiments. In particular, any reference to memory, database, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processors involved in the various embodiments provided herein may be, but are not limited to, general-purpose processors, central processing units (CPUs), graphics processing units (GPUs), digital signal processors (DSPs), programmable logic devices (PLDs), data processing logic devices based on quantum computing, and the like.

[0133] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0134] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A search association word display method, characterized in that: include: Obtain multiple associated words based on the search words entered by the user in the music application; Obtain the relevance of each of the associative words with the search word, the historical interaction characteristics, the real-time interaction characteristics, and the surge interaction characteristics of each of the associative words; the historical interaction characteristics represent the characteristics of the interaction behaviors implemented by all users in the music application on the music resources corresponding to the associative words in a first historical time period; the real-time interaction characteristics represent the characteristics of the interaction behaviors implemented by all users in the music application on the music resources corresponding to the associative words in a current time period; the surge interaction characteristics represent the characteristics of the interaction behaviors implemented by all users in the music application on the music resources corresponding to the associative words in a second historical time period, the number of which the statistical index of the increase is greater than a preset threshold; sorting the associated words according to the relevance of each associated word to the search term, the historical interaction feature of each associated word, the real-time interaction feature of each associated word, and the soaring interaction feature of each associated word to obtain a sorting result of the multiple associated words; According to the ranking results of the plurality of associated words, each of the associated words is displayed in the music application.

2. The method according to claim 1, characterized in that Acquiring the real-time interaction feature of each of the associated words includes: Real-time acquisition of data related to the interactive behaviors of all users of the music application on the music resources corresponding to the associated words within a current time period; Statistical processing is performed on the relevant data of the interactive behavior acquired in real time for each of the associative words to obtain the real-time interactive features of each of the associative words.

3. The method according to claim 1, characterized in that Obtaining the surge interaction feature of each of the associated words includes: Obtaining data related to interaction behaviors performed by all users of the music application on the music resources corresponding to the associated words within a second historical time period; For each of the associated words, statistical processing is performed on the data related to the interaction behavior of the associated word in the second historical time period according to different time granularities to obtain statistical data on the interaction behavior of the associated word at each time granularity; By comparing the interactive behavior statistics of the associated words at the same time granularity, the surge interaction features of the associated words are obtained.

4. The method according to claim 1, wherein Obtaining the historical interaction features of each of the associated words includes: Obtaining data related to interaction behaviors performed by all users of the music application on the music resources corresponding to the associated words within a first historical time period; Statistical processing is performed on the relevant data of the interaction behavior of each of the associative words in the first historical time period to obtain the historical interaction feature of each of the associative words.

5. The method according to claim 1, wherein The obtaining of multiple associative words includes: A database is searched, and a plurality of associative words associated with the search word are obtained according to the relevance between each entry in the database and the search word.

6. The method according to claim 5, characterized in that The entries in the database are also updated in the following way: Regularly obtain relevant data on search interaction behaviors performed by all users of the music application; Extracting a plurality of to-be-processed terms from the regularly acquired data related to the search interaction behavior; the to-be-processed terms include all the search terms input by the users and the associated terms clicked by the users; The quality score of each of the to-be-processed terms is determined, target terms whose quality scores meet the conditions are extracted from the multiple to-be-processed terms, and the database is updated according to the target terms.

7. The method according to claim 6, characterized in that Determining the quality score of each of the to-be-processed terms includes: Determining the search intent, term accuracy, and interactive behavior data of each of the terms to be processed; the term accuracy represents the degree of matching between the term to be processed and the terms in the database; The quality score of each of the to-be-processed terms is determined according to the search intent, the term accuracy and the interactive behavior data.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the search association word display method according to any one of claims 1 to 7 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the search association word display method according to any one of claims 1 to 7 are implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the search association word display method according to any one of claims 1 to 7 are implemented.

Images