Music retrieval method, computer device and program product

Abstract

The invention relates to a music retrieval method, computer equipment and a program product, and the method comprises the steps: obtaining a search term in response to a search input operation of a user; respectively matching the search word with the first word bank and the second word bank to obtain a first matching result and a second matching result; the first lexicon comprises a plurality of first associational words, the second lexicon comprises a plurality of second associational words, the second associational words are obtained by splitting at least one character of at least one associational word in the first lexicon, and both the first matching result and the second matching result comprise at least one associational word; obtaining an initial associational word before splitting corresponding to each associational word in the second matching result; and outputting a search result based on each associational word in the first matching result and the initial associational word corresponding to each associational word in the second matching result. By adopting the method, the probability that the determined search associational word is accurately matched with the search word input by the user can be improved.

Description

Technical Field

[0001] The present application relates to the field of Internet technology, and in particular to a music retrieval method, apparatus, computer equipment, computer-readable storage medium, and computer program product. Background Art

[0002] Search associated words refer to the words that users want to search for based on the partial phrases they enter into the search box. Traditional methods for recalling search associated words usually determine the recalled associated words by matching the search words with candidate associated words, or determine the recalled associated words based on the relevance of the search words with candidate associated words, the user's profile, etc., and display them as recommendation results.

[0003] However, for users who use handwriting input methods, especially elderly users, when the user inputs slowly, half-written words often appear. At this time, it will be difficult to recall the associated words that the user is satisfied with through traditional recall methods, causing the user to delete and re-enter, affecting search efficiency. Summary of the invention

[0004] Based on this, it is necessary to provide a music retrieval method, apparatus, computer device, computer-readable storage medium and computer program product to address the technical problem that the accuracy of the associated words recalled by the above method in the handwriting input method scenario is low.

[0005] In a first aspect, the present application provides a music retrieval method. The method comprises:

[0006] In response to a user's search input operation, obtaining a search term;

[0007] Matching the search term with a first word library and a second word library respectively to obtain a first matching result and a second matching result; the first word library includes a plurality of first associated words, the second word library includes a plurality of second associated words, the second associated words are obtained by splitting at least one character of at least one associated word in the first word library, and the first matching result and the second matching result both include at least one associated word;

[0008] Obtaining initial associative words before splitting corresponding to each associative word in the second matching result;

[0009] Based on each associative word in the first matching result and an initial associative word corresponding to each associative word in the second matching result, a search result is output.

[0010] In one embodiment, the method further comprises:

[0011] Selecting target associative words from the first word library;

[0012] Splitting the characters in the target associative word that meet the splitting condition to obtain the split target associative word;

[0013] The split target associative words form the second word library;

[0014] A mapping relationship between the target associative words and the split target associative words is established to obtain a mapping set.

[0015] In one embodiment, the step of selecting a target associative word from the first word library includes:

[0016] Obtaining historical search records of each associative word in the first word library;

[0017] Determining the search index value of each of the associated words according to the historical search records;

[0018] The target associative word is selected from the first word library according to the search index value.

[0019] In one embodiment, before splitting the characters in the target associative word that meet the splitting condition to obtain the split target associative word, the method further includes:

[0020] Acquire historical search data of historical users; the historical search data includes historical search terms and associated terms selected by the historical users for the historical search terms;

[0021] According to the historical search data, counting incorrectly input characters and the error frequencies of the incorrectly input characters;

[0022] The characters whose error frequency reaches a preset condition are determined as characters that meet the splitting condition.

[0023] In one embodiment, the step of splitting the characters in the target associative word that meet the splitting condition to obtain the split target associative word includes:

[0024] According to the radical splitting method, characters in the target associative word that meet the splitting condition are split to obtain the split target associative word.

[0025] In one embodiment, obtaining the initial associative words before splitting corresponding to each associative word in the second matching result includes:

[0026] According to each associated word in the second matching result, the mapping set is queried to obtain the initial associated words before splitting corresponding to each associated word.

[0027] In one embodiment, matching the search term with the first word library and the second word library respectively to obtain the first matching result and the second matching result includes:

[0028] Matching the search term with each associative term in the first word library in a prefix matching or non-prefix matching manner to obtain the first matching result;

[0029] The search term is matched with each associated term in the second vocabulary in a prefix matching or non-prefix matching manner to obtain the second matching result.

[0030] In one embodiment, after outputting the search results based on each associative word in the first matching result and the initial associative word corresponding to each associative word in the second matching result, the method further includes:

[0031] sorting the initial associative words corresponding to the associative words in the first matching result and the associative words in the second matching result to obtain a sorting result;

[0032] The initial associated words corresponding to each associated word in the first matching result and each associated word in the second matching result are displayed according to the sorting result.

[0033] In one embodiment, after obtaining the search term in response to the user's search input operation, the method further includes:

[0034] According to the search word, a preset mapping relationship table is queried to obtain a conversion word associated with the search word; the mapping relationship table stores different presentation forms of the same character;

[0035] The search word and the conversion word are matched with the first word library and the second word library to obtain a first matching result and a second matching result.

[0036] In a second aspect, the present application also provides a music retrieval device. The device comprises:

[0037] An acquisition module, used for acquiring a search term in response to a user's search input operation;

[0038] a matching module, configured to match the search term with a first word library and a second word library respectively, to obtain a first matching result and a second matching result; the first word library includes a plurality of first associated words, the second word library includes a plurality of second associated words, the second word library is obtained by splitting at least one character of at least one associated word in the first word library, and the first matching result and the second matching result both include at least one associated word;

[0039] A mapping module, used for obtaining the initial associative words before splitting corresponding to each associative word in the second matching result;

[0040] The determination module is used to output search results based on each associative word in the first matching result and an initial associative word corresponding to each associative word in the second matching result.

[0041] In a third aspect, the present application further provides a computer device. The computer device includes a memory and a processor, the memory stores a computer program, and the processor implements the following steps when executing the computer program:

[0042] In response to a user's search input operation, obtaining a search term;

[0043] Matching the search term with a first word library and a second word library respectively to obtain a first matching result and a second matching result; the first word library includes a plurality of first associated words, the second word library includes a plurality of second associated words, the second word library is obtained by splitting at least one character of at least one associated word in the first word library, and the first matching result and the second matching result both include at least one associated word;

[0044] Obtaining initial associative words before splitting corresponding to each associative word in the second matching result;

[0045] Based on each associative word in the first matching result and an initial associative word corresponding to each associative word in the second matching result, a search result is output.

[0046] In a fourth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0047] In response to a user's search input operation, obtaining a search term;

[0048] Matching the search term with a first word library and a second word library respectively to obtain a first matching result and a second matching result; the first word library includes a plurality of first associated words, the second word library includes a plurality of second associated words, the second word library is obtained by splitting at least one character of at least one associated word in the first word library, and the first matching result and the second matching result both include at least one associated word;

[0049] Obtaining initial associative words before splitting corresponding to each associative word in the second matching result;

[0050] Based on each associative word in the first matching result and an initial associative word corresponding to each associative word in the second matching result, a search result is output.

[0051] In a fifth aspect, the present application further provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0052] In response to a user's search input operation, obtaining a search term;

[0053] Matching the search term with a first word library and a second word library respectively to obtain a first matching result and a second matching result; the first word library includes a plurality of first associated words, the second word library includes a plurality of second associated words, the second word library is obtained by splitting at least one character of at least one associated word in the first word library, and the first matching result and the second matching result both include at least one associated word;

[0054] Obtaining initial associative words before splitting corresponding to each associative word in the second matching result;

[0055] Based on each associative word in the first matching result and an initial associative word corresponding to each associative word in the second matching result, a search result is output.

[0056] The above-mentioned music retrieval method, device, computer equipment, storage medium and computer program product obtain a second vocabulary by splitting at least one character of at least one associative word in a traditional first vocabulary. When determining the search associative word, on the basis of matching the search word with the first vocabulary to obtain a first matching result, it is also matched with the second vocabulary with character split to obtain a second matching result, thereby achieving accurate matching when the search word is only half of a character, and further obtaining the initial associative words before splitting corresponding to each associative word in the second matching result, so as to predict the correct search word that the user wants to input, and finally using each associative word in the first matching result and the initial associative words corresponding to each associative word in the second matching result as the search result for the search word input by the user, thereby achieving the determination of associative words from two dimensions of unsplit characters and split characters, thereby improving the probability of accurate matching between the determined search associative words and the search word input by the user, improving the user's search efficiency, and solving the problem of low accuracy of search associative words returned by traditional methods when handwriting input is slow or input is wrong. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 is a schematic diagram of a flow chart of a music retrieval method in one embodiment;

[0058] Figure 2 is a flowchart of a process of generating a first vocabulary, a second vocabulary and a mapping set in one embodiment;

[0059] Figure 3A schematic diagram of a process of online matching and determining search association words in one embodiment;

[0060] Figure 4 A schematic diagram of displaying results of searching for associative words in one embodiment;

[0061] Figure 5 is a structural block diagram of a music retrieval device in one embodiment;

[0062] Figure 6 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0063] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0064] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein.

[0065] In one embodiment, Figure 1 As shown, a music retrieval method is provided. This embodiment uses the method applied 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, laptops, smart phones, tablets, Internet of Things devices and portable wearable devices. The Internet of Things devices can be smart speakers, smart TVs, smart air conditioners, smart car-mounted devices, etc. Portable wearable devices can be smart watches, smart bracelets, head-mounted devices, etc. The server can be implemented as an independent server or a server cluster composed of multiple servers. In this embodiment, the method includes the following steps:

[0066] Step S110, obtaining a search term in response to a search input operation by the user.

[0067] The search term may be understood as a keyword entered by a user in a search box of an application (such as a music application). For example, the search term may be a singer's name, a song title, or a partial word or character in an album title.

[0068] Specifically, the user performs an input operation in the search box of an application with a search function such as an application or a browser. For example, when the user inputs "bow", the terminal can respond to this input operation and obtain that the search term input by the user is "bow".

[0069] In one implementation, after obtaining the search term input by the user, it can be further preprocessed. For example, the search term can be standardized in dimensions such as case and simplified and traditional Chinese characters, and then the associated words of the preprocessed search term can be determined.

[0070] Step S120: Match the search term with the first thesaurus and the second thesaurus respectively to obtain a first matching result and a second matching result. The first thesaurus includes multiple first associated words, and the second thesaurus includes multiple second associated words. The second thesaurus is obtained by splitting at least one character of at least one associated word in the first thesaurus. Both the first matching result and the second matching result include at least one associated word.

[0071] Among them, an associated word refers to a word entry that may be associated with the search term input by the user. For example, when the user inputs "king" in the search box, the associated word obtained can be "princess consort", and when the user inputs "country", the associated word can be "country", etc. That is, the associated word can be the complete expression of the search term.

[0072] In specific implementation, considering that in the case of a handwriting input method, etc., due to slow input writing, the search term input by the user may only be a part of the Chinese character that the user wants to input. For example, when the user wants to input "sunny", but only inputs "day", therefore, after obtaining the normal first thesaurus in this solution, at least one character of at least one associated word in the first thesaurus is further split to split one character into multiple parts, and thus the second thesaurus composed of the split associated words is obtained. For example, for the associated word "love song" in the first thesaurus, it can be split into "heart green brother owe", and similar splitting operations are performed on other associated words in the first thesaurus, and the thesaurus composed of each split associated word is used as the second thesaurus.

[0073] After obtaining the search term input by the user, the search term is matched with the normal first thesaurus and the split second thesaurus respectively, and the associated word in the first thesaurus that matches the search term can be recorded as the first matching result, and the associated word in the second thesaurus that matches the search term can be recorded as the second matching result. Among them, the number of associated words included in the first matching result and the second matching result can be one or multiple.

[0074] It should be noted that each associated word in the second thesaurus may contain multiple characters. For example, "sunny day" has two characters. In the case where an associated word has multiple characters, not every character in the associated word necessarily has to be split. At least one character being split is sufficient. For example, the character "晴" in the associated word "sunny day" can be split into "日青", while the character "天" remains unsplit.

[0075] Step S130: Obtain the initial associated words before splitting corresponding to each associated word in the second matching result.

[0076] In specific implementation, the second matching result is an associated word with at least one character split, and it also corresponds to the initial associated word before splitting. For example, the second matching result is "日青天", and the corresponding initial associated word is "sunny day". Since when the search term is part of a single character, the user doesn't actually intend to input that search term, but rather it's due to slow input operations. Therefore, after determining the split second matching result that matches the user-entered search term, it is necessary to restore each associated word in the second matching result to its unsplit initial state as the initial associated word and then display it to the user.

[0077] Step S140: Output a search result based on the associated words in the first matching result and the initial associated words corresponding to the associated words in the second matching result.

[0078] In specific implementation, after obtaining the first matching result obtained through normal matching and the initial associated words obtained through character split matching, the associated words obtained by these two methods can be displayed to the user as the search result for the user-entered search term.

[0079] In the above music retrieval method, the second thesaurus is obtained by splitting at least one character of at least one associated word in the traditional first thesaurus. When determining the search associated word, on the basis of matching the search term with the first thesaurus to obtain the first matching result, it is simultaneously matched with the second thesaurus with characters split to obtain the second matching result. Thus, accurate matching can be achieved in the case where the search term is only half of a character. Further, obtain the initial associated words before splitting corresponding to each associated word in the second matching result to predict the correct search term that the user wants to input. Finally, use the associated words in the first matching result and the initial associated words corresponding to the associated words in the second matching result as the search result for the user-entered search term, realizing the determination of associated words from two dimensions of unsplit characters and split characters. Thereby, the probability of accurate matching between the determined search associated word and the user-entered search term can be increased, improving the user's search efficiency, and solving the problem that the accuracy of the search associated words returned by the traditional method is relatively low when inputting by hand due to slow input or input errors.

[0080] In an exemplary embodiment, the method further includes: screening target associated words from a first thesaurus; splitting characters in the target associated words that meet the splitting conditions to obtain the split target associated words, and forming a second thesaurus with the split target associated words; establishing a mapping relationship between the target associated words and the split target associated words to obtain a mapping set.

[0081] In specific implementation, considering that splitting the associated words and then establishing the mapping will additionally occupy space, if each associated word in the first thesaurus is split, the data volume will increase sharply. Therefore, when splitting the associated words in the first thesaurus, screening can be performed according to preset screening conditions, and some of the associated words in the first thesaurus can be selected as the target associated words for splitting.

[0082] Further, when splitting the selected target associated words, they can be split into multiple parts according to the composition of the characters. However, the composition of some characters is relatively simple, and the composition of some characters is relatively complex. If each character of the associated word is split, the data volume will increase sharply further. Therefore, for any target associated word, it is not necessary to split each character of the target associated word, but characters that meet the splitting conditions can be selected for splitting, and the split characters and the unsplit characters in the target associated word are still sorted in the original character order of the target associated word to obtain the split target associated word. For example, if the target associated word is "sunny day", splitting the character "sunny" in it gives "day blue", and the character "day" does not need to be split. Thus, the split target associated word can be expressed as "day blue day". Performing similar splitting operations for each target associated word can obtain the split target associated words corresponding to each target associated word. These split target associated words are formed into a second thesaurus, and a mapping relationship can be established between each target associated word and the corresponding split target associated word to obtain a mapping set, so that in practical applications, the initial state of the split target associated word can be restored according to the mapping set.

[0083] In one implementation manner, the splitting of a character can be further refined. For example, splitting the character "emotion" into "heart radical, six, month". The order of the parts after splitting the character can also be changed. For example, splitting the character "emotion" into "blue, heart radical".

[0084] In this embodiment, by screening out the target associative words from the first vocabulary, the data volume of the associative words that need to be split can be reduced, and the problem of a sudden increase in data volume occupying too much space can be avoided. By splitting the characters in the target associative words that meet the splitting conditions without splitting each character, the data volume of the obtained second vocabulary can be further reduced. By establishing a mapping relationship between the target associative words and the split target associative words, a mapping set is obtained, so that the mapping set can be directly queried later to determine the initial associative words corresponding to each associative word in the matched second matching result, thereby ensuring restoration efficiency and accuracy.

[0085] In an exemplary embodiment, the above step of filtering out the target associative words from the first vocabulary specifically includes: obtaining historical search records of each associative word in the first vocabulary; determining the search index value of each associative word based on the historical search records; and filtering out the target associative words from the first vocabulary based on the search index value.

[0086] The search index value may be understood as an index that characterizes the search status of the associated word. For example, the search index value may be search popularity, which may indicate the frequency and number of times the associated word is searched, reflecting the popularity of the associated word.

[0087] In a specific implementation, the target associative words can be selected from the first word library according to the search index value of each associative word by obtaining the historical search records of each associative word in the first word library. According to the historical search records, the search index value of each associative word is determined, and the associative words whose search index value is greater than a preset threshold are selected from the first word library, or a certain proportion of associative words with higher search index values ​​are selected as the target associative words.

[0088] More specifically, the historical search records may include records of search times, timestamps, interaction data and other indicators. According to the historical search records, the search index value of each associated word can be determined by at least one of the search times, growth rate, user click rate and other indicators. For example, the total number of searches for each associated word can be directly counted, and the search index value can be determined according to the relationship that the number of searches is proportional to the popularity, that is, the more searches, the higher the popularity. The number of searches can also be weighted by the timestamp to obtain the search index value, such as giving a higher weight to the number of searches with a closer time, that is, determining the weight according to the relationship that the time is proportional to the weight, so as to weight the number of searches. The search index value can also be determined by weighting the index values ​​of each indicator, etc.

[0089] In this embodiment, based on the historical search records of each associated word in the first thesaurus, the search index value of each associated word is determined, and then the target associated words are screened out from the first thesaurus according to the search index value, so as to further perform character splitting, so that the associated words in the second thesaurus obtained after splitting are associated words with higher popularity, thereby improving the success rate of matching between the second thesaurus and the search term input by the user.

[0090] In one exemplary embodiment, before splitting the characters that meet the splitting conditions in the target associated word to obtain the split target associated word, it further includes: obtaining the historical search data of historical users; the historical search data includes historical search terms and the associated words selected by historical users for the historical search terms; according to the historical search data, counting the input error characters and the error frequencies of the input error characters; determining the characters with error frequencies reaching the preset conditions as the characters that meet the splitting conditions.

[0091] In specific implementation, the composition of each character is different, and the probability of input error during the input operation is also different. The composition of some characters is relatively simple, and during the input operation, generally, the situation of only inputting a part will not occur. For example, the character "天" is generally not miswritten, so it does not need to be split into "一大". Therefore, whether each character in the associated word needs to be split can be determined by counting the historical search data in the terminal, and the characters that are often input incorrectly are used as the characters that meet the splitting conditions.

[0092] More specifically, first obtain the historical search terms input by historical users during the search operation, and the associated words selected for the historical search terms. If the selected associated word is inconsistent with the historical search term, it indicates that the historical search term is input incorrectly. For example, if the input historical search term is "亻" and the selected associated word is "信仰", it indicates that the actual search term they want to search for is "信" or "仰", then these two characters are input error characters. Count several characters in a similar situation and the error frequencies of each character according to the historical search data, and determine the characters with error frequencies reaching the predetermined threshold as the characters that meet the splitting conditions.

[0093] In this embodiment, by obtaining the historical search data of historical users, counting the input error characters and the error frequencies of the input error characters according to the historical search data, and determining the characters with error frequencies reaching the preset conditions as the characters that meet the splitting conditions, the characters split in the second thesaurus are characters that are prone to input errors, thereby increasing the probability of successful matching with the search term input by the user and recommending accurate associated words for the user.

[0094] In an exemplary embodiment, splitting the characters in the target associated word that meet the splitting conditions to obtain the split target associated word includes: splitting the characters in the target associated word that meet the splitting conditions in the way of splitting by radicals to obtain the split target associated word.

[0095] In specific implementation, when determining the characters that meet the splitting conditions and splitting them, the radicals of the characters can be recognized first, and the characters are split into radicals and the remaining parts according to the radicals. For example, the radical of the character "汉" is "氵", and the remaining part is "又". After completion of splitting, all the split parts are combined to form the split target associated word. For example, the split target associated word corresponding to the character "汉" is "氵又".

[0096] In this embodiment, by splitting the characters in the target associated word that meet the splitting conditions in the way of splitting by radicals, each part of the split characters conforms to the user's writing habit, thereby increasing the probability of successful matching between the split target associated word and the search term input by the user.

[0097] In an exemplary embodiment, the step of obtaining the initial associated words before splitting corresponding to each associated word in the second matching result includes: querying the mapping set according to each associated word in the second matching result to obtain the initial associated words before splitting corresponding to each associated word in the second matching result.

[0098] In specific implementation, after obtaining the second matching result that matches the search term input by the user, the mapping set constructed when building the second thesaurus can be directly queried according to each associated word in the second matching result. This mapping set stores the mapping relationship between the split associated words and the associated words before splitting. Therefore, the initial associated words before splitting corresponding to each associated word in the second matching result can be obtained through querying.

[0099] In this embodiment, determining the initial associated words before splitting corresponding to each associated word in the second matching result by querying the mapping set can ensure the query efficiency and the accuracy of restoring the state before splitting of the second matching result.

[0100] In an exemplary embodiment, the above step S120 respectively matches the search term with the first thesaurus and the second thesaurus to obtain the first matching result and the second matching result, including: matching the search term with each associated word in the first thesaurus in the way of prefix matching or non-prefix matching to obtain the first matching result; matching the search term with each associated word in the second thesaurus in the way of prefix matching or non-prefix matching to obtain the second matching result.

[0101] Prefix matching means that the search term must be the beginning of the associated term being queried. Non-prefix matching allows the search term to appear at any position in the associated term being queried.

[0102] In a specific implementation, when the search term is matched with the first word library in a prefix matching manner, each associative word in the first word library is traversed, and for each associative word, it is checked whether it starts with the search term. If so, it is determined to belong to the first matching result, otherwise, it does not belong, thereby obtaining the matched associative words, which constitute the first matching result. When the search term is matched with the first word library in a non-prefix matching manner, each associative word in the first word library is traversed, and for each associative word, it is checked whether it contains the search term. If so, it is determined to belong to the first matching result, otherwise, it does not belong, thereby obtaining the matched associative words, which constitute the first matching result.

[0103] Similarly, when the search term is matched with the second word library in a prefix matching manner, each associative word in the second word library is traversed, and for each associative word, it is checked whether it begins with the search term. If so, it is determined to belong to the second matching result, otherwise, it does not belong, thereby obtaining the matched associative words, which constitute the second matching result. When the search term is matched with the second word library in a non-prefix matching manner, each associative word in the second word library is traversed, and for each associative word, it is checked whether it contains the search term. If so, it is determined to belong to the second matching result, otherwise, it does not belong, thereby obtaining the matched associative words, which constitute the second matching result.

[0104] In this embodiment, through prefix matching or non-prefix matching, it is possible to accurately identify associated words related to the search word input by the user, which can help the user quickly find the desired information, thereby improving the accuracy and relevance of the search.

[0105] In an exemplary embodiment, after outputting the search results based on the initial associative words corresponding to the associative words in the first matching result and the associative words in the second matching result, the above step S140 also includes: sorting the initial associative words corresponding to the associative words in the first matching result and the associative words in the second matching result to obtain sorted results; and displaying the initial associative words corresponding to the associative words in the first matching result and the second matching result according to the sorted results.

[0106] It is understandable that since the associative words in the second vocabulary are obtained through character splitting and are constructed words, they usually have no or very little search volume or popularity. Therefore, when sorting, the popularity of the initial associative words before splitting mapped to each associative word in the second matching result is used for sorting.

[0107] In specific implementation, sorting the associated words in the first matching result and the initial associated words corresponding to the associated words in the second matching result can be performed by one or more of the interaction metrics such as the search popularity or click-through rate, play rate, completion rate, and play volume of both. When using multiple metrics for sorting, different importance weights can be assigned to each metric, and the weighted sum method can be used to calculate the comprehensive score and sort according to the comprehensive score. Alternatively, a sorting model can be pre-trained, the metric values of each metric are input into the sorting model to obtain a score, and sorting is performed according to this score.

[0108] For example, assume the search term is "day", the associated word matched in the first matching result is "Sunset Boulevard", the associated word matched in the second matching result is "Sunny Sky", and the corresponding initial associated word is "sunny day". Then, the popularity of "sunny day" and the popularity of "Sunset Boulevard" are used for sorting. If it is determined that the popularity of "sunny day" is higher than that of "Sunset Boulevard", the final sorting result is: R["sunny day", "Sunset Boulevard"], and these two associated words are displayed according to this sorting result.

[0109] In this embodiment, by sorting the associated words in the first matching result and the initial associated words corresponding to the associated words in the second matching result, and displaying the associated words in the first matching result and the initial associated words corresponding to the associated words in the second matching result according to the sorting result, more relevant associated words can be ranked in the front, enabling users to find the options they are interested in faster, reducing information overload and screening time, improving the search efficiency and search experience, and enhancing the quality and display effect of the search association results.

[0110] In an exemplary embodiment, after the above step S110 obtains the search term in response to the user's search input operation, it further includes: querying a preset mapping relationship table according to the search term to obtain a conversion word associated with the search term; the mapping relationship table stores different presentation forms of the same character; both the search term and the conversion word are matched with the first word library and the second word library to obtain the first matching result and the second matching result.

[0111] Specifically, considering that the same character may have different presentation forms. For example, the character "亻" may be written as "人". Therefore, a mapping relationship table of such characters with different presentation forms can be pre-constructed. After obtaining the search term input by the user, the mapping relationship table can be queried to obtain the conversion word associated with the search term. When obtaining the matching results by matching with the first word library and the second word library, both the search term and the conversion word can be matched with the first word library and the second word library to increase the possibility that the matching associated words match the user's search intention.

[0112] In this embodiment, by using a mapping relation table, the user's search can not only match the originally input vocabulary, but also find its related conversion words, so that even if the user's input is not completely correct or the word usage is inappropriate, corresponding results can still be provided through the mapping relation, thereby improving the accuracy of the search.

[0113] In one embodiment, for the convenience of those skilled in the art to understand the embodiments of the present application, specific examples in conjunction with the accompanying drawings will be described below. Refer to Figure 2 , which shows a schematic flowchart of the generation process of the first thesaurus, the second thesaurus, and the mapping set, including the following steps:

[0114] Step S210, generate a normal first thesaurus, denoted as Da. Specifically, the first thesaurus can be constructed according to the historical search data in the application. The first thesaurus may include a series of entries such as the names of singers, lyrics, song names, album names, etc. searched.

[0115] Step S220, determine the search popularity of each associated word according to the historical search records of the associated words in the first thesaurus; screen out the target associated words from the first thesaurus according to the search popularity.

[0116] Specifically, let the data set composed of the target associated words be Db.

[0117] Step S230, perform a splitting process on the characters in the target associated words that meet the splitting conditions to obtain the split target associated words, and form a second thesaurus.

[0118] Specifically, at least one character of each target associated word in the data set Db is split to obtain a second thesaurus Dc. Among them, whether each character is split can determine whether it is a character that meets the splitting conditions. The determination process of the character that meets the splitting conditions is as follows: obtain the historical search data of historical users; the historical search data includes historical search terms and the associated words selected by historical users for the historical search terms; according to the historical search data, count the input error characters and the error frequency of the input error characters; determine the characters with the error frequency reaching the preset conditions as the characters that meet the splitting conditions.

[0119] Step S240, establish a mapping relationship between the target associated words and the split target associated words to obtain a mapping set.

[0120] Specifically, establish a mapping relationship between the second thesaurus Dc and the data set Db to obtain a mapping set M. For example: data set Db["love song"], data set Dc["忄青哥欠"], mapping set M[("忄青哥欠","love song")].

[0121] In this method, by screening out target associated words from the first thesaurus, the amount of data of the associated words that need to be split can be reduced, avoiding the problem that the data volume explodes and occupies too much space. By splitting the characters that meet the splitting conditions in the target associated words, instead of splitting each character, the data volume of the obtained second thesaurus can be further reduced. By establishing a mapping relationship between the target associated words and the split target associated words, a mapping set is obtained, so that the initial associated words corresponding to each associated word in the second matching result can be directly queried later, ensuring the restoration efficiency and accuracy. By using the historical search records of each associated word in the first thesaurus, the search popularity of each associated word is determined, and then the target associated words are screened out from the first thesaurus according to the search popularity for further character splitting, so that the associated words in the second thesaurus obtained after splitting are associated words with higher popularity, thereby improving the success rate of matching the second thesaurus with the search term input by the user. By obtaining the historical search data of historical users, according to the historical search data, the misinput characters and the error frequencies of the misinput characters are counted, and the characters whose error frequencies reach the preset conditions are determined as the characters that meet the splitting conditions, so that the characters split in the second thesaurus are characters that are easy to misinput, thereby increasing the probability of successful matching with the search term input by the user and recommending accurate associated words for the user.

[0122] Reference Figure 3 , shows a schematic flowchart of online matching to determine search associated words, including the following steps:

[0123] Step S310, in response to the user's search input operation, obtain the search term.

[0124] Specifically, after obtaining the search term, preprocessing can be performed to obtain the preprocessed search term.

[0125] Step S320, match the search term with each associated word in the first thesaurus to obtain a first matching result. Among them, the matching method is prefix matching or non-prefix matching.

[0126] Step S330, match the search term with each associated word in the second thesaurus to obtain a second matching result. Among them, the matching method is prefix matching or non-prefix matching.

[0127] For example, if the search term input by the user is "日", the "Sunset Boulevard" can be matched from the first thesaurus Da, that is, the first matching result Ra, and the "日青天" can be matched from the second thesaurus Dc, that is, the second matching result Rc.

[0128] Step S340, query the mapping set to obtain the initial associated words before splitting corresponding to the second matching result.

[0129] Specifically, if the second matching result Rc is not empty, the original word corresponding to the second matching result Rc is obtained through the mapping set M. For example, the initial associated word corresponding to "日青天" is "晴天".

[0130] Step S350: After mixing and sorting each associated word in the first matching result with the initial associated words corresponding to each associated word in the second matching result, display each associated word in the first matching result and the initial associated words corresponding to each associated word in the second matching result according to the sorting result.

[0131] Reference Figure 4 , is a schematic diagram of the display result of the search associated words provided for an embodiment. As Figure 4 shown, the input search term is "亻", and the search associated words are sorted in order as lullaby, Chinese Paladin, side face, belief.

[0132] In this method, the second word library is obtained by splitting at least one character of at least one associated word in the traditional first word library. When determining the search associated words, on the basis of obtaining the first matching result by matching the search term with the first word library, it is simultaneously matched with the second word library after character splitting to obtain the second matching result. Thus, accurate matching can be achieved when the search term is only half of a character. Further, the initial associated words before splitting corresponding to each associated word in the second matching result are obtained to realize the prediction of the correct search term that the user wants to input. Finally, the first matching result and the initial associated words are used as the search associated results for the search term input by the user, realizing the determination of associated words from two dimensions of unsplit characters and split characters. Thereby, the probability of accurate matching between the determined search associated words and the search term input by the user can be increased, and the search efficiency of the user can be improved; finally, by sorting each associated word and the initial associated words in the first matching result and displaying each associated word and the initial associated words in the first matching result according to the sorting result, more relevant associated words can be ranked in the front, enabling the user to find the options of interest faster, reducing information overload and screening time, improving the search efficiency and search experience, and enhancing the quality and display effect of the search associated results.

[0133] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed 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 part of the steps or stages in other steps.

[0134] Based on the same inventive concept, the embodiment of the present application also provides a music retrieval device for implementing the music retrieval method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more music retrieval device embodiments provided below can refer to the limitations on the music retrieval method above, and will not be repeated here.

[0135] In one embodiment, Figure 5 As shown, a music retrieval device is provided, including: an acquisition module 510, a matching module 520, a mapping module 530 and a determination module 540, wherein:

[0136] The acquisition module 510 is used to acquire a search term in response to a search input operation by a user;

[0137] The matching module 520 is used to match the search word with the first word library and the second word library respectively to obtain a first matching result and a second matching result; the first word library includes a plurality of first associated words, the second word library includes a plurality of second associated words, the second word library is obtained by splitting at least one character of at least one associated word in the first word library, and the first matching result and the second matching result both include at least one associated word;

[0138] A mapping module 530 is used to obtain the initial associative words before splitting corresponding to each associative word in the second matching result;

[0139] The determination module 540 is used to output search results based on each associative word in the first matching result and the initial associative word corresponding to each associative word in the second matching result.

[0140] In one of the embodiments, the device also includes a construction module, which is used to filter out target associative words from a first vocabulary; split the characters in the target associative words that meet the splitting conditions to obtain the split target associative words, and form the split target associative words into a second vocabulary; establish a mapping relationship between the target associative words and the split target associative words to obtain a mapping set.

[0141] In one embodiment, the construction module is further used to obtain historical search records of each associative word in the first vocabulary; determine the search index value of each associative word according to the historical search records; and select the target associative word from the first vocabulary according to the search index value.

[0142] In one of the embodiments, the construction module is also used to obtain historical search data of historical users; the historical search data includes historical search terms and associative words selected by historical users for the historical search terms; based on the historical search data, the incorrectly input characters and the error frequency of the incorrectly input characters are counted; and the characters whose error frequency reaches a preset condition are determined as characters that meet the splitting condition.

[0143] In one of the embodiments, the construction module is further used to split the characters in the target associative word that meet the splitting conditions in a radical splitting manner to obtain the split target associative word.

[0144] In one embodiment, the mapping module 530 is further configured to query a mapping set according to each associative word in the second matching result to obtain an initial associative word before splitting corresponding to each associative word.

[0145] In one embodiment, the matching module 520 is further used to match the search term with each associative term in the first vocabulary library in a prefix matching or non-prefix matching manner to obtain a first matching result; and to match the search term with each associative term in the second vocabulary library in a prefix matching or non-prefix matching manner to obtain a second matching result.

[0146] In one of the embodiments, the device also includes a display module, which is used to sort the initial association words corresponding to each associative word in the first matching result and each associative word in the second matching result to obtain a sorting result; and display the initial association words corresponding to each associative word in the first matching result and each associative word in the second matching result according to the sorting result.

[0147] In one embodiment, the matching module 520 is further used to query a preset mapping relationship table based on the search term to obtain a conversion word associated with the search term; the mapping relationship table stores different presentation forms of the same character; the search term and the conversion word are matched with the first vocabulary and the second vocabulary to obtain a first matching result and a second matching result.

[0148] Each module in the above-mentioned music retrieval device can be implemented in whole or in part by 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 a computer device in the form of software, so that the processor can call and execute the operations corresponding to each module.

[0149] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 6 As shown. The computer device includes a processor, a memory and a network interface connected through a system bus. Among them, 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 an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data in the music retrieval process. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a music retrieval method is implemented.

[0150] Those skilled in the art will understand that Figure 6 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 those shown in the figure, or combine certain components, or have a different arrangement of components.

[0151] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0152] 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.

[0153] 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.

[0154] 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 relevant laws, regulations and standards of relevant countries and regions.

[0155] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and 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 embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium 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), magnetoresistive 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. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0156] The technical features of the above embodiments may 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.

[0157] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A music retrieval method, characterized in that: The method comprises: In response to a user's search input operation, obtaining a search term; Matching the search term with a first word library and a second word library respectively to obtain a first matching result and a second matching result; the first word library includes a plurality of first associated words, the second word library includes a plurality of second associated words, the second associated words are obtained by splitting at least one character of at least one associated word in the first word library, and the first matching result and the second matching result both include at least one associated word; Obtaining initial associative words before splitting corresponding to each associative word in the second matching result; Based on each associative word in the first matching result and an initial associative word corresponding to each associative word in the second matching result, a search result is output.

2. The method according to claim 1, characterized in that The method further comprises: Selecting target associative words from the first word library; Splitting the characters in the target associative word that meet the splitting condition to obtain the split target associative word; The split target associative words form the second word library; A mapping relationship between the target associative words and the split target associative words is established to obtain a mapping set.

3. The method according to claim 2, characterized in that The step of selecting a target associative word from the first word library includes: Obtaining historical search records of each associative word in the first word library; Determining the search index value of each of the associated words according to the historical search records; The target associative word is selected from the first word library according to the search index value.

4. The method according to claim 2, characterized in that: Before splitting the characters in the target associative word that meet the splitting condition to obtain the split target associative word, the method further includes: Acquire historical search data of historical users; the historical search data includes historical search terms and associated terms selected by the historical users for the historical search terms; According to the historical search data, counting incorrectly input characters and the error frequencies of the incorrectly input characters; The characters whose error frequency reaches a preset condition are determined as characters that meet the splitting condition.

5. The method according to claim 2, characterized in that: The step of splitting the characters in the target associative word that meet the splitting condition to obtain the split target associative word includes: According to the radical splitting method, characters in the target associative word that meet the splitting condition are split to obtain the split target associative word.

6. The method according to claim 2, characterized in that The step of obtaining the initial associative words before splitting corresponding to each associative word in the second matching result includes: According to each associated word in the second matching result, the mapping set is queried to obtain the initial associated words before splitting corresponding to each associated word.

7. The method according to claim 1, characterized in that The step of matching the search term with the first word library and the second word library respectively to obtain a first matching result and a second matching result includes: Matching the search term with each associative term in the first word library in a prefix matching or non-prefix matching manner to obtain the first matching result; The search term is matched with each associated term in the second vocabulary in a prefix matching or non-prefix matching manner to obtain the second matching result.

8. The method according to claim 1, characterized in that: After outputting the search results based on the initial associative words corresponding to the associative words in the first matching results and the associative words in the second matching results, the method further includes: sorting the initial associative words corresponding to the associative words in the first matching result and the associative words in the second matching result to obtain a sorting result; The initial associated words corresponding to each associated word in the first matching result and each associated word in the second matching result are displayed according to the sorting result.

9. The method according to claim 1, characterized in that: After obtaining the search term in response to the user's search input operation, the method further includes: According to the search word, a preset mapping relationship table is queried to obtain a conversion word associated with the search word; the mapping relationship table stores different presentation forms of the same character; The search word and the conversion word are matched with the first word library and the second word library to obtain a first matching result and a second matching result.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the music retrieval method described in any one of claims 1 to 9 are implemented.

11. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the music retrieval method described in any one of claims 1 to 9 are implemented.

Images