Similar word determination method and device, electronic device, and storage medium

Abstract

The present invention provides a similar word determination method and device, electronic device, and storage medium, and relates to the field of computer technology. The similar word determination method includes: obtaining historical user behavior data; performing identification processing on the historical user behavior data to obtain search-related data corresponding to the historical user behavior data; and screening the search-related data to determine similar words corresponding to the historical user behavior data. The technical solution of the embodiment of the present invention can not only improve the accuracy of similar words, but also make similar words more complete, thereby improving the user experience.

Description

Background Art

[0002] With the rapid development of the Internet industry and the widespread application of e-commerce technology, the variety of items on e-commerce websites has become richer and their quantity has grown rapidly.

[0003] When users purchase items on e-commerce websites, they like to use the search function to find items. However, since the search terms used by users cannot be standardized, the accuracy of the search term matching results is low, resulting in users being unable to search for the items they need, which reduces the user experience.

[0004] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0005] The purpose of the embodiments of the present invention is to provide a similar word determination method, a similar word determination device, an electronic device and a computer-readable storage medium, thereby at least to a certain extent overcoming the problems of low accuracy of matching results of existing search services and incomplete types of matched items.

[0006] Other features and advantages of the present invention will become apparent from the following detailed description, or may be learned in part by practice of the present invention.

[0007] According to a first aspect of an embodiment of the present invention, a similar word determination method is provided, comprising: obtaining historical user behavior data; performing identification processing on the historical user behavior data to obtain search association data corresponding to the historical user behavior data; and performing screening processing on the search association data to determine similar words corresponding to the historical user behavior data.

[0008] In some example embodiments of the present invention, based on the aforementioned scheme, the historical user behavior data is identified and processed to obtain search-related data corresponding to the historical user behavior data, including: performing word segmentation processing on the historical user behavior data to identify multiple keywords in the historical user behavior data; and obtaining search-related data corresponding to the historical user behavior data from the multiple keywords according to a preset item name vocabulary.

[0009] In some example embodiments of the present invention, based on the aforementioned scheme, the historical user behavior data includes search transaction data and search click data; the search associated data includes transaction associated phrases and click associated phrases; obtaining the search associated data corresponding to the historical user behavior data from the multiple keywords according to a preset item name vocabulary includes: obtaining the search terms and transaction item name terms in the search transaction data based on a preset item name vocabulary; combining the search terms with the transaction item name terms to determine the multiple associated phrases corresponding to the search transaction data; summing up the transaction volume for each of the multiple associated phrases according to the search transaction data to obtain the transaction volume corresponding to each associated phrase; sorting the multiple associated phrases according to the transaction volume to obtain transaction associated phrases that meet preset conditions.

[0010] In some example embodiments of the present invention, based on the aforementioned scheme, the obtaining of search-related data corresponding to the historical user behavior data from the multiple keywords according to a preset item name vocabulary also includes: obtaining the search terms and clicked item name terms in the search click data based on the preset item name vocabulary; combining the search terms with the clicked item name terms to determine multiple associated phrases corresponding to the search click data; summing up the click volume for each of the multiple associated phrases according to the search click data to obtain the click volume corresponding to each associated phrase; and sorting the multiple associated phrases according to the click volume to obtain click-related phrases that meet preset conditions.

[0011] In some example embodiments of the present invention, based on the aforementioned scheme, the search-related data is screened and processed to determine similar words corresponding to the historical user behavior data, which further includes: performing statistical calculations on the search-related data based on the data volume corresponding to the historical user behavior data to determine statistical data corresponding to the search-related data; and performing intersection operations on the statistical data to determine intersection result data of the search-related data.

[0012] In some example embodiments of the present invention, based on the aforementioned scheme, after determining the intersection result data of the search associated data, the search associated data is screened to determine the similar words corresponding to the historical user behavior data, including: determining the joint ratio corresponding to the search associated data based on the intersection result data; performing a Wilson transformation on the joint ratio to obtain a Wilson confidence interval of the search associated data; and filtering the Wilson confidence interval based on a preset threshold to determine the similar words corresponding to the historical user behavior data.

[0013] In some example embodiments of the present invention, based on the aforementioned scheme, performing an intersection operation on the statistical data through the search associated data to determine the intersection result data of the search associated data also includes: performing a union operation on the statistical data to determine the union result data of the search associated data.

[0014] According to a second aspect of an embodiment of the present invention, a similar word determination device is provided, comprising: an acquisition unit for acquiring historical user behavior data; an identification unit for performing identification processing on the historical user behavior data to obtain search-related data corresponding to the historical user behavior data; and a screening unit for performing screening processing on the search-related data to determine similar words corresponding to the historical user behavior data.

[0015] According to a third aspect of an embodiment of the present invention, there is provided an electronic device, comprising: a processor; and a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, any one of the similar word determination methods described above is implemented.

[0016] According to a fourth aspect of an embodiment of the present invention, there is provided a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the similar word determination method according to any one of the above items is implemented.

[0017] The technical solution provided by the embodiment of the present invention may have the following beneficial effects:

[0018] The similar word determination method in the exemplary embodiment of the present invention obtains historical user behavior data from the historical search log, identifies the historical user behavior data to obtain search-related data corresponding to the historical user behavior data, and filters the search-related data to obtain corresponding similar words. On the one hand, the historical user behavior data is identified to obtain search-related data, and the search-related data is filtered to obtain similar words corresponding to the historical user behavior data, which can eliminate the influence of interference data and improve the accuracy of similar words; on the other hand, obtaining corresponding similar words from historical user behavior data can meet the user's search input habits, further improve the accuracy of similar words, make similar words more complete, and improve the user's experience.

[0019] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The drawings herein are incorporated into and constitute a part of the specification, showing embodiments consistent with the present invention, and together with the specification, are used to explain the principles of the present invention. Obviously, the drawings described below are only some embodiments of the present invention, and for those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0021] Figure 1 A schematic diagram schematically illustrates a similar word determination method according to some embodiments of the present invention;

[0022] Figure 2 A schematic diagram schematically illustrates a process of determining similar words according to some embodiments of the present invention;

[0023] Figure 3 A schematic diagram schematically illustrates a method for processing user search transaction data according to some embodiments of the present invention;

[0024] Figure 4 A schematic diagram schematically shows a method for processing user search click data according to some embodiments of the present invention;

[0025] Figure 5 A schematic diagram schematically shows a similar word determination device according to some embodiments of the present invention;

[0026] Figure 6 A schematic diagram of the structure of a computer system of an electronic device according to some embodiments of the present invention is shown;

[0027] Figure 7 A schematic diagram of a computer-readable storage medium according to some embodiments of the present invention is schematically shown. DETAILED DESCRIPTION

[0028] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that the present invention will be more comprehensive and complete and fully convey the concept of the example embodiments to those skilled in the art.

[0029] In addition, the described features, structures or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present invention. However, those skilled in the art will appreciate that the technical solution of the present invention can be practiced without one or more of the specific details, or other methods, components, devices, steps, etc. may be adopted. In other cases, known methods, devices, implementations or operations are not shown or described in detail to avoid blurring various aspects of the present invention.

[0030] Furthermore, the drawings are only schematic illustrations and are not necessarily drawn to scale. The block diagrams shown in the drawings are only functional entities and do not necessarily correspond to physically separate entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0031] With the rapid development of the Internet industry and the widespread application of e-commerce technology, the types of items on e-commerce websites have become richer and more complete. The continuous growth in the number of items, the difficulty of standardizing and maintaining item information, and the high degree of personalization of search terms entered by users during searches have led to a low accuracy rate in matching search terms with items.

[0032] One way to improve this is to use "similar item words" to match the search terms entered by users. For example, if there are two similar item words for "small pillow", namely "small hug pillow" and "small cushion", when the user's search term includes "small pillow", the search results corresponding to "small pillow", "small hug pillow" and "small cushion" are returned. However, since similar item word libraries are mostly collected and organized manually, the similarity relationship between item name words is not sound and complete, and the accuracy rate is low, so it cannot accurately match the items that users want to search for, which reduces the user experience.

[0033] Based on this, in this exemplary embodiment, a similar word determination method is first provided. Figure 1 The following schematic diagram shows a similar word determination method according to some embodiments of the present invention. Figure 1 As shown, the similar word determination method may include the following steps:

[0034] Step S110, obtaining historical user behavior data;

[0035] Step S120, performing identification processing on the historical user behavior data to obtain search-related data corresponding to the historical user behavior data;

[0036] Step S130, screening the search-related data to determine similar words corresponding to the historical user behavior data.

[0037] According to the similar word determination method in this example embodiment, on the one hand, historical user behavior data is identified and processed to obtain search-related data, and the search-related data is filtered and processed to obtain similar words corresponding to the historical user behavior data, which can eliminate the influence of interference data and improve the accuracy of similar words; on the other hand, corresponding similar words are obtained from historical user behavior data to meet the user's input habits, further improve the accuracy of similar words, make similar words more complete, and improve the user's usage experience.

[0038] Next, the similar word determination method in this exemplary embodiment will be further described.

[0039] Continue to refer Figure 1 As shown, in step S110, historical user behavior data is obtained.

[0040] In some exemplary embodiments of the present invention, the system collects and stores the user's operation behavior data from the user search log. When the collected and stored valid user operation behavior data (i.e., the operation behavior data such as the user collecting, purchasing, clicking on a link, etc. after entering the search term can be regarded as valid data, and the data of the user exiting the search result page or re-entering the search term without operating the search result after entering the search term can be regarded as invalid data, of course, valid or invalid data also has different judgment methods according to different actual situations, and the present invention does not make special restrictions on this) reaches a preset threshold, the valid user operation behavior data is obtained as the historical user behavior data of the exemplary embodiment of the present invention. The historical user behavior data can be search transaction data (i.e., the user purchases a search result after entering the search term) and search click data (i.e., the user clicks on a link after entering the search term). Of course, the historical user behavior data can also be the operation behavior data of other users on the search results, for example, the historical user behavior data can be the operation behavior data of additional purchases or collections of the search results, and the present invention does not make special restrictions on this.

[0041] In step S120, the historical user behavior data is identified and processed to obtain search-related data corresponding to the historical user behavior data.

[0042] In some example embodiments of the present invention, since the historical user behavior data includes longer phrases or sentences that cannot be recognized by the system, for example, the search content input by the user is "sturdy and durable three-hole clamp", the server cannot recognize the sentence, so it is necessary to perform recognition processing on the historical user behavior data. Recognition processing can refer to the process of processing the phrases or sentences that cannot be recognized by the system in the historical user behavior data through relevant algorithms or models to generate keywords that can be recognized by the system, such as performing word segmentation processing on the historical user behavior data. Of course, recognition processing can also be other processing methods, algorithms or models. For example, recognition processing can refer to processing through a translation word alignment algorithm, and the present invention does not make special restrictions on this. Search-related data can refer to keyword groups obtained after the historical user behavior data is recognized and processed. For example, search-related data can refer to transaction-related phrases obtained from search transaction data, or click-related phrases obtained from search click data. The present invention does not make special restrictions on this.

[0043] Specifically, the system performs word segmentation processing on the historical user behavior data, identifies and determines multiple keywords in the historical user behavior data, and obtains search-related data corresponding to the historical user behavior data from multiple keywords according to a preset item name vocabulary. Word segmentation processing may refer to the process of processing historical user behavior data through relevant word segmentation processing algorithms, such as performing word segmentation processing on the search content "sturdy and durable three-hole clamp" through Word2vec (Word to Vector, a related model for generating word vectors), and generating keyword vectors such as "sturdy", "durable", and "three-hole clamp". The item name vocabulary may refer to a vocabulary including all item names that is set up in advance by relevant staff according to actual conditions. The system extracts the item name words contained in multiple keywords from the item name word library to form search-related phrases. For example, the user search content "sturdy and durable three-hole folder" in the historical user behavior data is segmented to generate keyword vectors such as "sturdy", "durable", and "three-hole folder". The search result links "sturdy folder" and "durable document folder" of the user's purchase operation behavior are segmented to generate "sturdy", "durable", "folder" and "document folder". The aforementioned multiple keyword vectors are filtered through the item name word library to obtain two search-related data of "three-hole folder, folder" and "three-hole folder, document folder" (the search-related data can be in the form of a phrase "search word, item name word". Of course, the search-related data can also be in other forms, such as a table form, which is not specifically limited in the present invention). By performing word segmentation on the historical user behavior data, not only can the system more accurately identify the search content entered by the user and the item name words in the search results, but also the efficiency of the system in processing data can be improved.

[0044] Furthermore, the system obtains the search terms and transaction item names in the search transaction data according to the item name word library, and combines the search terms with the transaction item names to determine multiple associated phrases corresponding to the search transaction data. For example, the user search content "sturdy and durable three-hole folder" in the historical user behavior data is segmented to generate keyword vectors such as "sturdy", "durable", and "three-hole folder". The search result links "sturdy folder" and "durable folder" of the user's purchase operation behavior are segmented to generate "sturdy", "durable", "folder" and "folder". The aforementioned multiple keyword vectors are filtered through the item name word library to obtain two search associated phrases "three-hole folder, folder" and "three-hole folder, folder". Similarly, for the item name word "folder" in the search result link of the user's purchase operation behavior, there are multiple search terms "three-hole folder" and "folder", and two associated phrases "folder, three-hole folder" and "folder, folder" are obtained (the phrase form is "item name word, search word"). By combining search terms and transaction item names, the accuracy of the data can be improved, key data will not be missed, and the integrity of the data can be ensured.

[0045] The system calculates the sum of the transaction volume of each of the multiple associated phrases obtained in the above steps according to the search transaction data, determines the transaction volume corresponding to each associated phrase, sorts the multiple associated phrases according to the transaction volume (the sorting may refer to descending sorting or ascending sorting, which is not specifically limited), and obtains the transaction associated phrases that meet the preset conditions. The preset conditions may refer to the numerical values ​​for screening multiple associated phrases according to the actual situation. For example, the preset condition K is 100, that is, the system sorts the multiple associated phrases in descending order according to the transaction volume corresponding to each associated phrase, and obtains the associated phrases ranked in the top 100 as the transaction associated phrases for subsequent calculation (search associated data). Of course, it is also possible that the system sorts the multiple associated phrases in ascending order according to the transaction volume corresponding to each associated phrase, removes the associated phrases ranked in the top 100, and uses the remaining associated phrases as transaction associated phrases. By screening the associated phrases obtained by the transaction volume, invalid data such as 0 transaction volume or missing data can be eliminated in the data, ensuring the accuracy of the data and improving the efficiency of system processing.

[0046] Optionally, the system obtains the search terms and clicked item name terms in the search click data based on the item name vocabulary, and combines the search terms with the clicked item name terms to determine multiple associated phrases corresponding to the search click data. The system sums up the click volume for each of the multiple associated phrases based on the search click data to determine the click volume corresponding to each associated phrase, sorts the multiple associated phrases based on the click volume, and obtains click-associated phrases that meet preset conditions. The specific implementation method corresponding to this step can refer to the aforementioned step of obtaining transaction-associated phrases, which will not be described in detail here. By combining the search terms and clicked item name terms, the accuracy of the data can be improved and the integrity of the data can be ensured. By screening the obtained associated phrases based on the click volume, invalid data such as zero click volume or missing data can be eliminated, thereby ensuring the accuracy of the data and improving the efficiency of system processing.

[0047] In step S130, the search-related data is screened to determine similar words corresponding to the historical user behavior data.

[0048] In some example embodiments of the present invention, the system performs screening processing on the search-related data obtained by performing identification processing on the historical user behavior data. The screening processing may refer to a series of processing processes that can extract similar words in the search-related data. For example, the screening processing may include statistical processing, conversion processing, filtering processing, etc., and the present invention does not specifically limit this. The system obtains a similar word list (lexicon) corresponding to the historical user behavior data through screening processing on the search-related data, and uses the similar word list (lexicon) for the corresponding search service to improve the accuracy of the item word matched to the search word entered by the user.

[0049] Specifically, the system performs statistical calculations on the search-related data according to the data volume corresponding to the historical user behavior data to determine the statistical data corresponding to the search-related data; and performs intersection operations on the statistical data to determine the intersection result data of the search-related data. Statistical data may refer to feature data obtained by performing simple statistical calculations on the search-related data, such as performing statistical calculations on the transaction item name words corresponding to each search term based on the search-related data, the transaction volume corresponding to the transaction item name words corresponding to each search term in the transaction-related data (denoted as a), the total transaction volume corresponding to each search term (the sum of the order volumes corresponding to all transaction item names of the current search term, denoted as b), and the corresponding ratio a / b, and the obtained statistical data is recorded as "search term, transaction item name word, transaction volume a corresponding to the transaction item name word, total transaction volume b corresponding to the search term, ratio a / b". For example, the results of the statistical data may be recorded in the form of a table, as shown in Table 1:

[0050] Table 1 Statistics of transaction item names corresponding to each search term

[0051]

[0052] Similarly, based on the search-related data, the search terms corresponding to each transaction item name are statistically calculated, and the obtained statistical data is recorded as "transaction item name, search term, transaction volume c corresponding to the search term, total order volume d corresponding to the transaction item name, ratio c / d". For example, the statistical data can be recorded in the form of a table, as shown in Table 2:

[0053] Table 2 Statistics of search terms corresponding to each trading item name

[0054]

[0055] Based on the search terms in Table 1 (e.g., the search term "three-hole folder") and the transaction item names in Table 2 (e.g., the transaction item name "folder"), the system performs intersection calculation on the statistical data of the transaction item name corresponding to each search term and the statistical data of the search term corresponding to each transaction item name, that is, only the records of the item name phrases that appear in both statistical data are retained, and the obtained intersection result data records are "item name term 1, item name term 2, transaction volume a corresponding to the transaction item name, total transaction volume b corresponding to the search term, ratio a / b, transaction volume c corresponding to the search term, total order volume d corresponding to the transaction item name, ratio c / d", for example, the statistical data of the examples in Table 1 and Table 2 are intersected to obtain "three-hole folder, folder, 3, 12, 0.25, 1, 3, 0.33". By performing intersection operation on the statistical data corresponding to the search-related data, the similar words finally obtained can be made more accurate, and can be applied to search service scenarios with high accuracy requirements.

[0056] Furthermore, after determining the intersection result data of the search related data, the system determines the joint ratio corresponding to the search related data based on the intersection result data; the joint ratio is subjected to Wilson transformation to obtain the Wilson confidence interval value of the search related data (the Wilson confidence interval is a statistical measure that shows the degree to which the true value of the measured parameter has a certain probability of falling around the measurement result, that is, the confidence interval gives the degree of credibility of the measured value of the measured parameter, which can also be considered as the credibility of the conclusion). The system calculates the joint ratio (denoted as e) based on the intersection result data obtained in the above steps (that is, "item name 1, item name 2, transaction volume a corresponding to the transaction item name, total transaction volume b corresponding to the search term, ratio a / b, transaction volume c corresponding to the search term, total order volume d corresponding to the transaction item name, ratio c / d"). 1 ), as shown in formula (1):

[0057]

[0058] The obtained joint ratio is subjected to Wilson transformation to obtain the Wilson confidence interval value (denoted as e), as shown in formula (2):

[0059]

[0060] Where n is the sum of the total transaction volume b corresponding to the search term and the total order volume d corresponding to the transaction item name, and z is the confidence level. The z-statistic of .

[0061] The system filters the Wilson confidence interval value according to the preset threshold to determine the similar words corresponding to the historical user behavior data. The preset threshold can refer to the value determined by the relevant staff to filter the Wilson confidence interval value according to the actual situation. For example, if the preset threshold is 0.1, only data with e greater than or equal to 0.1 will be retained. At this time, the filtered data is recorded as "item name word 1, item name word 2, joint ratio e 1 ". By calculating the joint ratio of statistical data and filtering the data according to the Wilson execution interval value corresponding to the joint ratio, the accuracy and completeness of the data can be further improved.

[0062] Optionally, a union operation may be performed on the statistical data to determine the union result data of the search-related data. The calculation steps of the union result data are the same as the calculation steps of the aforementioned intersection result data, except that the final result is different, and the detailed process will not be repeated here. Similarly, the corresponding joint ratio (denoted as f) is calculated based on the union result data, and a Wilson transformation is performed to obtain the Wilson confidence interval value of the union result data, and the Wilson confidence interval value is filtered and processed according to a preset threshold to obtain the result data, which is recorded as "item name word 1, item name word 2, joint ratio f". By performing a union operation on the search-related data, the coverage of the obtained similar words can be larger, and the recall rate of similar words can be improved, which can be applied to search service scenarios with higher recall requirements.

[0063] refer to Figure 2 As shown, Figure 2 A schematic diagram of a process for determining similar words according to some embodiments of the present invention is schematically shown, and the steps in the diagram are described in detail below.

[0064] Step S210, processing the search transaction data in the historical user behavior data to obtain transaction-related phrases corresponding to the search transaction data;

[0065] Step S220, processing the search click data in the historical user behavior data to obtain click-related phrases corresponding to the search click data;

[0066] Step S230, statistically calculating and filtering the transaction-related phrases obtained in step S210 and the click-related phrases obtained in step S220, and merging the obtained result data;

[0067] Step S240, sorting the result data obtained in step S230 to obtain a similar word list (lexicon).

[0068] refer to Figure 3 As shown, Figure 3 A schematic diagram of a method for processing user search transaction data according to some embodiments of the present invention is schematically shown, and the steps described in the diagram are described in detail below.

[0069] Step S310, obtaining search content and search results from the user search transaction log (i.e., search transaction data in the user's historical behavior data), and calculating the transaction volume corresponding to the search content and search results;

[0070] Step S320, sorting the search results corresponding to each search content in descending order according to the transaction volume obtained in step S310, and obtaining the top K search results corresponding to each search content;

[0071] Step S330, determining the search content and the product words (search words) corresponding to the search results based on the preset item name word library, and obtaining the top K search result product words (trading item name words) corresponding to each search content product word;

[0072] Step S340, performing data preprocessing and statistical calculation on the K search result product words corresponding to each search content product word in step S330 to obtain corresponding statistical data;

[0073] Step S350: Similarly, sort the search content corresponding to each search result in descending order according to the transaction volume obtained in step S310, and obtain the top K search content corresponding to each search result;

[0074] Step S360, determining the search results and product words corresponding to the search content based on the preset item name word library, and obtaining the top K search content product words corresponding to each search result product word;

[0075] Step S370, performing data preprocessing and statistical calculation on the K search content product words corresponding to each search result product word in step S360 to obtain corresponding statistical data;

[0076] Step S380, calculating the joint ratio corresponding to the statistical data obtained in step 340 and step S370, performing Wilson transformation on the joint ratio to obtain a Wilson confidence interval value, and screening the Wilson confidence interval value according to a preset threshold;

[0077] Step S390, organizing similar words contained in the search transaction data corresponding to the statistical data obtained by the screening process in step S380 to generate a similar word list (thesaurus).

[0078] refer to Figure 4 As shown, Figure 4 A schematic diagram of a method for processing user search click data according to some embodiments of the present invention is schematically shown, and the steps described in the diagram are described in detail below.

[0079] Step S410, obtaining search content and search results from the user search click log (i.e., search click data in the user's historical behavior data), and calculating the transaction volume corresponding to the search content and search results;

[0080] Step S420, sorting the search results corresponding to each search content in descending order according to the number of clicks obtained in step S410, and obtaining the top K search results corresponding to each search content;

[0081] Step S430, determining the search content and the product words (search words) corresponding to the search results based on the preset item name word library, and obtaining the top K search result product words (trading item name words) corresponding to each search content product word;

[0082] Step S440, performing data preprocessing and statistical calculation on the K search result product words corresponding to each search content product word in step S430 to obtain corresponding statistical data;

[0083] Step S450: Similarly, sort the search content corresponding to each search result in descending order according to the number of clicks obtained in step S410, and obtain the top K search content corresponding to each search result;

[0084] Step S460, determining the search results and product words corresponding to the search content based on the preset item name word library, and obtaining the top K search content product words corresponding to each search result product word;

[0085] Step S470, performing data preprocessing and statistical calculation on the K search content product words corresponding to each search result product word in step S460 to obtain corresponding statistical data;

[0086] Step S480, calculating the joint ratio corresponding to the statistical data obtained in step S440 and step S470, performing Wilson transformation on the joint ratio to obtain a Wilson confidence interval value, and screening the Wilson confidence interval value according to a preset threshold;

[0087] Step S490, organizing similar words contained in the search click data corresponding to the statistical data obtained by the screening process in step S480 to generate a similar word list (thesaurus).

[0088] It should be noted that, in the embodiment of the present invention, only the processing process of the search transaction data and the search click data in the historical user behavior data is illustrated in detail. The historical user behavior data may also be the operation behavior data such as additional purchase or collection of the search results. Therefore, the processing process of the historical user behavior data may also include the processing process of the user's operation behavior data such as additional purchase or collection of the search results. The detailed processing process refers to the processing process of the search transaction data and the search click data, which will not be repeated here.

[0089] It should be noted that, although the steps of the method of the present invention are described in a specific order in the drawings, this does not require or imply that the steps must be performed in this specific order, or that all the steps shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps, etc.

[0090] In addition, in this exemplary embodiment, a similar word determination device is also provided. Figure 5 As shown, the similar word determination device 500 includes: an acquisition unit 510 for acquiring historical user behavior data; an identification unit 520 for performing identification processing on the historical user behavior data to obtain search-related data corresponding to the historical user behavior data; and a screening unit 530 for performing screening processing on the search-related data to determine similar words corresponding to the historical user behavior data.

[0091] In an exemplary embodiment of the present invention, based on the aforementioned scheme, the identification unit 520 also includes: a word segmentation processing unit, used to perform word segmentation processing on the historical user behavior data to identify multiple keywords in the historical user behavior data; a search-related data acquisition unit, used to acquire search-related data corresponding to the historical user behavior data from the multiple keywords according to a preset item name vocabulary.

[0092] In an exemplary embodiment of the present invention, based on the aforementioned scheme, the search associated data acquisition unit is configured to: acquire the search terms and transaction item name terms in the search transaction data based on the item name vocabulary; combine the search terms with the transaction item name terms to determine a plurality of associated phrases corresponding to the search transaction data; sum up the transaction volume of each of the plurality of associated phrases according to the search transaction data to determine the transaction volume corresponding to each associated phrase; sort the plurality of associated phrases according to the transaction volume to acquire transaction associated phrases that meet preset conditions.

[0093] In an exemplary embodiment of the present invention, based on the aforementioned scheme, the search associated data acquisition unit is configured to: acquire the search terms and clicked item name terms in the search click data based on a preset item name vocabulary; combine the search terms with the clicked item name terms to determine a plurality of associated phrases corresponding to the search click data; sum up the click volume for each of the plurality of associated phrases according to the search click data to acquire the click volume corresponding to each associated phrase; and sort the plurality of associated phrases according to the click volume to acquire click associated phrases that meet preset conditions.

[0094] In an exemplary embodiment of the present invention, based on the aforementioned scheme, the screening unit 530 is configured to: perform statistical calculations on the search-related data according to the data volume corresponding to the historical user behavior data to determine the statistical data corresponding to the search-related data; and perform intersection operations on the statistical data to determine the intersection result data of the search-related data.

[0095] In an exemplary embodiment of the present invention, based on the aforementioned scheme, the screening unit 530 is configured to: determine the joint ratio corresponding to the search-related data according to the intersection result data; perform Wilson transformation on the joint ratio to obtain the Wilson confidence interval value of the search-related data; and screen the Wilson confidence interval value according to a preset threshold to determine similar words corresponding to the historical user behavior data.

[0096] The specific details of each module of the similar word determination device mentioned above have been described in detail in the corresponding similar word determination method, so they will not be repeated here.

[0097] It should be noted that, although several modules or units of similar word determination device are mentioned in the above detailed description, this division is not mandatory. In fact, according to an embodiment of the present invention, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into being embodied by multiple modules or units.

[0098] In addition, in an exemplary embodiment of the present disclosure, an electronic device capable of implementing the above-mentioned similar word determination method is also provided.

[0099] It will be appreciated by those skilled in the art that various aspects of the present invention may be implemented as systems, methods or program products. Therefore, various aspects of the present invention may be specifically implemented in the following forms, namely: complete hardware embodiments, complete software embodiments (including firmware, microcode, etc.), or embodiments combining hardware and software aspects, which may be collectively referred to herein as "circuits", "modules" or "systems".

[0100] Refer to the following Figure 6 An electronic device 600 according to such an embodiment of the present invention is described. Figure 6 The electronic device 600 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0101] like Figure 6 As shown, the electronic device 600 is in the form of a general computing device. The components of the electronic device 600 may include but are not limited to: at least one processing unit 610, at least one storage unit 620, a bus 630 connecting different system components (including the storage unit 620 and the processing unit 610), and a display unit 640.

[0102] The storage unit stores program codes, which can be executed by the processing unit 610, so that the processing unit 610 performs the steps according to various exemplary embodiments of the present invention described in the above “Exemplary Method” section of this specification. For example, the processing unit 610 can perform the following steps: Figure 1 In step S110 shown in the figure, historical user behavior data is obtained; in step S120, the historical user behavior data is identified and processed to obtain search-related data corresponding to the historical user behavior data; in step S130, the search-related data is screened and processed to determine similar words corresponding to the historical user behavior data.

[0103] The storage unit 620 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 621 and / or a cache memory unit 622 , and may further include a read-only memory unit (ROM) 623 .

[0104] The storage unit 620 may also include a program / utility 624 having a set (at least one) of program modules 625, such program modules 625 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0105] Bus 630 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0106] The electronic device 600 may also communicate with one or more external devices 670 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 600, and / or communicate with any device that enables the electronic device 600 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 650. Furthermore, the electronic device 600 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter 660. As shown, the network adapter 660 communicates with other modules of the electronic device 600 via a bus 630. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 600, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0107] Through the description of the above embodiments, it is easy for those skilled in the art to understand that the example embodiments described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the embodiment of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the embodiment of the present disclosure.

[0108] In an exemplary embodiment of the present disclosure, a computer-readable storage medium is also provided, on which a program product capable of implementing the above method of the present specification is stored. In some possible embodiments, various aspects of the present invention can also be implemented in the form of a program product, which includes a program code, and when the program product is run on a terminal device, the program code is used to enable the terminal device to perform the steps according to various exemplary embodiments of the present invention described in the above "Exemplary Method" section of the present specification.

[0109] refer to Figure 7As shown, a program product 700 for implementing the similar word determination method according to an embodiment of the present invention is described, which can adopt a portable compact disk read-only memory (CD-ROM) and include program code, and can be run on a terminal device, such as a personal computer. However, the program product of the present invention is not limited thereto. In this document, a readable storage medium can be any tangible medium containing or storing a program, which can be used by or in combination with an instruction execution system, an apparatus or a device.

[0110] The program product may use any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0111] Computer readable signal media may include data signals propagated in baseband or as part of a carrier wave, in which readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Readable signal media may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0112] The program code embodied on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the foregoing.

[0113] Program code for performing the operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).

[0114] In addition, the above-mentioned figures are only schematic illustrations of the processes included in the method according to an exemplary embodiment of the present invention, and are not intended to be limiting. It is easy to understand that the processes shown in the above-mentioned figures do not indicate or limit the time sequence of these processes. In addition, it is also easy to understand that these processes can be performed synchronously or asynchronously, for example, in multiple modules.

[0115] Through the description of the above embodiments, it is easy for those skilled in the art to understand that the example embodiments described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the embodiment of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a touch terminal, or a network device, etc.) to execute the method according to the embodiment of the present disclosure.

[0116] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The specification and examples are to be considered exemplary only, and the true scope and spirit of the present disclosure are indicated by the claims.

[0117] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A method for determining similar words, characterized in that: include: Obtain historical user behavior data; Performing identification processing on the historical user behavior data to obtain search-related data corresponding to the historical user behavior data; Performing statistical calculations on the search-related data according to the data volume corresponding to the historical user behavior data to determine the statistical data corresponding to the search-related data; performing an intersection operation on the statistical data to determine the intersection result data of the search-related data; determining the joint ratio corresponding to the search-related data according to the intersection result data; performing a Wilson transformation on the joint ratio to obtain a Wilson confidence interval value of the search-related data; The Wilson confidence interval value is screened according to a preset threshold value to determine similar words corresponding to the historical user behavior data.

2. The similar word determination method according to claim 1, characterized in that: Performing identification processing on the historical user behavior data to obtain search-related data corresponding to the historical user behavior data includes: Performing word segmentation processing on the historical user behavior data to identify multiple keywords in the historical user behavior data; The search-related data corresponding to the historical user behavior data is obtained from the multiple keywords according to a preset object name vocabulary.

3. The similar word determination method according to claim 2, characterized in that: The historical user behavior data includes search transaction data and search click data; the search association data includes transaction-related phrases and click-related phrases; Acquiring search-related data corresponding to the historical user behavior data from the plurality of keywords according to a preset object name vocabulary includes: Based on the item name word library, obtaining the search words and transaction item name words in the search transaction data; Combining the search word with the transaction item name word to determine a plurality of associated word groups corresponding to the search transaction data; Performing a summation calculation on the transaction volume of each associated phrase in the plurality of associated phrases according to the search transaction data to determine the transaction volume corresponding to each associated phrase; The plurality of associated phrases are sorted according to the transaction volume to obtain transaction associated phrases that meet preset conditions.

4. The similar word determination method according to claim 3, characterized in that: The step of acquiring search-related data corresponding to the historical user behavior data from the plurality of keywords according to the preset object name word library further includes: Based on the item name word library, obtaining the search words and clicked item name words in the search click data; Combining the search word with the clicked item name word to determine a plurality of associated phrases corresponding to the search click data; According to the search click data, the click amount of each associated phrase in the plurality of associated phrases is summed up to determine the click amount corresponding to each associated phrase; The plurality of associated phrases are sorted according to the click volume to obtain click-associated phrases that meet preset conditions.

5. The similar word determination method according to claim 1, characterized in that: Performing an intersection operation on the statistical data through the search-related data to determine the intersection result data of the search-related data also includes: A union operation is performed on the statistical data to determine union result data of the search related data.

6. A similar word determination device, characterized in that: include: An acquisition unit, used to acquire historical user behavior data; an identification unit, configured to perform identification processing on the historical user behavior data to obtain search-related data corresponding to the historical user behavior data; A screening unit is used to perform statistical calculations on the search-related data according to the data volume corresponding to the historical user behavior data to determine the statistical data corresponding to the search-related data; perform an intersection operation on the statistical data to determine the intersection result data of the search-related data; determine the joint ratio corresponding to the search-related data according to the intersection result data; perform a Wilson transformation on the joint ratio to obtain a Wilson confidence interval value of the search-related data; The Wilson confidence interval value is screened according to a preset threshold value to determine similar words corresponding to the historical user behavior data.

7. An electronic device comprising: processor; as well as A memory having computer-readable instructions stored thereon, wherein the computer-readable instructions, when executed by the processor, implement the similar word determination method according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the similar word determination method according to any one of claims 1 to 5.

Images