An e-commerce search processing system and method

By building a personalized search processing model and real-time update of the search database, the problems of poor user experience and low search accuracy in existing e-commerce search processing technologies are solved, and a higher user experience and e-commerce platform transaction rate is achieved.

CN115269988BActive Publication Date: 2024-12-03JIANGSU HUAZE WEIFU TECH DEV CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210924850.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-08-02
Publication Date
2024-12-03
Estimated Expiration
2042-08-02

AI Technical Summary

Technical Problem

The existing e-commerce search processing technology adopts a fixed method, resulting in poor user experience, low search accuracy, and lack of personalized processing.

Method used

An e-commerce search processing system and method was designed, including obtaining modules, processing modules, building modules, building modules, search modules, sorting modules, display modules, preprocessing modules, training modules and subsequent search processing modules, by building personalized Search processing model, update the search database in real time, provide search guidance and hot word detection, enhance user experience and search accuracy.

Benefits of technology

It improves the user's search experience and search processing accuracy, and improves the product transaction rate of e-commerce platforms.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115269988B_ABST
    Figure CN115269988B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of e-commerce search processing technology, and in particular to an e-commerce search processing system and method. In view of the problem that the current existing e-commerce search processing technology adopts a fixed search processing method and lacks personalized search processing for users, resulting in poor user search experience and low search processing accuracy, the following scheme is proposed, wherein an e-commerce search processing system includes an acquisition module, a processing module, an establishment module, a construction module, a retrieval module, a sorting module, a display module, an e-commerce search processing module, a preprocessing module, a training module, and a subsequent search processing module. The purpose of the present invention is to acquire the search preference of the user and construct a personalized search processing model to enhance the user's search processing experience, and train the constructed model to enhance the accuracy of the search processing. At the same time, through the subsequent search processing, search guidance is provided to the user, thereby improving the commodity transaction rate of the e-commerce platform.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of e-commerce search processing, and in particular to an e-commerce search processing system and method. Background Art

[0002] With the rapid development of the Internet, e-commerce has become the choice of more and more people. When it comes to e-commerce, you may think of Taobao online shopping, Jingdong delivery, Meituan ordering... In fact, in layman's terms, e-commerce is to use computer networks to carry out business activities in accordance with certain technologies, models, and standards, so as to help companies reduce operating costs and achieve differentiated competitive advantages. It can be said that in the field of business, it represents the future development direction of productivity. Based on the rapid development of the Internet in recent years, the development of e-commerce has been very rapid in any country or region. Compared with foreign countries, my country's e-commerce started relatively late. However, this does not affect the rapid development of my country's e-commerce, because my country's e-commerce currently has a large consumer group and advanced Internet technology, which provides an important foundation and driving force for the development of e-commerce. Because e-commerce has great development prospects, many investors in my country have invested in the field of e-commerce, which has also made the competition among e-commerce increasingly fierce.

[0003] However, the existing e-commerce search processing technology currently uses a fixed search processing method and lacks personalized search processing for users, resulting in poor user search experience and low search processing accuracy. Therefore, we propose an e-commerce search processing system and method to solve the above problems. Summary of the invention

[0004] The purpose of the present invention is to solve the problems that the current existing e-commerce search processing technology adopts a fixed search processing method, lacks personalized search processing for users, leads to poor user search experience and low search processing accuracy, and proposes an e-commerce search processing system and method.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] An e-commerce search processing system comprises an acquisition module, a processing module, an establishment module, a construction module, a retrieval module, a sorting module, a display module, an e-commerce search processing module, a pre-processing module, a training module, and a subsequent search processing module, wherein the acquisition module is respectively connected to the processing module, the e-commerce search processing module, and the display module, the processing module is connected to the establishment module, the establishment module is connected to the construction module, the construction module is respectively connected to the retrieval module, the display module, and the training module, the retrieval module is connected to the sorting module, the sorting module is respectively connected to the display module and the pre-processing module, the e-commerce search processing module is connected to the pre-processing module, the training module is connected to the subsequent search processing module, wherein the subsequent search processing module comprises a real-time update unit, a guide unit, and a detection unit;

[0007] Preferably, the acquisition module is used to acquire the existing e-commerce product information, user search terms and new search weights of the e-commerce platform, the processing module is used to process the acquired e-commerce product information, the establishment module is used to establish a product database framework using the acquired classification directory as a storage directory, the construction module is used to build a product database and a personalized search processing model, the retrieval module is used to search the retrieval database, the sorting module is used to sort the products according to the acquired search weights, the display module is used to display the products through the obtained sorting list, the e-commerce search processing module is used to perform e-commerce search processing, the pre-processing module is used to perform Query pre-processing on the acquired user search terms, the training module is used to train the constructed personalized search processing model, and the subsequent search processing module is used to perform real-time update of the retrieval database, pre-search guidance and real-time hot word detection;

[0008] The present invention also provides an e-commerce search processing method, comprising the following steps:

[0009] S1: Build a product database: Professionals build a product database;

[0010] S2: Search: The e-commerce platform searches through the search database;

[0011] S3: e-commerce search processing: the e-commerce platform performs e-commerce search processing;

[0012] S4: Building a personalized search processing model: The product platform automatically builds a personalized search processing model;

[0013] S5: Subsequent search processing: The e-commerce platform performs subsequent search processing;

[0014] Preferably, in S1, a commodity database is constructed by professionals, wherein before constructing the commodity database, professionals first obtain the existing e-commerce commodity information on the e-commerce platform, wherein the e-commerce commodity information includes commodity name, commodity variety, commodity specification, commodity style, commodity color, commodity new technology, and commodity new material, and professionals process the obtained e-commerce commodity information, wherein during the processing, a classification catalog is obtained by computer according to the existing commodity classification, and the obtained classification catalog is used as a storage catalog to establish a commodity database framework, and at the same time, professionals store the obtained e-commerce commodity information through the commodity data framework to construct the commodity database, wherein the obtained e-commerce commodity information needs to be stored in the corresponding classification catalog during storage;

[0015] Preferably, in S2, the e-commerce platform searches through a search database, wherein the search database is a database of all user search records stored by the e-commerce platform, and when searching, the search weights of different classification directories are first obtained through searching, and the obtained search weights are used as the first search weights, and at the same time, the search weights of users for the same product name in different classification directories are obtained through searching, and the obtained search weights are used as the second search weights, and the products are sorted according to the obtained search weights, and the products are displayed through sorting, wherein when sorting, the e-commerce platform first performs preliminary sorting according to the first search weight, and after the preliminary sorting is completed, the e-commerce platform performs final sorting according to the second search weight, and when sorting, the classification directories and product names with large weights are sorted in the front, and the classification directories and product names with small weights are sorted in the back, and a product sorting list is obtained through sorting, and the e-commerce platform displays the products through the obtained sorting list;

[0016] Preferably, in S3, the e-commerce platform performs e-commerce search processing, wherein the e-commerce search engine first obtains the user's search terms when performing the e-commerce search processing, and performs Query preprocessing on the obtained user search terms, wherein the search engine first identifies the search terms when performing the Query preprocessing, and performs processing based on the identification results, wherein the identification results show that the obtained search terms contain stop words, then the search engine removes the stop words, and the identification results show that the obtained search terms do not contain stop words, then the search terms are normalized, wherein the normalization processing is that the search engine performs special symbol filtering, case normalization, traditional Chinese to simplified Chinese, and full-width to half-width on the search terms, and after the normalization processing is completed, the search engine performs pinyin conversion and synonym replacement on the search terms, and at the same time performs word segmentation, completion, rewriting and error correction on the search terms to obtain the product name and the category directory where the product is located, obtains a new search weight through the obtained product name and category directory, sorts through the obtained search weight to obtain a new sorted list, and displays the products through the new sorted list;

[0017] Preferably, in S4, the commodity platform automatically constructs a personalized search processing model, wherein when constructing the model, the e-commerce platform first constructs the search processing model through the constructed commodity database, and after the search processing model is constructed, the e-commerce platform obtains the user's search records, purchase records, browsing records and return records, and analyzes the obtained information, obtains the user's purchase preferences through the analysis results, increases the classification directory corresponding to the commodity and the search weight of the commodity through the obtained purchase preferences, and obtains the user's personalized commodity ranking through the obtained new classification directory and commodity search weight, and stores the obtained user's personalized commodity ranking into the search model to construct a personalized search processing model. After the personalized search processing model is constructed, the constructed personalized search processing model is trained by professionals, wherein during the training, the professionals preset the search commodity name, obtain the commodity display arrangement order by inputting the preset commodity name, and the professionals view the order, make a judgment based on the viewing result, and process the judgment result, wherein if the viewing result shows that the first place in the commodity ranking list is different from the preset search commodity name, it is judged as personalized. If the search processing model is wrong, and the checking result shows that the first place in the product sorting list is the same as the preset search product name, the personalized search processing model is judged to be correct. If the judgment result is that the personalized search processing model is wrong, the training is continued, and the training result is judged. The search processing accuracy of the personalized search processing model is calculated by the judgment result. The training is stopped until the search processing accuracy of the personalized search processing model is higher than 98%. The personalized search processing model is used. If the judgment result is that the personalized search processing model is correct, the constructed personalized search processing model is continued to be trained, and the training result is checked. The judgment is made by checking the result, and the processing is performed by the judgment result. If the checking result is that the personalized search processing model is correct, the professional records the number of consecutive judgment results. If the checking result is that the personalized search processing model is wrong, the professional calculates and obtains the search processing accuracy of the personalized search processing model. If the judgment result is that the personalized search processing model is correct for more than 100 consecutive times and the search processing accuracy of the personalized search processing model is higher than 98%, the training is stopped, and the personalized search processing model is used.

[0018] Preferably, in S5, the e-commerce platform performs subsequent search processing, wherein the subsequent search processing includes real-time updating of the retrieval database, pre-search guidance, and real-time hot word detection. The updating of the retrieval database is performed by the e-commerce platform to obtain the search records of all users of the platform in real time, and store the real-time obtained data. The e-commerce platform performs real-time retrieval to update the first search proportion and the second search proportion data in real time, and obtains a real-time commodity ranking list through real-time data update. The e-commerce platform displays the user searched commodities through the obtained real-time commodity ranking list. During the pre-search guidance, the e-commerce platform obtains the commodity search hot words of the upper platform in real time, compares the obtained commodity search hot words with the user's purchase preferences, and processes the comparison results. , where the comparison result shows that there are overlapping elements between the obtained product search hot words and the user's purchase preferences, the e-commerce platform extracts the product search hot words containing the overlapping elements, and sorts the extracted product search hot words in the front row of the hot word search display area. The real-time hot word detection is performed by the e-commerce platform to detect the product hot words in real time and obtain the detection results. The e-commerce platform uses the obtained product hot words as product sub-labels, where the product sub-labels are equivalent to the product names when the user searches. The products are sorted according to the user's retrieval preferences and displayed through the sorting. At the same time, the e-commerce platform performs calculations to obtain the product transaction rate, and compares the obtained results with the product transaction rate data generated when the e-commerce platform closes the hot word search display area.

[0019] Compared with the prior art, the present invention has the following beneficial effects:

[0020] 1. By obtaining the user's search preferences and building a personalized search processing model, the user's search processing experience is enhanced, and the constructed model is trained to enhance the accuracy of search processing.

[0021] 2. Providing search guidance to users through subsequent search processing improves the commodity transaction rate of the e-commerce platform.

[0022] The purpose of the present invention is to build a personalized search processing model by acquiring the user's search preferences, enhance the user's search processing experience, and train the constructed model to enhance the accuracy of search processing. At the same time, through subsequent search processing, search guidance is provided to users, thereby improving the commodity transaction rate of the e-commerce platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 This is a structural diagram of an e-commerce search processing system proposed by the present invention;

[0024] Figure 2 This is a flow chart of an e-commerce search processing method proposed by the present invention;

[0025] Figure 3 A simulation diagram for building a commodity database of an e-commerce search processing system and method proposed by the present invention;

[0026] Figure 4 This is a query preprocessing diagram of an e-commerce search processing system and method proposed by the present invention. DETAILED DESCRIPTION

[0027] The technical solutions in the embodiments of the present invention will be described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0028] Embodiment 1

[0029] Reference Figure 1-4 An e-commerce search processing system includes an acquisition module, a processing module, an establishment module, a construction module, a retrieval module, a sorting module, a display module, an e-commerce search processing module, a preprocessing module, a training module, and a subsequent search processing module. The acquisition module is connected to the processing module, the e-commerce search processing module, and the display module respectively. The processing module is connected to the establishment module, the establishment module is connected to the construction module, the construction module is connected to the retrieval module, the display module, and the training module respectively. The retrieval module is connected to the sorting module, the sorting module is connected to the display module and the preprocessing module respectively, the e-commerce search processing module is connected to the preprocessing module, and the training module is connected to the subsequent search processing module, wherein the subsequent search processing module includes a real-time update unit, a guide unit, and a detection unit, wherein the acquisition module is used to obtain the existing E-commerce product information, user search terms and new search weights, the processing module is used to process the acquired e-commerce product information, the establishment module is used to establish a product database framework using the acquired classification directory as a storage directory, the construction module is used to build a product database and a personalized search processing model, the retrieval module is used to search the retrieval database, the sorting module is used to sort the products according to the acquired search weights, the display module is used to display the products through the obtained sorting list, the e-commerce search processing module is used to perform e-commerce search processing, the pre-processing module is used to perform Query pre-processing on the acquired user search terms, the training module is used to train the constructed personalized search processing model, and the subsequent search processing module is used to perform real-time update of the retrieval database, pre-search guidance and real-time hot word detection;

[0030] The present invention also provides an e-commerce search processing method, comprising the following steps:

[0031] S1: Constructing a commodity database: Professionals construct a commodity database, wherein before constructing the commodity database, the professionals first obtain the existing e-commerce commodity information on the e-commerce platform, wherein the e-commerce commodity information includes commodity name, commodity variety, commodity specification, commodity style, commodity color, commodity new technology, and commodity new material, and the professionals process the obtained e-commerce commodity information, wherein during the processing, a classification catalog is obtained by computer according to the existing commodity classification, and the obtained classification catalog is used as a storage catalog to establish a commodity database framework, and at the same time, the professionals store the obtained e-commerce commodity information through the commodity data framework to construct the commodity database, wherein the obtained e-commerce commodity information needs to be stored in the corresponding classification catalog during storage;

[0032] S2: Searching: The e-commerce platform searches through a search database, wherein the search database is a database of all user search records stored by the e-commerce platform, and when searching, first obtains the search weights of different classification directories through searching, and uses the obtained search weights as the first search weights, and at the same time obtains the search weights of users for the same product name in different classification directories through searching, and uses the obtained search weights as the second search weights, sorts the products according to the obtained search weights, and displays the products through sorting, wherein when sorting, the e-commerce platform first performs preliminary sorting according to the first search weights, and after the preliminary sorting is completed, the e-commerce platform performs final sorting according to the second search weights, and when sorting, the classification directories and product names with large weights are sorted in the front, and the classification directories and product names with small weights are sorted in the back, and a product sorting list is obtained through sorting, and the e-commerce platform displays the products through the obtained sorting list;

[0033] S3: E-commerce search processing: The e-commerce platform performs e-commerce search processing, wherein the e-commerce search engine first obtains the user's search terms, and performs Query preprocessing on the obtained user search terms, wherein the search engine first identifies the search terms during Query preprocessing, and performs processing based on the identification results, wherein the identification results show that the obtained search terms contain stop words, then the search engine removes the stop words, and the identification results show that the obtained search terms do not contain stop words, then the search terms are normalized, wherein the search engine performs special symbol filtering, case normalization, traditional Chinese to simplified Chinese, and full-width to half-width on the search terms, and after the normalization is completed, the search engine performs pinyin conversion and synonym replacement on the search terms, and simultaneously performs word segmentation, completion, rewriting, and error correction on the search terms to obtain the product name and the category directory where the product is located, obtains a new search weight through the obtained product name and category directory, sorts through the obtained search weight to obtain a new sorted list, and displays the products through the new sorted list;

[0034] S4: Constructing a personalized search processing model: The commodity platform automatically constructs a personalized search processing model, wherein the e-commerce platform first constructs a search processing model through a constructed commodity database when constructing the model, and after the search processing model is constructed, the e-commerce platform obtains the user's search records, purchase records, browsing records, and return records, and analyzes the obtained information, and obtains the user's purchase preferences through the analysis results, and increases the classification directory corresponding to the commodity and the search weight of the commodity through the obtained purchase preferences, and obtains the user's personalized commodity ranking through the obtained new classification directory and commodity search weight, and stores the obtained user's personalized commodity ranking into the search model to construct a personalized search processing model. After the personalized search processing model is constructed, the constructed personalized search processing model is trained by professionals, wherein during the training, the professionals preset the search commodity name, and obtain the commodity display arrangement order by inputting the preset commodity name, and the professionals view the order, make a judgment based on the viewing result, and process the judgment result, wherein the viewing result shows that the first place in the commodity ranking list is different from the preset search commodity name, then it is judged as The personalized search processing model is wrong. If the result of checking shows that the first name in the product sorting list is the same as the preset search product name, the personalized search processing model is judged to be correct. If the result of judging that the personalized search processing model is wrong, the training is continued, and the training result is judged. The search processing accuracy of the personalized search processing model is calculated by the judgment result. The training is stopped until the search processing accuracy of the personalized search processing model is higher than 98%. The personalized search processing model is used. If the result of judging that the personalized search processing model is correct, the constructed personalized search processing model is continued to be trained, and the training result is checked. The judgment is made by checking the result, and the processing is performed by the judgment result. If the result of checking is that the personalized search processing model is correct, the professional records the number of consecutive judgment results. If the result of checking is that the personalized search processing model is wrong, the professional calculates and obtains the search processing accuracy of the personalized search processing model. If the result of judging that the personalized search processing model is correct is higher than 100 times, the training is stopped, and the personalized search processing model is used.

[0035] S5: Subsequent search processing: The e-commerce platform performs subsequent search processing, wherein the subsequent search processing includes real-time updating of the retrieval database, pre-search guidance, and real-time hot word detection. The updating of the retrieval database is performed by the e-commerce platform to obtain the search records of all users of the platform in real time, and store the real-time obtained data. The e-commerce platform performs real-time retrieval to update the first search proportion and the second search proportion data in real time, and obtains a real-time commodity ranking list through real-time data update. The e-commerce platform displays the user's searched commodities through the obtained real-time commodity ranking list. During the pre-search guidance, the e-commerce platform obtains the commodity search hot words on the upper platform in real time, compares the obtained commodity search hot words with the user's purchase preferences, and processes the comparison results. , where the comparison result shows that there are overlapping elements between the obtained product search hot words and the user's purchase preferences, the e-commerce platform extracts the product search hot words containing the overlapping elements, and sorts the extracted product search hot words in the front row of the hot word search display area. The real-time hot word detection is performed by the e-commerce platform to detect the product hot words in real time and obtain the detection results. The e-commerce platform uses the obtained product hot words as product sub-labels, where the product sub-labels are equivalent to the product names when the user searches. The products are sorted according to the user's retrieval preferences and displayed through the sorting. At the same time, the e-commerce platform performs calculations to obtain the product transaction rate, and compares the obtained results with the product transaction rate data generated when the e-commerce platform closes the hot word search display area.

[0036] Embodiment 2

[0037] Reference Figure 1-4An e-commerce search processing system includes an acquisition module, a processing module, an establishment module, a construction module, a retrieval module, a sorting module, a display module, an e-commerce search processing module, a preprocessing module, a training module, and a subsequent search processing module. The acquisition module is connected to the processing module, the e-commerce search processing module, and the display module respectively. The processing module is connected to the establishment module, the establishment module is connected to the construction module, the construction module is connected to the retrieval module, the display module, and the training module respectively. The retrieval module is connected to the sorting module, the sorting module is connected to the display module and the preprocessing module respectively, the e-commerce search processing module is connected to the preprocessing module, and the training module is connected to the subsequent search processing module, wherein the subsequent search processing module includes a real-time update unit, a guide unit, and a detection unit, wherein the acquisition module is used to obtain the existing E-commerce product information, user search terms and new search weights, the processing module is used to process the acquired e-commerce product information, the establishment module is used to establish a product database framework using the acquired classification directory as a storage directory, the construction module is used to build a product database and a personalized search processing model, the retrieval module is used to search the retrieval database, the sorting module is used to sort the products according to the acquired search weights, the display module is used to display the products through the obtained sorting list, the e-commerce search processing module is used to perform e-commerce search processing, the pre-processing module is used to perform Query pre-processing on the acquired user search terms, the training module is used to train the constructed personalized search processing model, and the subsequent search processing module is used to perform real-time update of the retrieval database, pre-search guidance and real-time hot word detection;

[0038] The present invention also provides an e-commerce search processing method, comprising the following steps:

[0039] S1: Build a product database: Professionals build a product database;

[0040] S2: Searching: The e-commerce platform searches through a search database, wherein the search database is a database of all user search records stored by the e-commerce platform, and when searching, first obtains the search weights of different classification directories through searching, and uses the obtained search weights as the first search weights, and at the same time obtains the search weights of users for the same product name in different classification directories through searching, and uses the obtained search weights as the second search weights, sorts the products according to the obtained search weights, and displays the products through sorting, wherein when sorting, the e-commerce platform first performs preliminary sorting according to the first search weights, and after the preliminary sorting is completed, the e-commerce platform performs final sorting according to the second search weights, and when sorting, the classification directories and product names with large weights are sorted in the front, and the classification directories and product names with small weights are sorted in the back, and a product sorting list is obtained through sorting, and the e-commerce platform displays the products through the obtained sorting list;

[0041] S3: E-commerce search processing: The e-commerce platform performs e-commerce search processing, wherein the e-commerce search engine first obtains the user's search terms, and performs Query preprocessing on the obtained user search terms, wherein the search engine first identifies the search terms during Query preprocessing, and performs processing based on the identification results, wherein the identification results show that the obtained search terms contain stop words, then the search engine removes the stop words, and the identification results show that the obtained search terms do not contain stop words, then the search terms are normalized, wherein the search engine performs special symbol filtering, case normalization, traditional Chinese to simplified Chinese, and full-width to half-width on the search terms, and after the normalization is completed, the search engine performs pinyin conversion and synonym replacement on the search terms, and simultaneously performs word segmentation, completion, rewriting, and error correction on the search terms to obtain the product name and the category directory where the product is located, obtains a new search weight through the obtained product name and category directory, sorts through the obtained search weight to obtain a new sorted list, and displays the products through the new sorted list;

[0042] S4: Constructing a personalized search processing model: The commodity platform automatically constructs a personalized search processing model, wherein the e-commerce platform first constructs a search processing model through a constructed commodity database when constructing the model, and after the search processing model is constructed, the e-commerce platform obtains the user's search records, purchase records, browsing records, and return records, and analyzes the obtained information, and obtains the user's purchase preferences through the analysis results, and increases the classification directory corresponding to the commodity and the search weight of the commodity through the obtained purchase preferences, and obtains the user's personalized commodity ranking through the obtained new classification directory and commodity search weight, and stores the obtained user's personalized commodity ranking into the search model to construct a personalized search processing model. After the personalized search processing model is constructed, the constructed personalized search processing model is trained by professionals, wherein during the training, the professionals preset the search commodity name, and obtain the commodity display arrangement order by inputting the preset commodity name, and the professionals view the order, make a judgment based on the viewing result, and process the judgment result, wherein the viewing result shows that the first place in the commodity ranking list is different from the preset search commodity name, then it is judged as The personalized search processing model is wrong. If the result of checking shows that the first name in the product sorting list is the same as the preset search product name, the personalized search processing model is judged to be correct. If the result of judging that the personalized search processing model is wrong, the training is continued, and the training result is judged. The search processing accuracy of the personalized search processing model is calculated by the judgment result. The training is stopped until the search processing accuracy of the personalized search processing model is higher than 98%. The personalized search processing model is used. If the result of judging that the personalized search processing model is correct, the constructed personalized search processing model is continued to be trained, and the training result is checked. The judgment is made by checking the result, and the processing is performed by the judgment result. If the result of checking is that the personalized search processing model is correct, the professional records the number of consecutive judgment results. If the result of checking is that the personalized search processing model is wrong, the professional calculates and obtains the search processing accuracy of the personalized search processing model. If the result of judging that the personalized search processing model is correct is higher than 100 times, the training is stopped, and the personalized search processing model is used.

[0043] S5: Subsequent search processing: The e-commerce platform performs subsequent search processing, wherein the subsequent search processing includes real-time updating of the retrieval database, pre-search guidance, and real-time hot word detection. The updating of the retrieval database is performed by the e-commerce platform to obtain the search records of all users of the platform in real time, and store the real-time obtained data. The e-commerce platform performs real-time retrieval to update the first search proportion and the second search proportion data in real time, and obtains a real-time commodity ranking list through real-time data update. The e-commerce platform displays the user's searched commodities through the obtained real-time commodity ranking list. During the pre-search guidance, the e-commerce platform obtains the commodity search hot words on the upper platform in real time, compares the obtained commodity search hot words with the user's purchase preferences, and processes the comparison results. , where the comparison result shows that there are overlapping elements between the obtained product search hot words and the user's purchase preferences, the e-commerce platform extracts the product search hot words containing the overlapping elements, and sorts the extracted product search hot words in the front row of the hot word search display area. The real-time hot word detection is performed by the e-commerce platform to detect the product hot words in real time and obtain the detection results. The e-commerce platform uses the obtained product hot words as product sub-labels, where the product sub-labels are equivalent to the product names when the user searches. The products are sorted according to the user's retrieval preferences and displayed through the sorting. At the same time, the e-commerce platform performs calculations to obtain the product transaction rate, and compares the obtained results with the product transaction rate data generated when the e-commerce platform closes the hot word search display area.

[0044] Embodiment 3

[0045] Reference Figure 1-4An e-commerce search processing system includes an acquisition module, a processing module, an establishment module, a construction module, a retrieval module, a sorting module, a display module, an e-commerce search processing module, a preprocessing module, a training module, and a subsequent search processing module. The acquisition module is connected to the processing module, the e-commerce search processing module, and the display module respectively. The processing module is connected to the establishment module, the establishment module is connected to the construction module, the construction module is connected to the retrieval module, the display module, and the training module respectively. The retrieval module is connected to the sorting module, the sorting module is connected to the display module and the preprocessing module respectively, the e-commerce search processing module is connected to the preprocessing module, and the training module is connected to the subsequent search processing module, wherein the subsequent search processing module includes a real-time update unit, a guide unit, and a detection unit, wherein the acquisition module is used to obtain the existing E-commerce product information, user search terms and new search weights, the processing module is used to process the acquired e-commerce product information, the establishment module is used to establish a product database framework using the acquired classification directory as a storage directory, the construction module is used to build a product database and a personalized search processing model, the retrieval module is used to search the retrieval database, the sorting module is used to sort the products according to the acquired search weights, the display module is used to display the products through the obtained sorting list, the e-commerce search processing module is used to perform e-commerce search processing, the pre-processing module is used to perform Query pre-processing on the acquired user search terms, the training module is used to train the constructed personalized search processing model, and the subsequent search processing module is used to perform real-time update of the retrieval database, pre-search guidance and real-time hot word detection;

[0046] The present invention also provides an e-commerce search processing method, comprising the following steps:

[0047] S1: Constructing a commodity database: Professionals construct a commodity database, wherein before constructing the commodity database, the professionals first obtain the existing e-commerce commodity information on the e-commerce platform, wherein the e-commerce commodity information includes commodity name, commodity variety, commodity specification, commodity style, commodity color, commodity new technology, and commodity new material, and the professionals process the obtained e-commerce commodity information, wherein during the processing, a classification catalog is obtained by computer according to the existing commodity classification, and the obtained classification catalog is used as a storage catalog to establish a commodity database framework, and at the same time, the professionals store the obtained e-commerce commodity information through the commodity data framework to construct the commodity database, wherein the obtained e-commerce commodity information needs to be stored in the corresponding classification catalog during storage;

[0048] S2: E-commerce search processing: The e-commerce platform performs e-commerce search processing, wherein the e-commerce search engine first obtains the user's search terms, and performs Query preprocessing on the obtained user search terms, wherein the search engine first identifies the search terms during Query preprocessing, and performs processing based on the identification results, wherein the identification results show that the obtained search terms contain stop words, then the search engine removes the stop words, and the identification results show that the obtained search terms do not contain stop words, then the search terms are normalized, wherein the search engine performs special symbol filtering, case normalization, traditional Chinese to simplified Chinese, and full-width to half-width on the search terms, and after the normalization is completed, the search engine performs pinyin conversion and synonym replacement on the search terms, and simultaneously performs word segmentation, completion, rewriting, and error correction on the search terms to obtain the product name and the category directory where the product is located, obtains a new search weight through the obtained product name and category directory, sorts through the obtained search weight to obtain a new sorted list, and displays the products through the new sorted list;

[0049] S3: Constructing a personalized search processing model: The commodity platform automatically constructs a personalized search processing model, wherein the e-commerce platform first constructs a search processing model through the constructed commodity database. After the search processing model is constructed, the e-commerce platform obtains the user's search records, purchase records, browsing records and return records, and analyzes the obtained information. The user's purchase preference is obtained through the analysis result, and the classification directory corresponding to the commodity and the search weight of the commodity are increased through the obtained purchase preference. The user's personalized commodity ranking is obtained through the obtained new classification directory and commodity search weight, and the obtained user's personalized commodity ranking is stored in the search model to construct a personalized search processing model. After the personalized search processing model is constructed, the constructed personalized search processing model is trained by professionals. During the training, the professionals preset the search commodity name, and obtain the commodity display arrangement order by inputting the preset commodity name. The professionals view the order, make a judgment based on the viewing result, and process the judgment result. If the viewing result shows that the first place in the commodity ranking list is different from the preset search commodity name, it is judged as The personalized search processing model is wrong. If the result of checking shows that the first name in the product sorting list is the same as the preset search product name, the personalized search processing model is judged to be correct. If the result of judging that the personalized search processing model is wrong, the training is continued, and the training result is judged. The search processing accuracy of the personalized search processing model is calculated by the judgment result. The training is stopped until the search processing accuracy of the personalized search processing model is higher than 98%. The personalized search processing model is used. If the result of judging that the personalized search processing model is correct, the constructed personalized search processing model is continued to be trained, and the training result is checked. The judgment is made by checking the result, and the processing is performed by the judgment result. If the result of checking is that the personalized search processing model is correct, the professional records the number of consecutive judgment results. If the result of checking is that the personalized search processing model is wrong, the professional calculates and obtains the search processing accuracy of the personalized search processing model. If the result of judging that the personalized search processing model is correct is higher than 100 times, the training is stopped, and the personalized search processing model is used.

[0050] S4: Subsequent search processing: The e-commerce platform performs subsequent search processing, wherein the subsequent search processing includes real-time updating of the retrieval database, pre-search guidance, and real-time hot word detection. The updating of the retrieval database is performed by the e-commerce platform to obtain the search records of all users of the platform in real time, and store the real-time obtained data. The e-commerce platform performs real-time retrieval to update the first search proportion and the second search proportion data in real time, and obtains a real-time commodity ranking list through real-time data update. The e-commerce platform displays the user's searched commodities through the obtained real-time commodity ranking list. During the pre-search guidance, the e-commerce platform obtains the commodity search hot words of the upper platform in real time, compares the obtained commodity search hot words with the user's purchase preferences, and processes the comparison results. , where the comparison result shows that there are overlapping elements between the obtained product search hot words and the user's purchase preferences, the e-commerce platform extracts the product search hot words containing the overlapping elements, and sorts the extracted product search hot words in the front row of the hot word search display area. The real-time hot word detection is performed by the e-commerce platform to detect the product hot words in real time and obtain the detection results. The e-commerce platform uses the obtained product hot words as product sub-labels, where the product sub-labels are equivalent to the product names when the user searches. The products are sorted according to the user's retrieval preferences and displayed through the sorting. At the same time, the e-commerce platform performs calculations to obtain the product transaction rate, and compares the obtained results with the product transaction rate data generated when the e-commerce platform closes the hot word search display area.

[0051] Embodiment 4

[0052] Reference Figure 1-4An e-commerce search processing system includes an acquisition module, a processing module, an establishment module, a construction module, a retrieval module, a sorting module, a display module, an e-commerce search processing module, a preprocessing module, a training module, and a subsequent search processing module. The acquisition module is connected to the processing module, the e-commerce search processing module, and the display module respectively. The processing module is connected to the establishment module, the establishment module is connected to the construction module, the construction module is connected to the retrieval module, the display module, and the training module respectively. The retrieval module is connected to the sorting module, the sorting module is connected to the display module and the preprocessing module respectively, the e-commerce search processing module is connected to the preprocessing module, and the training module is connected to the subsequent search processing module, wherein the subsequent search processing module includes a real-time update unit, a guide unit, and a detection unit, wherein the acquisition module is used to obtain the existing E-commerce product information, user search terms and new search weights, the processing module is used to process the acquired e-commerce product information, the establishment module is used to establish a product database framework using the acquired classification directory as a storage directory, the construction module is used to build a product database and a personalized search processing model, the retrieval module is used to search the retrieval database, the sorting module is used to sort the products according to the acquired search weights, the display module is used to display the products through the obtained sorting list, the e-commerce search processing module is used to perform e-commerce search processing, the pre-processing module is used to perform Query pre-processing on the acquired user search terms, the training module is used to train the constructed personalized search processing model, and the subsequent search processing module is used to perform real-time update of the retrieval database, pre-search guidance and real-time hot word detection;

[0053] The present invention also provides an e-commerce search processing method, comprising the following steps:

[0054] S1: Constructing a commodity database: Professionals construct a commodity database, wherein before constructing the commodity database, the professionals first obtain the existing e-commerce commodity information on the e-commerce platform, wherein the e-commerce commodity information includes commodity name, commodity variety, commodity specification, commodity style, commodity color, commodity new technology, and commodity new material, and the professionals process the obtained e-commerce commodity information, wherein during the processing, a classification catalog is obtained by computer according to the existing commodity classification, and the obtained classification catalog is used as a storage catalog to establish a commodity database framework, and at the same time, the professionals store the obtained e-commerce commodity information through the commodity data framework to construct the commodity database, wherein the obtained e-commerce commodity information needs to be stored in the corresponding classification catalog during storage;

[0055] S2: Searching: The e-commerce platform searches through a search database, wherein the search database is a database of all user search records stored by the e-commerce platform, and when searching, first obtains the search weights of different classification directories through searching, and uses the obtained search weights as the first search weights, and at the same time obtains the search weights of users for the same product name in different classification directories through searching, and uses the obtained search weights as the second search weights, sorts the products according to the obtained search weights, and displays the products through sorting, wherein when sorting, the e-commerce platform first performs preliminary sorting according to the first search weights, and after the preliminary sorting is completed, the e-commerce platform performs final sorting according to the second search weights, and when sorting, the classification directories and product names with large weights are sorted in the front, and the classification directories and product names with small weights are sorted in the back, and a product sorting list is obtained through sorting, and the e-commerce platform displays the products through the obtained sorting list;

[0056] S3: E-commerce search processing: The e-commerce platform performs e-commerce search processing, wherein the e-commerce search engine first obtains the user's search terms, and performs Query preprocessing on the obtained user search terms, wherein the search engine first identifies the search terms during Query preprocessing, and performs processing based on the identification results, wherein the identification results show that the obtained search terms contain stop words, then the search engine removes the stop words, and the identification results show that the obtained search terms do not contain stop words, then the search terms are normalized, wherein the search engine performs special symbol filtering, case normalization, traditional Chinese to simplified Chinese, and full-width to half-width on the search terms, and after the normalization is completed, the search engine performs pinyin conversion and synonym replacement on the search terms, and simultaneously performs word segmentation, completion, rewriting, and error correction on the search terms to obtain the product name and the category directory where the product is located, obtains a new search weight through the obtained product name and category directory, sorts through the obtained search weight to obtain a new sorted list, and displays the products through the new sorted list;

[0057] S4: Constructing a personalized search processing model: The commodity platform automatically constructs a personalized search processing model, wherein the e-commerce platform first constructs a search processing model through a constructed commodity database, and after the search processing model is constructed, the e-commerce platform obtains the user's search records, purchase records, browsing records, and return records, and analyzes the obtained information, obtains the user's purchase preferences through the analysis results, increases the classification catalog corresponding to the commodity and the search weight of the commodity through the obtained purchase preferences, and obtains the user's personalized commodity ranking through the obtained new classification catalog and commodity search weight, and stores the obtained user's personalized commodity ranking into the search model to construct a personalized search processing model. After the personalized search processing model is constructed, professionals train the constructed personalized search processing model;

[0058] S5: Subsequent search processing: The e-commerce platform performs subsequent search processing, wherein the subsequent search processing includes real-time updating of the retrieval database, pre-search guidance, and real-time hot word detection. The updating of the retrieval database is performed by the e-commerce platform to obtain the search records of all users of the platform in real time, and store the real-time obtained data. The e-commerce platform performs real-time retrieval to update the first search proportion and the second search proportion data in real time, and obtains a real-time commodity ranking list through real-time data update. The e-commerce platform displays the user's searched commodities through the obtained real-time commodity ranking list. During the pre-search guidance, the e-commerce platform obtains the commodity search hot words on the upper platform in real time, compares the obtained commodity search hot words with the user's purchase preferences, and processes the comparison results. , where the comparison result shows that there are overlapping elements between the obtained product search hot words and the user's purchase preferences, the e-commerce platform extracts the product search hot words containing the overlapping elements, and sorts the extracted product search hot words in the front row of the hot word search display area. The real-time hot word detection is performed by the e-commerce platform to detect the product hot words in real time and obtain the detection results. The e-commerce platform uses the obtained product hot words as product sub-labels, where the product sub-labels are equivalent to the product names when the user searches. The products are sorted according to the user's retrieval preferences and displayed through the sorting. At the same time, the e-commerce platform performs calculations to obtain the product transaction rate, and compares the obtained results with the product transaction rate data generated when the e-commerce platform closes the hot word search display area.

[0059] Embodiment 5

[0060] Reference Figure 1-4An e-commerce search processing system includes an acquisition module, a processing module, an establishment module, a construction module, a retrieval module, a sorting module, a display module, an e-commerce search processing module, a preprocessing module, a training module, and a subsequent search processing module. The acquisition module is connected to the processing module, the e-commerce search processing module, and the display module respectively. The processing module is connected to the establishment module, the establishment module is connected to the construction module, the construction module is connected to the retrieval module, the display module, and the training module respectively. The retrieval module is connected to the sorting module, the sorting module is connected to the display module and the preprocessing module respectively, the e-commerce search processing module is connected to the preprocessing module, and the training module is connected to the subsequent search processing module, wherein the subsequent search processing module includes a real-time update unit, a guide unit, and a detection unit, wherein the acquisition module is used to obtain the existing E-commerce product information, user search terms and new search weights, the processing module is used to process the acquired e-commerce product information, the establishment module is used to establish a product database framework using the acquired classification directory as a storage directory, the construction module is used to build a product database and a personalized search processing model, the retrieval module is used to search the retrieval database, the sorting module is used to sort the products according to the acquired search weights, the display module is used to display the products through the obtained sorting list, the e-commerce search processing module is used to perform e-commerce search processing, the pre-processing module is used to perform Query pre-processing on the acquired user search terms, the training module is used to train the constructed personalized search processing model, and the subsequent search processing module is used to perform real-time update of the retrieval database, pre-search guidance and real-time hot word detection;

[0061] The present invention also provides an e-commerce search processing method, comprising the following steps:

[0062] S1: Constructing a commodity database: Professionals construct a commodity database, wherein before constructing the commodity database, the professionals first obtain the existing e-commerce commodity information on the e-commerce platform, wherein the e-commerce commodity information includes commodity name, commodity variety, commodity specification, commodity style, commodity color, commodity new technology, and commodity new material, and the professionals process the obtained e-commerce commodity information, wherein during the processing, a classification catalog is obtained by computer according to the existing commodity classification, and the obtained classification catalog is used as a storage catalog to establish a commodity database framework, and at the same time, the professionals store the obtained e-commerce commodity information through the commodity data framework to construct the commodity database, wherein the obtained e-commerce commodity information needs to be stored in the corresponding classification catalog during storage;

[0063] S2: Searching: The e-commerce platform searches through a search database, wherein the search database is a database of all user search records stored by the e-commerce platform, and when searching, first obtains the search weights of different classification directories through searching, and uses the obtained search weights as the first search weights, and at the same time obtains the search weights of users for the same product name in different classification directories through searching, and uses the obtained search weights as the second search weights, sorts the products according to the obtained search weights, and displays the products through sorting, wherein when sorting, the e-commerce platform first performs preliminary sorting according to the first search weights, and after the preliminary sorting is completed, the e-commerce platform performs final sorting according to the second search weights, and when sorting, the classification directories and product names with large weights are sorted in the front, and the classification directories and product names with small weights are sorted in the back, and a product sorting list is obtained through sorting, and the e-commerce platform displays the products through the obtained sorting list;

[0064] S3: E-commerce search processing: The e-commerce platform performs e-commerce search processing, wherein the e-commerce search engine first obtains the user's search terms, and performs Query preprocessing on the obtained user search terms, wherein the search engine first identifies the search terms during Query preprocessing, and performs processing based on the identification results, wherein the identification results show that the obtained search terms contain stop words, then the search engine removes the stop words, and the identification results show that the obtained search terms do not contain stop words, then the search terms are normalized, wherein the search engine performs special symbol filtering, case normalization, traditional Chinese to simplified Chinese, and full-width to half-width on the search terms, and after the normalization is completed, the search engine performs pinyin conversion and synonym replacement on the search terms, and simultaneously performs word segmentation, completion, rewriting, and error correction on the search terms to obtain the product name and the category directory where the product is located, obtains a new search weight through the obtained product name and category directory, sorts through the obtained search weight to obtain a new sorted list, and displays the products through the new sorted list;

[0065] S4: Constructing a personalized search processing model: The commodity platform automatically constructs a personalized search processing model, wherein the e-commerce platform first constructs a search processing model through a constructed commodity database when constructing the model, and after the search processing model is constructed, the e-commerce platform obtains the user's search records, purchase records, browsing records, and return records, and analyzes the obtained information, and obtains the user's purchase preferences through the analysis results, and increases the classification directory corresponding to the commodity and the search weight of the commodity through the obtained purchase preferences, and obtains the user's personalized commodity ranking through the obtained new classification directory and commodity search weight, and stores the obtained user's personalized commodity ranking into the search model to construct a personalized search processing model. After the personalized search processing model is constructed, the constructed personalized search processing model is trained by professionals, wherein during the training, the professionals preset the search commodity name, and obtain the commodity display arrangement order by inputting the preset commodity name, and the professionals view the order, make a judgment based on the viewing result, and process the judgment result, wherein the viewing result shows that the first place in the commodity ranking list is different from the preset search commodity name, then it is judged as The personalized search processing model is wrong. If the checking result shows that the first place in the product sorting list is the same as the preset search product name, the personalized search processing model is judged to be correct. If the judgment result is that the personalized search processing model is wrong, continue to train and judge the training result. Calculate the search processing accuracy of the personalized search processing model through the judgment result. Stop training until the search processing accuracy of the personalized search processing model is higher than 98%, and use the personalized search processing model. If the judgment result is that the personalized search processing model is correct, continue to train the constructed personalized search processing model, and check the training results. Make a judgment by checking the results and process by checking the results. If the checking result is that the personalized search processing model is correct, the professional will record the consecutive number of judgment results. If the checking result is that the personalized search processing model is wrong, the professional will calculate and obtain the search processing accuracy of the personalized search processing model. If the judgment result is that the personalized search processing model is correct for more than 100 consecutive times and the search processing accuracy of the personalized search processing model is higher than 98%, stop training and use the personalized search processing model.

[0066] The e-commerce search processing system and method in Embodiment 1, Embodiment 2, Embodiment 3, Embodiment 4 and Embodiment 5 were tested, and the following results were obtained:

[0067]

[0068] The e-commerce search processing method system and method obtained by Example 1, Example 2, Example 3, Example 4 and Example 5 have significantly improved the search processing accuracy and product transaction rate compared with the existing system and method, and Example 1 is the best example.

[0069] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. An e-commerce search processing method, characterized in that: The following steps are involved: S1: Build a product database: Professionals build a product database; S2: Search: The e-commerce platform searches through the search database; S3: e-commerce search processing: the e-commerce platform performs e-commerce search processing; S4: Building a personalized search processing model: The product platform automatically builds a personalized search processing model; S5: Subsequent search processing: The e-commerce platform performs subsequent search processing; In S1, a commodity database is constructed by professionals, wherein before constructing the commodity database, professionals first obtain the existing e-commerce commodity information on the e-commerce platform, wherein the e-commerce commodity information includes commodity name, commodity variety, commodity specification, commodity style, commodity color, commodity new technology, and commodity new material, and professionals process the obtained e-commerce commodity information, wherein during the processing, a classification catalog is obtained by computer according to the existing commodity classification, and the obtained classification catalog is used as a storage catalog to establish a commodity database framework, and at the same time, professionals store the obtained e-commerce commodity information through the commodity data framework to construct the commodity database, wherein the obtained e-commerce commodity information needs to be stored in the corresponding classification catalog during storage; In S2, the e-commerce platform searches through a search database, wherein the search database is a database of all user search records stored by the e-commerce platform, and when searching, the search weights of different classification directories are first obtained through searching, and the obtained search weights are used as the first search weights, and at the same time, the search weights of users for the same product name in different classification directories are obtained through searching, and the obtained search weights are used as the second search weights, and the products are sorted according to the obtained search weights, and the products are displayed through sorting, wherein when sorting, the e-commerce platform first performs preliminary sorting according to the first search weight, and after the preliminary sorting is completed, the e-commerce platform performs final sorting according to the second search weight, and when sorting, the classification directories and product names with large weights are sorted in the front, and the classification directories and product names with small weights are sorted in the back, and a product sorting list is obtained through sorting, and the e-commerce platform displays the products through the obtained sorting list; In S3, the e-commerce platform performs e-commerce search processing, wherein the e-commerce search engine first obtains the user's search terms when performing the e-commerce search processing, and performs Query preprocessing on the obtained user search terms, wherein the search engine first identifies the search terms when performing the Query preprocessing, and performs processing based on the identification results, wherein the identification results show that the obtained search terms contain stop words, then the search engine removes the stop words, and the identification results show that the obtained search terms do not contain stop words, then the search terms are normalized, wherein the normalization processing is that the search engine performs special symbol filtering, case normalization, traditional Chinese to simplified Chinese, and full-width to half-width on the search terms, and after the normalization processing is completed, the search engine performs pinyin conversion and synonym replacement on the search terms, and simultaneously performs word segmentation, completion, rewriting, and error correction on the search terms to obtain the product name and the category directory where the product is located, obtains a new search weight through the obtained product name and category directory, sorts through the obtained search weight to obtain a new sorted list, and displays the products through the new sorted list; In S4, the commodity platform automatically constructs a personalized search processing model, wherein when constructing the model, the e-commerce platform first constructs the search processing model through the constructed commodity database. After the search processing model is constructed, the e-commerce platform obtains the user's search records, purchase records, browsing records and return records, and analyzes the obtained information. The user's purchase preference is obtained through the analysis result, and the classification directory corresponding to the commodity and the search weight of the commodity are increased through the obtained purchase preference. The user's personalized commodity ranking is obtained through the obtained new classification directory and commodity search weight, and the obtained user's personalized commodity ranking is stored in the search model to construct a personalized search processing model. After the personalized search processing model is constructed, the constructed personalized search processing model is trained by professionals. During the training, the professionals preset the search commodity name, and obtain the commodity display arrangement order by inputting the preset commodity name. The professionals view the order, make a judgment based on the viewing result, and process the judgment result. If the viewing result shows that the first place in the commodity ranking list is different from the preset search commodity name, it is judged as a personalized search. If the processing model is wrong, and the checking result shows that the first name in the product sorting list is the same as the preset search product name, the personalized search processing model is judged to be correct. If the judgment result is that the personalized search processing model is wrong, the training is continued, and the training result is judged. The search processing accuracy of the personalized search processing model is calculated by the judgment result. The training is stopped until the search processing accuracy of the personalized search processing model is higher than 98%. The personalized search processing model is used. If the judgment result is that the personalized search processing model is correct, the constructed personalized search processing model is continued to be trained, and the training result is checked. The judgment is made by checking the result, and the processing is performed according to the judgment result. If the checking result is that the personalized search processing model is correct, the professional records the number of consecutive judgment results. If the checking result is that the personalized search processing model is wrong, the professional calculates and obtains the search processing accuracy of the personalized search processing model. If the judgment result is that the personalized search processing model is correct for more than 100 consecutive times and the search processing accuracy of the personalized search processing model is higher than 98%, the training is stopped, and the personalized search processing model is used. In S5, the e-commerce platform performs subsequent search processing, wherein the subsequent search processing includes real-time updating of the retrieval database, pre-search guidance, and real-time hot word detection. The updating of the retrieval database is performed by the e-commerce platform to obtain the search records of all users of the platform in real time, and store the real-time obtained data. The e-commerce platform performs real-time retrieval to update the first search proportion and the second search proportion data in real time, and obtains a real-time commodity ranking list through real-time data update. The e-commerce platform displays the user's searched commodities through the obtained real-time commodity ranking list. During the pre-search guidance, the e-commerce platform obtains the commodity search hot words of the upper platform in real time, compares the obtained commodity search hot words with the user's purchase preferences, and processes the comparison results. If the comparison results show that the obtained product search hot words have overlapping elements with the user's purchase preferences, the e-commerce platform will extract the product search hot words containing the overlapping elements, and sort the extracted product search hot words in the front row of the hot word search display area. The real-time hot word detection is performed by the e-commerce platform to detect the product hot words in real time and obtain the detection results. The e-commerce platform uses the obtained product hot words as product sub-labels, wherein the product sub-label is equivalent to the product name when the user searches. The products are sorted according to the user's retrieval preferences and displayed according to the sorting. At the same time, the e-commerce platform performs calculations to obtain the product transaction rate, and compares the obtained results with the product transaction rate data generated when the e-commerce platform closes the hot word search display area.

2. An e-commerce search processing system, which is used to implement an e-commerce search processing method as described in claim 1, comprising an acquisition module, a processing module, a building module, a retrieval module, a sorting module, a display module, an e-commerce search processing module, a pre-processing module, a training module, and a subsequent search processing module, characterized in that: The acquisition module is connected to the processing module, the e-commerce search processing module, and the display module respectively; the processing module is connected to the establishment module, the establishment module is connected to the construction module, the construction module is connected to the retrieval module, the display module, and the training module respectively; the retrieval module is connected to the sorting module, the sorting module is connected to the display module and the preprocessing module respectively; the e-commerce search processing module is connected to the preprocessing module; the training module is connected to the subsequent search processing module, wherein the subsequent search processing module includes a real-time update unit, a guiding unit, and a detection unit; the acquisition module is used to obtain the existing e-commerce product information, the user's search terms, and the new search weight of the e-commerce platform; the processing module is used to obtain the e-commerce product information The establishment module is used to establish a commodity database framework by using the acquired classification directory as a storage directory, the construction module is used to construct a commodity database and a personalized search processing model, the retrieval module is used to search the retrieval database, the sorting module is used to sort the commodities according to the acquired search weights, the display module is used to display the commodities through the acquired sorting list, the e-commerce search processing module is used to perform e-commerce search processing, the pre-processing module is used to perform Query pre-processing on the acquired user search terms, the training module is used to train the constructed personalized search processing model, and the subsequent search processing module is used to perform real-time update of the retrieval database, pre-search guidance, and real-time hot word detection.