CN114820091A - Intelligent non-standard product classification system - Google Patents
Intelligent non-standard product classification system Download PDFInfo
- Publication number
- CN114820091A CN114820091A CN202110113218.8A CN202110113218A CN114820091A CN 114820091 A CN114820091 A CN 114820091A CN 202110113218 A CN202110113218 A CN 202110113218A CN 114820091 A CN114820091 A CN 114820091A
- Authority
- CN
- China
- Prior art keywords
- module
- standard
- picture
- attribute
- searched
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0641—Shopping interfaces
- G06Q30/0643—Graphical representation of items or shoppers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Library & Information Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Databases & Information Systems (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a non-standard product intelligent classification system, which comprises: the database module is used for receiving and recording preset non-standard information containing the incidence relation between the categories and the attributes; the data analysis processing module is used for analyzing the attributes of the non-standard articles to be searched, and comprises the steps of extracting the attributes of the non-standard articles to be searched and calculating the attribute proportion; the data matching processing module is used for matching the attribute of the non-standard object to be searched with the attribute of a preset non-standard object; the classification module is used for classifying the categories of the non-standard articles to be searched; the input module is used for inputting preset non-standard product information and non-standard product information to be searched; and the output module is used for outputting the matched preset non-standard information list. When a merchant does not put a certain commodity on the shelf, a buyer can search only by uploading a commodity photo, the system automatically refines and analyzes the commodity in the photo, then the analyzed picture is matched with the merchant commodity stored in the system, and the system automatically classifies the same or similar commodity catalog or commodity.
Description
Technical Field
The invention belongs to the technical field of search classification, and particularly relates to an intelligent non-standard product classification system.
Background
With the development of information technology, online shopping has become a main shopping mode, and the online massive commodities make the classification of the commodities into new problems. Traditional methods of sorting by brand and style have failed to meet the increasing demand for online shopping. The consumers generally search for their needs in online shopping to reduce the number of products to be browsed. Even so, there are still a large number of items entering the search results in some large online shopping malls. The consumers find out the specific type of commodities from thousands or even tens of thousands of searched commodities, such as a sea fishing needle, and the page-by-page turning shopping experience is unacceptable to the consumers.
The search technology for specific commodities that has been applied at present is based on the search of the text information of the commodities, specifically, the provider of the information inputs various information of the commodities in text form, and the consumer limits the content of the search by text during shopping and selects a specific commodity from the search results for detailed reference. The searching method has the disadvantages that a great deal of information such as shapes, textures, colors and the like contained in the pictures of the commodities is completely ignored, and the intuitive information provides great convenience for consumers in the online shopping process and plays a very important role in the final decision purchase of the customers.
Disclosure of Invention
Aiming at the defects in the prior art, the invention aims to provide an intelligent non-standard product classification system to solve the problems in the background art, so that a user can search, classify and match through uploading pictures, and can quickly obtain a commodity with high similarity to the retrieved pictures.
In order to achieve the purpose, the invention provides the following technical scheme: a non-standard intelligent classification system comprising:
the server side comprises a database module, a data analysis processing module and a data matching processing module, wherein the database module is used for receiving and recording preset non-standard information containing the incidence relation between the categories and the attributes;
the data analysis processing module is used for analyzing the attributes of the non-standard articles to be searched, and comprises the steps of extracting the attributes of the non-standard articles to be searched and calculating the attribute ratio;
the data matching processing module is used for matching the attribute of the non-standard article to be searched with the attribute of a preset non-standard article and judging whether the matching result is greater than or equal to a preset threshold value or not;
the classification module is used for classifying, and classifying the categories of the non-standard articles to be searched when the judgment is larger than or equal to a preset threshold value;
the input module is used for inputting preset non-standard product information and non-standard product information to be searched by a user;
the output module is used for outputting a preset non-standard article information list matched with the attributes of the non-standard articles to be searched;
the output end of the input module is respectively connected with the input end of the database module and the input end of the data analysis processing module, the output end of the database module and the output end of the data analysis processing module are respectively connected with the input end of the data matching processing module, the output end of the data matching processing module is respectively connected with the input end of the output module and the input end of the classification module, and the output end of the classification module is connected with the input end of the database module.
By adopting the system, when a buyer has a demand for purchasing the commodity and a merchant does not put on the commodity, the buyer only needs to upload the character information or the photo of the commodity through the input module, particularly the uploaded picture, the system can automatically refine and analyze the commodity in the picture, the commodity in the picture is matched with the commodity stored in the system database after analysis, and the system automatically classifies the same or similar commodity attribute or commodity and outputs the commodity. In addition, the system also has the function of automatically recording the pictures to be retrieved and classifying the pictures to enrich the database, thereby facilitating the classification and updating of the subsequent database.
Further, the data analysis processing module comprises a semantic analysis processing module and an image analysis processing module;
the picture analysis processing module comprises a picture preprocessing module, a picture identification processing module and a picture feature extraction module;
the picture preprocessing module comprises mode acquisition, analog-to-digital conversion, filtering, fuzzy elimination, noise reduction and geometric distortion correction;
the image recognition processing module is used for recognizing the preprocessed non-standard image by utilizing a pre-trained neural network;
the picture feature extraction module is used for extracting picture features of the identified non-standard picture by utilizing a pre-trained neural network;
the output end of the input module is respectively connected with the input end of the semantic analysis processing module and the input end of the picture preprocessing module, the output end of the semantic analysis processing module is connected with the input end of the data matching processing module, the output end of the picture preprocessing module is connected with the input end of the picture recognition processing module, and the output end of the picture recognition processing module is connected with the input end of the picture feature extraction module.
In the above scheme, the semantic analysis processing module is convenient for analysis, search and matching according to the character information input by the buyer, the picture analysis processing module is convenient for analysis, search and matching according to the picture information input by the buyer, and the picture preprocessing module is convenient for correcting the picture uploaded by the buyer so as to facilitate subsequent picture feature identification and feature extraction and lay a foundation for subsequent matching accuracy.
Further, the picture features include picture characters, picture colors, picture shapes, picture textures and the like.
Furthermore, the data analysis processing module also comprises an attribute definition module which is used for acquiring semantic features corresponding to the extracted picture features and defining the semantic features as attributes of the non-standard articles to be searched;
the input end of the attribute definition module is connected with the output end of the picture feature extraction module.
In the scheme, the attribute definition module is convenient for carrying out text description on the picture characteristics and endowing the picture characteristics with the attribute characteristics of the non-standard object to be searched, namely, a picture-text conversion process, is convenient for picture characteristic comparison, converts the picture characteristic comparison into the attribute characteristic comparison, and expands the picture characteristic comparison range.
Further, the data analysis processing module also comprises an attribute proportion calculation module and an attribute sorting module,
the attribute proportion calculation module is used for predicting the attributes of the non-standard articles to be searched and obtaining at least one prediction score of at least one attribute of the non-standard articles to be searched;
the attribute sorting module is used for sorting at least one prediction score of at least one attribute of the non-standard object to be searched from large to small to obtain the attribute of the non-standard object to be searched sorted from top to bottom;
the input end of the attribute proportion calculation module is connected with the output end of the attribute definition module, and the input end of the attribute sorting module is connected with the output end of the attribute proportion calculation module.
In the scheme, the attribute ratio calculation module and the attribute sorting module are convenient for sorting the picture characteristics so as to output the preset non-standard product with high similarity.
Further, the data matching processing module comprises a non-standard article comparison module, a similarity sorting module and a matching judgment module;
the non-standard article comparison module is used for comparing the attributes of the non-standard articles to be searched which are sorted from top to bottom with the attributes of the non-standard articles preset in the database module to obtain at least one similarity degree value;
the similarity sorting module is used for sorting the similarity degree values from large to small;
the matching judgment module is used for comparing the similarity degree values sorted from large to small with a preset threshold value, and sorting the preset non-standard articles of which the similarity degree values are larger than or equal to the preset threshold value from top to bottom according to the difference value to serve as matched preset non-standard articles;
the input end of the non-standard article comparison module is connected with the output end of the attribute sorting module, the input end of the similar sorting module is connected with the output end of the non-standard article comparison module, the input end of the matching judgment module is connected with the output end of the similar sorting module, and the input end of the output module is connected with the output end of the matching judgment module.
In the scheme, the setting of the similarity sorting module facilitates the high-priority matching of the similarity degree numerical value, and improves the matching efficiency.
Further, the similarity degree value is a single similarity degree value of the sorted first attributes or a combined similarity degree value of the sorted attributes from top to bottom. The preset can be the initial setting of the system or the later setting according to the requirement by presetting whether the set sequencing is a single similarity degree value or a combined similarity degree value.
Further, the classification module comprises an acquisition module, a classification definition module and a recording module,
the acquisition module is used for acquiring the classification category of the preset non-standard article with the maximum similarity degree value;
the classification definition module is used for defining classification categories as the classification categories of the non-standard articles to be searched;
the receiving and recording module is used for receiving and recording the non-standard articles to be searched with the classification categories;
the input end of the acquisition module is connected with the output end of the matching judgment module, the input end of the classification definition module is connected with the output end of the acquisition module, and the input end of the recording module is connected with the output end of the classification definition module.
In the scheme, the classification module is arranged, so that the system has the function of automatically recording the pictures to be retrieved and classifies the pictures to enrich the database, and the subsequent database classification and updating are facilitated.
Further, the information comprises text information and picture information of non-standard articles, and the information comprises non-standard article information actively uploaded by a user and passively uploaded non-standard article information triggered by actions of searching, browsing, purchasing or collecting of the user.
Further, the categories include clothing, shoe bags, mothers and babies, beauty cosmetics, general goods, food, fruits, sports outdoors, cell phone numbers, household appliances, medical health, home textiles, furniture home decoration, book entertainment and the like.
Further, the attributes include color, material, quality, type, style, brand, origin, and the like.
Further, the users include sellers and buyers.
Further, the preset non-standard product information list output by the output module and matched with the attributes of the non-standard products to be searched comprises preset non-standard product pictures and merchant links.
The invention has the following beneficial effects:
the invention relates to a non-standard product intelligent classification system, when a buyer has a demand for purchasing a product and a merchant does not put the product on shelf, the buyer only needs to upload the character information or the picture of the product through an input module, particularly the uploaded picture, the system can automatically refine and analyze the product in the picture, match the product with the product stored in a system database after analysis, and automatically classify the product attribute or the product which is the same as or similar to the product attribute or the product and output the product. In addition, the system also has the function of automatically recording the pictures to be retrieved and classifying the pictures to enrich the database, thereby facilitating the classification and updating of the subsequent database.
Drawings
FIG. 1 is a block diagram of a non-standard intelligent classification system according to the present invention;
FIG. 2 is a schematic diagram of the internal module structure of the data analysis processing module according to the present invention;
FIG. 3 is a schematic diagram of the internal module structure of the image analysis processing module according to the present invention;
FIG. 4 is a schematic diagram of the internal module structure of the data matching processing module and the classification module according to the present invention.
In the figure: 1. a server side; 11. a database module; 12. a data analysis processing module; 121. a semantic analysis processing module; 122. a picture analysis processing module; 1221. a picture preprocessing module; 1222. a picture identification processing module; 1223. a picture feature extraction module; 123. an attribute definition module; 124. an attribute proportion calculation module; 125. an attribute sorting module; 13. a data matching processing module; 131. a non-standard article comparison module; 132. a similarity sorting module; 133. a matching judgment module; 2. a classification module; 21. an acquisition module; 22. a classification definition module; 23. a recording module; 3. a recording module; 4. and an output module.
Detailed Description
The present invention will be described in detail below with reference to the accompanying drawings and examples.
Example 1
An intelligent non-standard product classification system, as shown in fig. 1, comprises:
the system comprises a server end 1, wherein the server end 1 comprises a database module 11, a data analysis processing module 12 and a data matching processing module 13, wherein the database module 11 is used for receiving and recording preset non-standard information containing the incidence relation between categories and attributes;
the data analysis processing module 12 is used for analyzing the attributes of the non-standard articles to be searched, including extracting the attributes of the non-standard articles to be searched and calculating the attribute ratio;
the data matching processing module 13 is configured to match the attribute of the non-standard object to be searched with the attribute of a preset non-standard object, and determine whether a matching result is greater than or equal to a preset threshold;
the classification module 2 is used for classifying, and when the judgment result is greater than or equal to a preset threshold value, classifying the categories of the non-standard articles to be searched;
the input module 3 is used for inputting preset non-standard product information and non-standard product information to be searched by a user;
the output module 4 is used for outputting a preset non-standard article information list matched with the attributes of the non-standard articles to be searched;
the output of type-in module 3 links to each other with database module 11's input and data analysis processing module 12's input respectively, database module 11's output and data analysis processing module 12's output all link to each other with data matching processing module 13's input, data matching processing module 13's output links to each other with output module 4's input and classification module 2's input respectively, classification module 2's output links to each other with database module 11's input.
The information comprises non-standard information actively uploaded by a user and passively uploaded non-standard information triggered by user searching, browsing, purchasing or collecting behaviors; the categories comprise clothes, shoe bags, mothers and babies, beauty cosmetics, general goods, food, fruits, sports outdoors, mobile phone numbers, household appliances, medical health, home textiles, furniture home decoration, book entertainment and the like; attributes include color, material, quality, type, style, brand, place of origin, etc.; users include sellers and buyers; the preset non-standard article information list which is output by the output module 4 and matched with the attribute of the non-standard article to be searched comprises a preset non-standard article picture and a merchant link.
By adopting the system, when a buyer has a demand for purchasing a commodity and a merchant does not put on the commodity, the buyer only needs to upload the character information or the photo of the commodity through the input module 3, particularly the uploaded picture, the system can automatically refine and analyze the commodity in the picture, match the analyzed commodity with the commodity stored in the system database, and automatically classify the same or similar commodity attribute or commodity and output the commodity. In addition, the system also has the function of automatically recording the pictures to be retrieved and classifying the pictures to enrich the database, thereby facilitating the classification and updating of the subsequent database.
Specifically, as shown in fig. 2, the data analysis processing module 12 includes a semantic analysis processing module 121, an image analysis processing module 122, an attribute definition module 123, an attribute proportion calculation module 124, and an attribute sorting module 125, where the attribute definition module 123 is configured to obtain semantic features corresponding to the extracted image features, and define the semantic features as attributes of the non-standard object to be searched, where specifically, the image features include image characters, image colors, image shapes, image textures, and the like; the attribute ratio calculation module 124 is used for predicting the attributes of the non-standard object to be searched to obtain at least one prediction score of at least one attribute of the non-standard object to be searched, the prediction score is calculated according to the characteristics of the picture, specifically, the characters of the picture are set to be a certain score, and the number of the picture color, the picture shape, the picture texture and the like are evaluated according to the area size of the picture occupied by the picture; the attribute sorting module 125 is configured to sort at least one prediction score of at least one attribute of the non-standard object to be searched from large to small, and obtain attributes sorted from top to bottom of the non-standard object to be searched;
the output end of the entry module 3 is connected to the input end of the semantic analysis processing module 121 and the input end of the picture analysis processing module 122, the output end of the semantic analysis processing module 121 is connected to the input end of the data matching processing module 13, the output end of the picture analysis processing module 122 is connected to the input end of the attribute definition module 123, the output end of the attribute definition module 123 is connected to the input end of the attribute proportion calculation module 124, and the output end of the attribute proportion calculation module 124 is connected to the input end of the attribute sorting module 125.
In the above scheme, the semantic analysis processing module 121 facilitates analysis, search and matching according to text information input by a buyer, the picture analysis processing module 122 facilitates analysis, search and matching according to picture information input by the buyer, the attribute definition module 123 facilitates text description of picture features, and gives attribute features of a non-standard object to be searched, namely, a picture-text conversion process, which facilitates picture feature comparison, converts the picture feature comparison into attribute feature comparison, and expands the picture feature comparison range. The attribute proportion calculation module 124 and the attribute sorting module 125 facilitate sorting the picture features so as to output the preset non-standard articles with high similarity.
Specifically, as shown in fig. 3, the picture analysis processing module 122 includes a picture preprocessing module 1221, a picture identification processing module 1222, and a picture feature extraction module 1223;
the picture preprocessing module 1221 includes mode acquisition, analog-to-digital conversion, filtering, blur elimination, noise reduction, and geometric distortion correction;
the picture recognition processing module 1222 is configured to recognize the preprocessed non-standard picture by using a pre-trained neural network;
the image feature extraction module 1223 is configured to perform image feature extraction on the identified non-standard image by using a pre-trained neural network;
the output end of the input module 3 is connected to the input end of the picture preprocessing module 1221, the output end of the picture preprocessing module 1221 is connected to the input end of the picture recognition processing module 1222, the output end of the picture recognition processing module 1222 is connected to the input end of the picture feature extraction module 1223, and the output end of the picture feature extraction module 1223 is connected to the input end of the attribute definition module 123.
In the above scheme, the image preprocessing module 1221 facilitates correction of the image uploaded by the buyer, so as to facilitate subsequent image feature identification and feature extraction, and lay a foundation for subsequent matching accuracy.
Specifically, as shown in fig. 4, the data matching processing module 13 includes a non-standard comparison module 131, a similarity sorting module 132, and a matching judgment module 133;
the non-standard article comparison module 131 is configured to compare the attributes of the non-standard articles to be searched sorted from top to bottom with the attributes of the non-standard articles preset in the database module 11 to obtain at least one similarity degree value;
a similarity sorting module 132, configured to sort the similarity degree values from large to small;
the matching judgment module 133 is configured to compare the similarity degree values sorted from large to small with a preset threshold, and sort the preset non-standard articles with the similarity degree values larger than or equal to the preset threshold from top to bottom according to the difference values to serve as matched preset non-standard articles;
the input end of the non-standard article comparison module 131 is connected with the output end of the attribute sorting module 125, the input end of the similar sorting module 132 is connected with the output end of the non-standard article comparison module 131, the input end of the matching judgment module 133 is connected with the output end of the similar sorting module 132, and the output end of the matching judgment module 133 is connected with the input end of the output module 4.
In the above scheme, the setting of the similarity sorting module 132 facilitates the priority matching with a high similarity degree value, and improves the matching efficiency.
Further, the similarity value is a single similarity value of the first attribute or a combined similarity value of the attributes sorted from top to bottom. The preset can be the initial setting of the system or the later setting according to the requirement by presetting whether the set sequencing is a single similarity degree value or a combined similarity degree value.
As shown in fig. 4 in particular, the classification module 2 includes an acquisition module 21, a classification definition module 22 and a listing module 23,
the obtaining module 21 is configured to obtain a classification category of the preset non-standard article with the largest similarity degree value;
the classification definition module 22 is used for defining the classification category as the classification category of the non-standard article to be searched;
the receiving and recording module is used for receiving and recording the non-standard articles to be searched with the classification categories;
the input end of the acquisition module 21 is connected to the output end of the matching judgment module 133, the input end of the classification definition module 22 is connected to the output end of the acquisition module 21, and the input end of the recording module 23 is connected to the output end of the classification definition module 22.
In the above scheme, the classification module 2 is arranged to enable the system to have the function of automatically recording the pictures to be retrieved and classify the pictures to enrich the database, so that the subsequent database classification and update are facilitated.
Finally, it should be noted that the above embodiments are only used for illustrating the technical solutions of the embodiments of the present invention and not for limiting the same, and although the embodiments of the present invention are described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions can be made on the technical solutions of the embodiments of the present invention, and these modifications or equivalent substitutions cannot make the modified technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. A non-standard intelligent classification system is characterized by comprising:
the server side comprises a database module, a data analysis processing module and a data matching processing module, wherein the database module is used for receiving and recording preset non-standard information containing the incidence relation between the categories and the attributes;
the data analysis processing module is used for analyzing the attributes of the non-standard articles to be searched, and comprises the steps of extracting the attributes of the non-standard articles to be searched and calculating the attribute ratio;
the data matching processing module is used for matching the attribute of the non-standard article to be searched with the attribute of a preset non-standard article and judging whether the matching result is greater than or equal to a preset threshold value or not;
the classification module is used for classifying, and classifying the categories of the non-standard articles to be searched when the judgment is larger than or equal to a preset threshold value;
the input module is used for inputting preset non-standard product information and non-standard product information to be searched by a user;
the output module is used for outputting a preset non-standard article information list matched with the attribute of the non-standard article to be searched;
the output end of the input module is respectively connected with the input end of the database module and the input end of the data analysis processing module, the output end of the database module and the output end of the data analysis processing module are respectively connected with the input end of the data matching processing module, the output end of the data matching processing module is respectively connected with the input end of the output module and the input end of the classification module, and the output end of the classification module is connected with the input end of the database module.
2. The intelligent non-standard article classification system according to claim 1,
the data analysis processing module comprises a semantic analysis processing module and an image analysis processing module;
the picture analysis processing module comprises a picture preprocessing module, a picture identification processing module and a picture feature extraction module;
the picture preprocessing module comprises filtering, blur elimination, noise reduction and geometric distortion correction;
the image recognition processing module is used for recognizing the preprocessed non-standard image by utilizing a pre-trained neural network;
the picture feature extraction module is used for extracting picture features of the identified non-standard picture by utilizing a pre-trained neural network;
the output end of the input module is respectively connected with the input end of the semantic analysis processing module and the input end of the picture preprocessing module, the output end of the semantic analysis processing module is connected with the input end of the data matching processing module, the output end of the picture preprocessing module is connected with the input end of the picture recognition processing module, and the output end of the picture recognition processing module is connected with the input end of the picture feature extraction module.
3. The intelligent non-standard article classification system according to claim 2, wherein the picture features include picture characters, picture colors, picture shapes and picture textures.
4. The intelligent non-standard product classification system according to claim 2, wherein the data analysis processing module further comprises an attribute definition module, configured to obtain semantic features corresponding to the extracted picture features, and define the semantic features as attributes of the non-standard products to be searched;
the input end of the attribute definition module is connected with the output end of the picture feature extraction module.
5. The intelligent non-standard article classification system according to claim 4, wherein the data analysis processing module further comprises an attribute proportion calculation module and an attribute sorting module,
the attribute proportion calculation module is used for predicting the attributes of the non-standard articles to be searched and obtaining at least one prediction score of at least one attribute of the non-standard articles to be searched;
the attribute sorting module is used for sorting at least one prediction score of at least one attribute of the non-standard object to be searched from large to small to obtain the attribute of the non-standard object to be searched sorted from top to bottom;
the input end of the attribute proportion calculation module is connected with the output end of the attribute definition module, and the input end of the attribute sorting module is connected with the output end of the attribute proportion calculation module.
6. The intelligent non-standard product classification system according to claim 5, wherein the data matching processing module comprises a non-standard product comparison module, a similarity sorting module and a matching judgment module;
the non-standard article comparison module is used for comparing the attributes of the non-standard articles to be searched which are sorted from top to bottom with the attributes of the non-standard articles preset in the database module to obtain at least one similarity degree value;
the similarity sorting module is used for sorting the similarity degree values from large to small;
the matching judgment module is used for comparing the similarity degree values sorted from large to small with a preset threshold value, and sorting the preset non-standard articles of which the similarity degree values are larger than or equal to the preset threshold value from top to bottom according to the difference values to serve as matched preset non-standard articles;
the input end of the non-standard article comparison module is connected with the output end of the attribute sorting module, the input end of the similar sorting module is connected with the output end of the non-standard article comparison module, the input end of the matching judgment module is connected with the output end of the similar sorting module, and the input end of the output module is connected with the output end of the matching judgment module.
7. The intelligent non-standard article classification system according to claim 6, wherein the similarity degree value is a single similarity degree value sorting the first attribute or a combined similarity degree value sorting the attributes from top to bottom.
8. The non-standard intelligent classification system according to claim 6, wherein the classification module comprises an acquisition module, a classification definition module and a listing module,
the acquisition module is used for acquiring the classification category of the preset non-standard article with the maximum similarity degree value;
the classification definition module is used for defining classification categories as the classification categories of the non-standard articles to be searched;
the receiving and recording module is used for receiving and recording the non-standard articles to be searched with the classification categories;
the input end of the acquisition module is connected with the output end of the matching judgment module, the input end of the classification definition module is connected with the output end of the acquisition module, and the input end of the recording module is connected with the output end of the classification definition module.
9. The intelligent non-standard product classification system according to claim 1, wherein the information comprises text information and picture information of non-standard products, and the information comprises non-standard product information actively uploaded by a user and non-standard product information passively uploaded triggered by user search, browse, purchase or collection.
10. The non-standard intelligent classification system according to claim 1, wherein the categories include clothing, shoe bags, mothers and babies, beauty cosmetics, department goods, food, fruits, sports outdoors, cell phones, digital products, household appliances, medical health, home textiles, furniture home decoration and book entertainment, and the attributes include color, material, quality, type, style, brand and place of production.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110113218.8A CN114820091A (en) | 2021-01-27 | 2021-01-27 | Intelligent non-standard product classification system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110113218.8A CN114820091A (en) | 2021-01-27 | 2021-01-27 | Intelligent non-standard product classification system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114820091A true CN114820091A (en) | 2022-07-29 |
Family
ID=82523749
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110113218.8A Pending CN114820091A (en) | 2021-01-27 | 2021-01-27 | Intelligent non-standard product classification system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114820091A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115861986A (en) * | 2022-12-21 | 2023-03-28 | 浙江由由科技有限公司 | Non-standard product intelligent identification and loss prevention method based on supermarket self-service checkout system |
-
2021
- 2021-01-27 CN CN202110113218.8A patent/CN114820091A/en active Pending
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115861986A (en) * | 2022-12-21 | 2023-03-28 | 浙江由由科技有限公司 | Non-standard product intelligent identification and loss prevention method based on supermarket self-service checkout system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107861972B (en) | Method and equipment for displaying full commodity result after user inputs commodity information | |
CN107748754B (en) | Knowledge graph perfecting method and device | |
KR20190029567A (en) | Method for recommending a product using style feature | |
CN110110181A (en) | A kind of garment coordination recommended method based on user styles and scene preference | |
KR20210098884A (en) | A method of providing a fashion item recommendation service using a body shape and purchase history | |
US20200257976A1 (en) | Algorithmic apparel recommendation | |
AU2010256641A1 (en) | System and method for learning user genres and styles and matching products to user preferences | |
CN112102037A (en) | Live E-commerce platform commodity content intelligent pushing management system based on big data | |
CN108446931B (en) | Garment enterprise management system and fitting feedback method based on same | |
US11037071B1 (en) | Cross-category item associations using machine learning | |
KR101992988B1 (en) | An online shopping mall system recommending apparel materials using dynamic learning method | |
CN107256513A (en) | Method and device is recommended in a kind of collocation of object | |
CN111986007A (en) | Method for commodity aggregation and similarity calculation | |
CN107861970A (en) | A kind of commodity picture searching method and device | |
CN110413825A (en) | Clap recommender system in street towards fashion electric business | |
Tahir et al. | E-commerce platform based on Machine Learning Recommendation System | |
CN114820091A (en) | Intelligent non-standard product classification system | |
US20210390430A1 (en) | Machine Learning System and Method for Garment Recommendation | |
CN112508114A (en) | Intelligent clothing recommendation system and method | |
CN112380450A (en) | Commodity library retrieval system based on cloud inventory | |
KR102495868B1 (en) | Fashion-related customized perfume recommendation system using ai | |
KR20210041733A (en) | Method, apparatus and computer program for fashion item recommendation | |
CN116091162A (en) | E-commerce commodity recommendation method and device | |
CN115392993A (en) | Big data intelligent management system based on live video commodity storage show | |
Park et al. | Clothing classification using CNN and shopping mall search system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |