Feature analysis method and device of intelligent interaction electronic product, electronic equipment, storage medium and program product

Abstract

The invention provides a feature analysis method and device of an intelligent interaction electronic product, electronic equipment, a storage medium and a program product. The feature analysis method comprises the following steps: determining a plurality of network data sources related to a target category, wherein the target category is a to-be-analyzed category in a plurality of categories of the intelligent interactive electronic product; acquiring product information data of a plurality of products belonging to a target category from the plurality of network data sources in a network acquisition mode; according to a data structure model used for describing the intelligent interactive electronic product from multiple dimensions, performing structured representation on the acquired product information data of each product to obtain product feature data of each product; performing similarity matching on the plurality of products according to the product feature data of the plurality of products, and performing fusion processing on the product feature data of the plurality of products according to a similarity matching result; and classifying the fused product feature data according to a preset product feature category system.

Description

Technical Field

[0001] The present disclosure generally relates to the fields of computer technology and design, and more specifically, to a feature analysis method, device, electronic device, storage medium and program product for an intelligent interactive electronic product. Background Art

[0002] With the rapid development of smart interactive electronic products, users' demands for product features are becoming increasingly diverse. Analyzing a large number of existing smart interactive electronic products will help optimize new product designs and improve the functions and performance of smart interactive electronic products. In the existing technology, product analysis often relies on limited data sources and manual analysis, which limits the comprehensiveness and accuracy of the analysis. Summary of the invention

[0003] An exemplary embodiment of the present disclosure is to provide a feature analysis method, device, electronic device, storage medium and program product for a smart interactive electronic product, which can automatically extract required information data of the smart interactive electronic product from multi-source network data and perform product feature analysis, thereby improving the comprehensiveness and accuracy of the feature analysis of the smart interactive electronic product.

[0004] According to a first aspect of an embodiment of the present disclosure, a feature analysis method for a smart interactive electronic product is provided, the feature analysis method comprising: determining a plurality of network data sources related to a target category, wherein the target category is a category to be analyzed among a plurality of categories of smart interactive electronic products; collecting product information data of a plurality of products belonging to the target category from the plurality of network data sources through a network collection method; performing a structured representation on the collected product information data of each product according to a data structure model for describing smart interactive electronic products from a plurality of dimensions to obtain product feature data of each product; performing similarity matching on the plurality of products according to the product feature data of the plurality of products, and performing fusion processing on the product feature data of the plurality of products according to the similarity matching results; and classifying the fused product feature data according to a preset product feature category system.

[0005] Optionally, the step of performing similarity matching on the multiple products based on the product feature data of the multiple products includes: for each product among the multiple products, using a pre-trained image representation model to map the image data in the product feature data of the product to a vector space to obtain the image representation space of the product, and using a pre-trained text representation model to map the text data in the product feature data of the product to a vector space to obtain the text representation space of the product; based on the image representation space and text representation space of each of the multiple products, performing similarity matching on the multiple products.

[0006] Optionally, the step of performing similarity matching on the multiple products based on the image representation space and text representation space of each of the multiple products includes: determining, for any two products among the multiple products, the similarity in the image representation space and the similarity in the text representation space between the any two products; taking a weighted sum of the similarity in the image representation space and the similarity in the text representation space to obtain a comprehensive similarity between the any two products; and determining that the any two products belong to the same product when the comprehensive similarity between the any two products is greater than a preset threshold.

[0007] Optionally, the step of fusing the product feature data of the multiple products based on the similarity matching result includes: merging and deduplicating the product feature data of the same product among the product feature data of the multiple products.

[0008] Optionally, the image representation model is a deep neural network based on contrastive learning, a model trained using an image feature dataset for smart interactive electronic products; the text representation model is a deep neural network based on contrastive learning, a model trained using a text feature dataset for smart interactive electronic products.

[0009] Optionally, the types of the multiple network data sources include at least one of the following: online e-commerce, social media platform, product forum.

[0010] Optionally, the network collection method includes at least one of the following items: web crawler technology, API call, and data exchange protocol.

[0011] Optionally, the type of the product information data includes at least one of the following items: product text description, product image, user evaluation, technical specifications, price information, and historical sales volume.

[0012] Optionally, the multiple dimensions include: product name, product attributes, product features, product price, and market trends.

[0013] Optionally, the product feature category system is a feature category system developed based on at least one of the type, function, and user demand of the smart interactive electronic product.

[0014] According to a second aspect of an embodiment of the present disclosure, a feature analysis device for a smart interactive electronic product is provided, the feature analysis device comprising: a data source determination unit, configured to determine a plurality of network data sources related to a target category, wherein the target category is a category to be analyzed among a plurality of categories of smart interactive electronic products; an information collection unit, configured to collect product information data of a plurality of products belonging to the target category from the plurality of network data sources through a network collection method; a feature data acquisition unit, configured to perform a structured representation on the collected product information data of each product according to a data structure model for describing smart interactive electronic products from a plurality of dimensions, so as to obtain product feature data of each product; a matching and fusion unit, configured to perform similarity matching on the plurality of products according to the product feature data of the plurality of products, and to perform fusion processing on the product feature data of the plurality of products according to the similarity matching results; and a classification unit, configured to classify the fused product feature data according to a preset product feature category system.

[0015] According to a third aspect of an embodiment of the present disclosure, a computer-readable storage medium storing instructions is provided, which, when executed by a processor of an electronic device, enables the electronic device to perform the feature analysis method for an intelligent interactive electronic product as described above.

[0016] According to a fourth aspect of an embodiment of the present disclosure, an electronic device is provided, comprising: at least one processor; and at least one memory storing computer executable instructions, wherein the computer executable instructions, when executed by the at least one processor, prompt the at least one processor to execute the feature analysis method for the smart interactive electronic product as described above.

[0017] According to a fifth aspect of an embodiment of the present disclosure, a computer program product is provided, comprising computer executable instructions, which, when executed by at least one processor, implement the feature analysis method for the intelligent interactive electronic product as described above.

[0018] According to the feature analysis method, device, electronic device, storage medium and program product of the intelligent interactive electronic product of the exemplary embodiment of the present disclosure, a method for analyzing the features of the intelligent interactive electronic product based on structured association analysis of multi-source network data is provided, which can utilize rich network data to mine different features of the same product from different data sources, and merge and classify these features, thereby improving the comprehensiveness and accuracy of the feature analysis of the intelligent interactive electronic product.

[0019] In the following description, some aspects and / or advantages of the general inventive concept will be set forth, and some other aspects and / or advantages will be known through the following description or the implementation of the general inventive concept. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] These and / or other aspects and advantages of the present application will become clearer and easier to understand from the following detailed description of the embodiments of the present application in conjunction with the accompanying drawings, in which:

[0021] Figure 1 A flow chart showing a method for analyzing features of a smart interactive electronic product according to an exemplary embodiment of the present disclosure;

[0022] Figure 2 A structural block diagram showing a feature analysis device for an intelligent interactive electronic product according to an exemplary embodiment of the present disclosure;

[0023] Figure 3 A structural block diagram of an electronic device according to an exemplary embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0024] Reference will now be made in detail to the embodiments of the present disclosure, examples of which are illustrated in the accompanying drawings, wherein like reference numerals refer to like parts throughout. The embodiments will be described below with reference to the drawings in order to explain the present disclosure.

[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0026] It should be noted that the phrase "at least one of the items" in the present disclosure includes three types of parallel situations: "any one of the items", "a combination of any number of the items", and "all of the items". For example, "including at least one of A and B" includes the following three parallel situations: (1) including A; (2) including B; (3) including A and B. Another example is "executing at least one of step 1 and step 2" which means the following three parallel situations: (1) executing step 1; (2) executing step 2; (3) executing step 1 and step 2.

[0027] Figure 1 A flow chart of a feature analysis method for a smart interactive electronic product according to an exemplary embodiment of the present disclosure is shown.

[0028] Reference Figure 1, in step S100, multiple network data sources related to the target category are determined.

[0029] The target category is a category to be analyzed among multiple categories of smart interactive electronic products. As an exemplary embodiment, a user selection operation of multiple categories of smart interactive electronic products may be received, and in response to the selection operation, the category selected by the user is used as the target category.

[0030] Smart interactive electronic products are smart electronic products that can interact. As an exemplary embodiment, multiple categories of smart interactive electronic products may include but are not limited to: mobile phones, wearable devices, smart homes, tablet computers, computers, multimedia playback devices, smart educational devices, and smart medical devices.

[0031] As an exemplary embodiment, the types of the plurality of network data sources may include, but are not limited to, at least one of the following items: an online e-commerce, a social media platform, and a product forum.

[0032] It should be understood that the data content covered by the multiple network data sources related to the target category needs to be closely related to the smart interactive electronic products of the category. For example, when the target category is smart medical equipment, the product forums in the multiple network data sources related to the target category can be product forums about medical equipment.

[0033] In step S200, product information data of a plurality of products belonging to a target category is collected from the plurality of network data sources through network collection.

[0034] As an exemplary embodiment, the network collection method may include but is not limited to at least one of the following items: web crawler technology, API call, and data exchange protocol.

[0035] As an exemplary embodiment, the type of product information data may include, but is not limited to, at least one of the following items: product text description, product image, user evaluation, technical specifications, price information, and historical sales volume.

[0036] According to an exemplary embodiment of the present disclosure, relevant data of products of the target category are automatically collected by means of a web crawler according to preset rules and keywords.

[0037] In step S300, the collected product information data of each product is structured according to a data structure model for describing intelligent interactive electronic products from multiple dimensions to obtain product feature data of each product.

[0038] As an exemplary embodiment, a data structure model for intelligent interactive electronic products can be constructed in advance according to the required product features. In other words, the data structure model defines multiple product features for describing the intelligent interactive electronic products from different dimensions (aspects).

[0039] As an exemplary embodiment, multiple dimensions (i.e., multiple product features) may include, but are not limited to: product name, product attributes, product features, product price, and market trend. It should be understood that each of the above dimensions can be further subdivided into multiple sub-dimensions. For example, market trend can be further subdivided into: the trend of price changes over time (i.e., price trend) and the trend of sales volume changes over time (i.e., sales volume trend).

[0040] As an exemplary embodiment, step S300 may include two parts: data preprocessing and data analysis. Regarding data preprocessing, the collected raw data may be cleaned, including methods such as data outlier detection and data deduplication, to remove useless information, such as irrelevant comments. Regarding data analysis, it is used to extract product feature-related data from the preprocessed data. For example, grammatical analysis, entity recognition and other methods may be used to further structure the preprocessed unstructured data, extract product key information belonging to preset dimensions (that is, extract corresponding product key information according to preset dimensions) to obtain product feature data, and use a management system such as a database to store structured product feature data for easy query and analysis.

[0041] According to an exemplary embodiment of the present disclosure, a network data structure model for intelligent interactive electronic products is constructed, and the collected product information data is automatically structured and represented in sequence.

[0042] In step S400, similarity matching is performed on the multiple products based on the product feature data of the multiple products.

[0043] As an exemplary embodiment, step S400 may include: for each product among the multiple products, using a pre-trained image representation model, mapping the image data in the product feature data of the product to a vector space to obtain the image representation space of the product, and using a pre-trained text representation model, mapping the text data in the product feature data of the product to a vector space to obtain the text representation space of the product; based on the image representation space and text representation space of each of the multiple products, performing similarity matching on the multiple products.

[0044] The image representation model and the text representation model are deep neural networks based on contrastive learning, which are trained on a specially constructed dataset for intelligent interactive electronic products. The image representation model and the text representation model can map product-related images or text content obtained from different data sources into a vector space.

[0045] As an example, the image representation model is a deep neural network based on contrastive learning, and is trained using an image feature dataset for smart interactive electronic products. As an example, the image representation model may use a convolutional neural network.

[0046] As an example, the text representation model is a deep neural network based on contrastive learning, and is a model trained using a text feature dataset for smart interactive electronic products. As an example, the text representation model can use a deep network composed of transformers.

[0047] As an exemplary embodiment, the step of performing similarity matching on the multiple products based on the image representation space and text representation space of each of the multiple products may include: determining, for any two products among the multiple products, the similarity in the image representation space and the similarity in the text representation space between the any two products; obtaining the comprehensive similarity between the any two products based on the similarity in the image representation space and the similarity in the text representation space; and determining that the any two products belong to the same product (i.e., are the same product) when the comprehensive similarity between the any two products is greater than a preset threshold.

[0048] It should be understood that if it is determined that product A and product B belong to the same product, and product B and product C also belong to the same product, then it can be considered that products A, B, and C all belong to the same product.

[0049] As an exemplary embodiment, the similarities of the image representation space and the text representation space can be calculated respectively by normalized Euclidean distance, and the two similarities can be fused to obtain a comprehensive similarity by assigning different weights, and then the final matching result (i.e., whether they belong to the same product) can be obtained based on the comprehensive similarity.

[0050] As an example, the similarity result S1 of the image representation space and the similarity result S2 of the text representation space may be fused, for example, in the following manner:

[0051] Sω1S1+ω2S2

[0052] Among them, ω1 and ω2 are weight parameters, which can be set according to actual conditions, specific needs and experience, and ensure that ω1+ω2=1.

[0053] According to an exemplary embodiment of the present disclosure, in order to achieve accurate discrimination of identical products and similar features, an image representation model and a text representation model are used to map the feature text and feature image of the product into a computable vector space to achieve subsequent similarity discrimination.

[0054] In step S500, the product feature data of the multiple products are fused according to the similarity matching result.

[0055] As an exemplary embodiment, step S500 may include: merging and deduplicating product feature data belonging to the same product among the product feature data of the plurality of products.

[0056] In view of the large number of products collected and analyzed from multiple network data sources, the present disclosure proposes to merge the data of the same product from different network data sources and remove duplicate content after merging. For example, product information data of product A is collected through the first network data source, product information data of product B is collected through the second network data source, and product information data of product C is collected through the third network data source. If it is determined that products A, B, and C are the same product, the product feature data of products A, B, and C are merged and de-duplicated.

[0057] According to an exemplary embodiment of the present disclosure, the products corresponding to the vectors and their features are associated using similarity matching results, features belonging to the same product are merged, and similar features are deduplicated.

[0058] In step S600, the fused product feature data is classified according to a preset product feature classification system.

[0059] As an exemplary embodiment, step S600 may include: for each product, determining the feature category to which each item of product feature data in the fused product feature data of the product belongs in the product feature category system, and classifying the product feature data into the feature category.

[0060] As an exemplary embodiment, the product feature category system is a feature category system developed based on at least one of the type, function, and user needs of the smart interactive electronic product. It defines the categories to which different product features belong. For example, material features, color features, shape features, appearance image features, etc. belong to the appearance feature category. By classifying the product feature data belonging to each product feature category, that is, grouping the specific data of product features belonging to the same feature category into one category, it is convenient for users to obtain product feature information more intuitively, systematically, and conveniently.

[0061] It should be understood that the specific feature categories corresponding to different categories may be different, which is related to at least one of the type, function, and user needs of the products in the category. For example, the functions of mobile phones and smart medical devices are quite different, and the corresponding feature categories will be different.

[0062] As an example, the product feature classification system may be developed by professional designers of smart interactive electronic products based on product types, functions, and user needs.

[0063] The classified product feature analysis results can be applied to multiple fields such as product design, market analysis, user recommendations, etc., providing support for design decisions of smart interactive electronic products.

[0064] In addition, the feature analysis method of the smart interactive electronic product according to the exemplary embodiment of the present disclosure may further include: outputting the classified product feature analysis results (ie, the classified product feature data of each product) to the user.

[0065] According to an exemplary embodiment of the present disclosure, a method for analyzing the features of smart interactive electronic products based on structured association analysis of multi-source network data is provided, which can utilize rich network data to mine different features of the same product from different data sources, and merge and classify these features, thereby improving the comprehensiveness and accuracy of the feature analysis of smart interactive electronic products.

[0066] Figure 2 A structural block diagram of a feature analysis device for an intelligent interactive electronic product according to an exemplary embodiment of the present disclosure is shown.

[0067] Reference Figure 2 According to an exemplary embodiment of the present disclosure, a feature analysis device for an intelligent interactive electronic product includes: a data source determination unit 100, an information collection unit 200, a feature data acquisition unit 300, a matching and fusion unit 400, and a classification unit 500.

[0068] Specifically, the data source determination unit 100 is configured to determine a plurality of network data sources related to a target category, wherein the target category is a category to be analyzed among a plurality of categories of smart interactive electronic products.

[0069] The information collection unit 200 is configured to collect product information data of a plurality of products belonging to the target category from the plurality of network data sources through network collection.

[0070] The feature data acquisition unit 300 is configured to perform structured representation on the collected product information data of each product according to a data structure model used to describe intelligent interactive electronic products from multiple dimensions, so as to obtain product feature data of each product.

[0071] The matching and fusion unit 400 is configured to perform similarity matching on the multiple products according to the product feature data of the multiple products, and perform fusion processing on the product feature data of the multiple products according to the similarity matching results.

[0072] The classification unit 500 is configured to classify the fused product feature data according to a preset product feature classification system.

[0073] As an exemplary embodiment, the matching fusion unit 400 can be configured to: for each product of the multiple products, use a pre-trained image representation model to map the image data in the product feature data of the product to a vector space to obtain the image representation space of the product, and use a pre-trained text representation model to map the text data in the product feature data of the product to a vector space to obtain the text representation space of the product; based on the image representation space and text representation space of each of the multiple products, perform similarity matching on the multiple products.

[0074] As an exemplary embodiment, the matching fusion unit 400 can be configured to: determine, for any two products among the multiple products, the similarity in the image representation space and the similarity in the text representation space between the any two products; perform weighted summation on the similarity in the image representation space and the similarity in the text representation space to obtain a comprehensive similarity between the any two products; and determine that the any two products belong to the same product when the comprehensive similarity between the any two products is greater than a preset threshold.

[0075] As an exemplary embodiment, the matching and fusion unit 400 may be configured to: merge and remove duplicates of the product feature data of the multiple products that belong to the same product.

[0076] As an exemplary embodiment, the image representation model is a deep neural network based on contrastive learning, and is a model trained using an image feature dataset for smart interactive electronic products; the text representation model is a deep neural network based on contrastive learning, and is a model trained using a text feature dataset for smart interactive electronic products.

[0077] As an exemplary embodiment, the types of the plurality of network data sources include at least one of the following items: an online e-commerce, a social media platform, and a product forum.

[0078] As an exemplary embodiment, the network collection method includes at least one of the following items: web crawler technology, API call, and data exchange protocol.

[0079] As an exemplary embodiment, the type of the product information data includes at least one of the following items: product text description, product picture, user evaluation, technical specifications, price information, and historical sales volume.

[0080] As an exemplary embodiment, the multiple dimensions include: product name, product attributes, product features, product price, and market trend.

[0081] As an exemplary embodiment, the product feature classification system is a feature classification system formulated according to at least one of the type, function, and user demand of the smart interactive electronic product.

[0082] It should be understood that the specific processing performed by the feature analysis device of the intelligent interactive electronic product according to the exemplary embodiment of the present disclosure has been referred to. Figure 1 The details have been described in detail and will not be repeated here.

[0083] It should be understood that the various units and modules in the feature analysis device of the intelligent interactive electronic product according to the exemplary embodiment of the present disclosure may be implemented as hardware components and / or software components.

[0084] Figure 3 A structural block diagram of an electronic device according to an exemplary embodiment of the present disclosure is shown.

[0085] Reference Figure 3 The electronic device includes: at least one memory 600 and at least one processor 700, wherein the at least one memory 600 stores a set of computer executable instructions, and when the computer executable instruction set is executed by the at least one processor 700, the feature analysis method of the intelligent interactive electronic product as described in the above exemplary embodiment is executed.

[0086] As an example, the electronic device may be a PC, a tablet device, a personal digital assistant, a smart phone, or other device capable of executing the above instruction set. Here, the electronic device is not necessarily a single electronic device, but may also be any device or circuit capable of executing the above instruction (or instruction set) individually or in combination. The electronic device may also be part of an integrated control system or system manager, or may be configured as a portable electronic device interconnected with a local or remote (e.g., via wireless transmission) interface.

[0087] In an electronic device, the processor 700 may include a central processing unit (CPU), a graphics processing unit (GPU), a programmable logic device, a dedicated processor system, a microcontroller or a microprocessor. As an example and not limitation, the processor 700 may also include an analog processor, a digital processor, a microprocessor, a multi-core processor, a processor array, a network processor, etc.

[0088] The processor 700 may execute instructions or codes stored in the memory 600, wherein the memory 600 may also store data. Instructions and data may also be sent and received over a network via a network interface device, wherein the network interface device may employ any known transmission protocol.

[0089] The memory 600 may be integrated with the processor 700, for example, by placing RAM or flash memory within an integrated circuit microprocessor or the like. In addition, the memory 600 may include a separate device, such as an external disk drive, a storage array, or any other storage device that can be used by a database system. The memory 600 and the processor 700 may be operatively coupled, or may communicate with each other, such as through an I / O port, a network connection, etc., so that the processor 700 can read files stored in the memory.

[0090] In addition, the electronic device may also include a video display (such as a liquid crystal display) and a user interaction interface (such as a keyboard, a mouse, a touch input device, etc.) All components of the electronic device may be connected to each other via a bus and / or a network.

[0091] According to an exemplary embodiment of the present disclosure, a computer-readable storage medium storing instructions may also be provided, wherein when the instructions are executed by at least one processor, the at least one processor is prompted to execute the feature analysis method of the intelligent interactive electronic product as described in the above exemplary embodiment. Examples of computer-readable storage media here include: read-only memory (ROM), random access programmable read-only memory (PROM), electrically erasable programmable read-only memory (EEPROM), random access memory (RAM), dynamic random access memory (DRAM), static random access memory (SRAM), flash memory, non-volatile memory, CD-ROM, CD-R, CD+R, CD-RW, CD+RW, DVD-ROM, DVD-R, DVD+R, DVD-RW, DVD+RW, DVD-RAM, BD-ROM, BD-R, BD-R LTH, BD-RE, Blu-ray or optical disk storage, hard disk drive (HDD), solid state drive (SSD), card storage (such as, multimedia card, secure digital (SD) card or extreme digital (XD) card), magnetic tape, floppy disk, magneto-optical data storage device, optical data storage device, hard disk, solid state disk and any other device, any other device is configured to store computer programs and any associated data, data files and data structures in a non-transitory manner and provide the computer programs and any associated data, data files and data structures to a processor or computer so that the processor or computer can execute the computer program. The computer program in the above-mentioned computer-readable storage medium can be run in an environment deployed in a computer device such as a client, a host, an agent device, a server, etc. In addition, in one example, the computer program and any associated data, data files and data structures are distributed on a networked computer system, so that the computer program and any associated data, data files and data structures are stored, accessed and executed in a distributed manner by one or more processors or computers.

[0092] According to an exemplary embodiment of the present disclosure, a computer program product may also be provided, and instructions in the computer program product may be executed by at least one processor to complete the feature analysis method of the intelligent interactive electronic product as described in the above exemplary embodiment.

[0093] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present disclosure are indicated by the following claims.

[0094] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A method for analyzing characteristics of intelligent interactive electronic products, characterized in that: The feature analysis method comprises: Determine multiple network data sources related to a target category, wherein the target category is a category to be analyzed among multiple categories of smart interactive electronic products; Collecting product information data of multiple products belonging to the target category from the multiple network data sources through network collection; According to the data structure model used to describe the intelligent interactive electronic products from multiple dimensions, the collected product information data of each product is structured to obtain product feature data of each product; According to the product feature data of the multiple products, similarity matching is performed on the multiple products, and according to the similarity matching result, the product feature data of the multiple products is fused; The fused product feature data is classified according to the preset product feature category system.

2. The feature analysis method according to claim 1, characterized in that: The step of performing similarity matching on the multiple products according to the product feature data of the multiple products comprises: For each of the multiple products, using a pre-trained image representation model, mapping image data in the product feature data of the product to a vector space to obtain an image representation space of the product, and using a pre-trained text representation model, mapping text data in the product feature data of the product to a vector space to obtain a text representation space of the product; Based on the image representation space and the text representation space of each of the multiple products, similarity matching is performed on the multiple products.

3. The feature analysis method according to claim 2, characterized in that: The step of performing similarity matching on the multiple products based on the image representation space and text representation space of each of the multiple products comprises: For any two products among the plurality of products, determining a similarity in an image representation space and a similarity in a text representation space between the any two products; Performing a weighted summation on the similarity of the image representation space and the similarity of the text representation space to obtain a comprehensive similarity between any two products; When the comprehensive similarity between the any two products is greater than a preset threshold, it is determined that the any two products belong to the same product.

4. The feature analysis method according to claim 3, characterized in that: The step of fusing the product feature data of the plurality of products according to the similarity matching result comprises: Among the product feature data of the multiple products, product feature data belonging to the same product are merged and deduplicated.

5. The feature analysis method according to claim 2, characterized in that: The image representation model is a deep neural network based on contrastive learning, and is a model trained using an image feature dataset for smart interactive electronic products; The text representation model is a deep neural network based on contrastive learning, and is a model trained using a text feature dataset for smart interactive electronic products.

6. The feature analysis method according to claim 1, characterized in that: The types of the plurality of network data sources include at least one of the following: online e-commerce, social media platform, product forum; And / or, the network collection method includes at least one of the following: web crawler technology, API call, data exchange protocol; And / or, the type of the product information data includes at least one of the following items: product text description, product image, user evaluation, technical specifications, price information, historical sales; And / or, the multiple dimensions include: product name, product attributes, product features, product price, market trend; And / or, the product feature classification system is a feature classification system developed based on at least one of the type, function, and user demand of the smart interactive electronic product.

7. A feature analysis device for an intelligent interactive electronic product, characterized in that: The feature analysis device comprises: A data source determination unit is configured to determine a plurality of network data sources related to a target category, wherein the target category is a category to be analyzed among a plurality of categories of intelligent interactive electronic products; An information collection unit is configured to collect product information data of a plurality of products belonging to the target category from the plurality of network data sources through a network collection method; A feature data acquisition unit is configured to perform a structured representation on the collected product information data of each product according to a data structure model for describing intelligent interactive electronic products from multiple dimensions, so as to obtain product feature data of each product; a matching and fusing unit configured to perform similarity matching on the multiple products according to the product feature data of the multiple products, and to perform fusing processing on the product feature data of the multiple products according to the similarity matching result; The classification unit is configured to classify the fused product feature data according to a preset product feature category system.

8. A computer-readable storage medium storing instructions, characterized in that: When the instruction is executed by a processor of an electronic device, the electronic device is enabled to execute the feature analysis method of an intelligent interactive electronic product as claimed in any one of claims 1 to 6.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; at least one memory storing computer executable instructions, Wherein, when the computer executable instructions are executed by the at least one processor, the at least one processor is prompted to execute the feature analysis method of the intelligent interactive electronic product as claimed in any one of claims 1 to 6.

10. A computer program product comprising computer executable instructions, characterized in that: When the computer executable instructions are executed by at least one processor, the feature analysis method for the intelligent interactive electronic product as claimed in any one of claims 1 to 6 is implemented.

Images