Product information acquisition method and device, storage medium and electronic equipment

Abstract

The application provides a product information acquisition method and device, a storage medium and electronic equipment, and relates to the technical field of data processing.The method comprises the following steps: acquiring question information of a target product; inputting the question information of the target product into a pre-trained search model, so that the pre-trained search model outputs answer information corresponding to the question information of the target product.The technical scheme provided by the application can more accurately and efficiently acquire product information, thereby meeting the needs of users.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a product information acquisition method, apparatus, storage medium, and electronic device. Background Technology

[0002] In existing technologies, users search for relevant product information by entering product keywords. However, because this method involves keyword matching within an existing database, the search results are often inaccurate. Users need to filter through numerous search results to find the information they want, or combine information from various results to obtain the desired information. Clearly, existing information retrieval methods are not convenient enough, and users spend a considerable amount of time obtaining the product information they need, failing to meet the current user needs given the large volume of product information available. Summary of the Invention

[0003] In view of the problems in the prior art, this application proposes a product information acquisition method, apparatus, storage medium and electronic device, which can acquire product information more accurately and efficiently, thereby meeting user needs.

[0004] To achieve the above objectives, the technical solution of the present invention is implemented as follows:

[0005] In a first aspect, embodiments of the present invention provide a product information acquisition method, the method comprising:

[0006] Obtain problem information about the target product;

[0007] The question information of the target product is input into a pre-trained search model so that the pre-trained search model outputs answer information corresponding to the question information of the target product.

[0008] In some embodiments, obtaining the problem information of the target product includes:

[0009] Obtain the user's input regarding the target product's problem information;

[0010] The method further includes:

[0011] The user-inputted question information about the target product is stored as input information.

[0012] Obtain the answer information that corresponds to the input information and meets the preset conditions as the output information;

[0013] Based on the input information and the output information, the pre-trained search model is improved to obtain the improved search model;

[0014] The step of inputting the question information of the target product into a pre-trained search model, so that the pre-trained search model outputs answer information corresponding to the question information of the target product, includes:

[0015] The question information of the target product is input into the improved search model so that the improved search model outputs the answer information corresponding to the question information of the target product.

[0016] In some embodiments, the user inputs the problem information of the target product in any of the following ways:

[0017] Text input, voice input, or input by clicking preset keywords.

[0018] In some embodiments, the answer information that meets the preset conditions includes: answer information whose matching degree with the input information exceeds a first preset threshold, or answer information whose user satisfaction with the answer information exceeds a second preset threshold.

[0019] In some embodiments, the pre-trained search model outputs answer information corresponding to the question information of the target product in the following manner:

[0020] Content recognition is performed on the problem information of the target product to obtain keyword information;

[0021] Based on the keyword information, product information matching the keyword information is searched in a pre-established product model database to obtain a product information set; wherein, the product model database stores the association information between each product and other products;

[0022] The product information set is output as the answer information corresponding to the question information of the target product.

[0023] In some embodiments, the step of performing content recognition on the problem information of the target product to obtain keyword information includes:

[0024] Hidden Markov Models are used to perform content recognition on the problem information of the target product to obtain the keyword information.

[0025] In some embodiments, the product model database is pre-established in the following manner:

[0026] Obtain basic information and external related information for each existing product;

[0027] Based on the basic information of each product, construct a product model for each product;

[0028] Based on the external association information of each product, a relationship between each product model is constructed; wherein, the relationship between each product model reflects the association information between each product and other products.

[0029] Secondly, embodiments of the present invention provide a product information acquisition device, the device comprising:

[0030] The first acquisition unit is used to acquire problem information about the target product;

[0031] The input unit is used to input the question information of the target product into a pre-trained search model, so that the pre-trained search model outputs the answer information corresponding to the question information of the target product.

[0032] Thirdly, embodiments of the present invention provide a computer-readable storage medium storing program code, which, when executed by a processor, implements the product information acquisition method as described in any of the above embodiments.

[0033] Fourthly, embodiments of the present invention provide an electronic device, the electronic device including a memory and a processor, the memory storing program code executable on the processor, the program code being executed by the processor to implement the product information acquisition method as described in any of the above embodiments.

[0034] The product information acquisition method, apparatus, storage medium, and electronic device provided in this invention acquire question information about a target product and input this question information into a pre-trained search model. The pre-trained search model then outputs answer information corresponding to the question information of the target product. This allows for convenient, fast, and efficient acquisition of product information based on the question information of the target product and the search model. Because the search model is pre-trained, it can accurately output information based on training samples, that is, accurately output answer information corresponding to the question information of the target product. Therefore, the technical solution provided in this invention, compared with existing information retrieval methods, can acquire product information more accurately and efficiently, thereby meeting user needs.

[0035] Furthermore, the technical solution provided by the embodiments of the present invention can also improve the search model based on the question information of the target product input by the user and the answer information corresponding to the question information and meeting the preset conditions. That is, the search model can be continuously improved during the process of the user inputting question information, thereby further improving the accuracy of the answer information of the target product output by the search model. Attached Figure Description

[0036] The scope of this invention can be better understood by reading the following detailed description of exemplary embodiments in conjunction with the accompanying drawings. The accompanying drawings are:

[0037] Figure 1 The method flow of this invention embodiment Figure 1 ;

[0038] Figure 2 The method flow of this invention embodiment Figure 2 ;

[0039] Figure 3 The device structure of this invention is shown in the embodiment. Figure 1 ;

[0040] Figure 4 The device structure of this invention is shown in the embodiment. Figure 2 . Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of the present invention clearer, the implementation method of the present invention will be described in detail below with reference to the accompanying drawings and embodiments, so that the process of how the present invention uses technical means to solve technical problems and achieve technical effects can be fully understood and implemented accordingly.

[0042] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0043] Example 1

[0044] Information storage and retrieval refers to the process of storing or structuring information using certain methods, matching precise key information from massive amounts of data based on user characteristics, meeting user needs in the storage and retrieval process, and obtaining highly targeted information.

[0045] With the rapid development of the Internet and e-commerce, the amount of data and information online has exploded. Enterprises and individuals are facing the challenge of storing massive amounts of data on the network. How to process this data effectively, establish interconnected models, meet user needs and business scenarios, build product correlations through user input, and gradually complete search models for each product to achieve the ideal effect of accurate search and perfect matching is a problem that urgently needs to be solved.

[0046] This invention provides a product information acquisition method. By establishing a product model, the relationship between products is strengthened, forming an independent yet interconnected network structure. By establishing a search model, pattern recognition is performed on user input text, voice, and behavior, and a recurrent neural network is used to increase the probability of the search event being selected. By learning from examples of user-input question information, the search model is continuously trained and improved so that the search model outputs product information with higher matching degree and greater accuracy.

[0047] like Figure 1 As shown, the product information acquisition method described in this embodiment includes steps S101 and S102. The specific contents of these steps are described in detail below:

[0048] Step S101: Obtain problem information for the target product;

[0049] In this embodiment, obtaining the question information of the target product includes: obtaining the question information of the target product input by the user. That is, in this embodiment, the user can directly input the question information of the target product to obtain the answer information of the target product corresponding to the question information, so as to make the information acquisition more intuitive for the user and thus make the operation more convenient.

[0050] To further facilitate user operation, in this embodiment, users can input the problem information of the target product using any of the following methods: text input, voice input, or input by clicking preset keywords. That is, in this embodiment, users can input the problem information of the target product via text, voice, or by clicking preset keywords on a preset page on the electronic device. Users can choose any of the above methods to input the problem information of the target product based on their actual situation.

[0051] For example, on a product's after-sales service platform, users can inquire about the product's after-sales policy, warranty period, past issues or solutions, historical information, and customer purchase records. Upon receiving this information, the system automatically outputs comprehensive and highly relevant answers.

[0052] Step S102: Input the question information of the target product into the pre-trained search model so that the pre-trained search model outputs the answer information corresponding to the question information of the target product.

[0053] In this embodiment, the search model is a recurrent neural network (RNN) model. The RNN model is trained using an error backpropagation algorithm. The input samples during training are the user-inputted question information, and the output samples are the answer information corresponding to that question information.

[0054] To obtain product information more accurately and efficiently, in this embodiment, the pre-trained search model outputs answer information corresponding to the question information of the target product in the following manner: content recognition is performed on the question information of the target product to obtain keyword information; based on the keyword information, product information matching the keyword information is searched in a pre-established product model database to obtain a product information set; wherein, the product model database stores the association information between each product and other products; the product information set is output as the answer information corresponding to the question information of the target product.

[0055] To more accurately and efficiently identify the content of the problem information of the target product, this embodiment describes the content identification of the problem information of the target product to obtain keyword information, which includes: using a Hidden Markov Model to identify the content of the problem information of the target product to obtain the keyword information.

[0056] To enable the product information in the product model database to be interconnected and obtain more complete and comprehensive product information, the product model database described in this embodiment is pre-established in the following manner: acquiring the basic information and external association information of each existing product; constructing a product model for each product based on the basic information of each product; and constructing the relationship between each product model based on the external association information of each product; wherein the relationship between each product model reflects the association information between each product and other products.

[0057] The product information in the product model database is constantly updated. Users develop search habits and search scope through repeated searches, and new product information is added to the search scope based on user behavior, forming new relational networks. That is, the product models described in this embodiment, and the relationships between them, are also continuously updated and improved based on user search content. The basic product information and external relational information described in this embodiment include product type, model, promotional and after-sales information, technical parameters, training service information, and market sales information.

[0058] Specifically, the user inputs a question about the target product, and the system uses a Hidden Markov Model (HMM) to perform content recognition on that question. For speech recognition, a corresponding HMM is generated for each word. The system evaluates and selects the HMM most likely to produce the pronunciation represented by the observed sequence, where each observed sequence consists of the speech of a single word. For text recognition, each sentence can be considered a given observable state, and an HMM is built to find the most likely word segmentation method. Then, based on the identified keywords, the system searches the product model database built using the above method for product information matching those keywords. All the searched product information is then output as the answer information corresponding to the question about the target product and displayed to the user.

[0059] Because each product model is interconnected, they can form a network structure. Thus, when a user searches for a product-related question, they can find not only the answer to the product itself but also information about related products, making the search model's output more comprehensive. In other words, this embodiment expands the search scope for user questions and recommends relevant information related to the user's current search query. For example, if a user searches for "air conditioner," it can recommend related series or products.

[0060] The stronger the correlation between products, the more and more detailed the information obtained, forming a complementary information effect. The more complete the product information, the more users can obtain information about all related products when they search for smart air conditioners.

[0061] To continuously improve the search model and further enhance the accuracy of the target product's output answer information, such as... Figure 2 As shown, the method described in this embodiment further includes:

[0062] Step S103: Store the problem information of the target product input by the user as input information;

[0063] Step S104: Obtain the answer information that corresponds to the input information and meets the preset conditions as the output information;

[0064] Step S105: Based on the input information and the output information, improve the pre-trained search model to obtain the improved search model.

[0065] In this embodiment, storing and recording the question information of the target product entered by the user each time is to allow the search model to continuously learn from search examples, so as to continuously train and improve the search model.

[0066] Based on the above technical solution, the method described in this embodiment of inputting the question information of the target product into a pre-trained search model so that the pre-trained search model outputs answer information corresponding to the question information of the target product includes: inputting the question information of the target product into the improved search model so that the improved search model outputs answer information corresponding to the question information of the target product.

[0067] In this embodiment, the search model can also continuously train and improve itself based on the user's input of the target product's question information, so that the search model can output more accurate product information in practical applications. Improving the search model belongs to the system's information-enhancing structured learning, which mainly relies on the machine's deep learning process on examples. Continuous learning will obtain more excellent learning results and better match the user's needs and questions, thereby improving the matching rate of its own search.

[0068] It should be noted that when the problem information of the target product obtained by the system is the problem information input by the user, there is no order of execution between the above steps S103 and S102, and they can be executed simultaneously. Figure 2 This is merely an example of one implementation method.

[0069] In order to ensure that the product information output by the search model meets user needs, the answer information that meets the preset conditions in this embodiment includes: answer information whose matching degree with the input information exceeds a first preset threshold, or answer information whose user satisfaction with the answer information exceeds a second preset threshold.

[0070] Specifically, in the actual use of the above search model, each different question asked by the user will be recorded and then matched with the known product information in the product model database to obtain answer information that corresponds to the question and has a high degree of matching. This will continuously train and improve the search model, ultimately achieving the effect of accurate search and gradually completing the pyramid of search solutions.

[0071] As user needs are gradually met, the search model will become more sophisticated, and different methods will be developed to keep pace with the times based on different user needs, problems, and solutions at different times. At the same time, historical information will be recorded, so as to achieve a steady and accurate search solution from phased progress.

[0072] Among them, the above-mentioned user needs are gradually met as the completeness and relevance of product information and the perfection of the search model are improved. When the product information output by the search model completely matches the user's input question, or when the number of times the user searches for similar questions about the product decreases, or when the satisfaction level with the search results reaches a high score, it means that the user's needs are largely met.

[0073] The different periods mentioned above refer to different stages of a user's product usage, such as the initial, middle, and later stages of using a new product, representing the user's familiarization process with the product. Different needs refer to the different product information users want to know; for example, the product information a user wants to know in the early stages of using a new product differs from the product information they want after using it for a period of time. In other words, users will encounter different problems during product use; these problems are the user-inputted questions. After searching a massive product model database, corresponding solutions will be found. As the system's search model is gradually improved, and each product has its own model, the search results will become increasingly accurate. However, problems are constantly arising, and products are also updated and upgraded to keep pace with the times. The technologies or solutions used will change accordingly, so the content users search for will change, and the product's service policies and handling methods will also change. These historical changes accumulate step by step, forming different stages and gradually being completed. The user's role and the system's own model are both improved based on these growth and changes.

[0074] The product information acquisition method provided in this invention obtains question information about a target product and inputs this question information into a pre-trained search model. The pre-trained search model then outputs answer information corresponding to the question information of the target product. This allows for convenient, fast, and efficient acquisition of product information based on the question information and the search model. Since the search model is pre-trained, it can accurately output information based on training samples, that is, accurately output answer information corresponding to the question information of the target product. Therefore, the technical solution provided in this invention, compared with existing information retrieval methods, can acquire product information more accurately and efficiently, thereby meeting user needs.

[0075] Furthermore, the technical solution provided by the embodiments of the present invention can also improve the search model based on the question information of the target product input by the user and the answer information corresponding to the question information and meeting the preset conditions. That is, the search model can be continuously improved during the process of the user inputting question information, thereby further improving the accuracy of the answer information of the target product output by the search model.

[0076] Example 2

[0077] Corresponding to the above method embodiments, the present invention also provides a product information acquisition device, such as... Figure 3 As shown, the device includes:

[0078] The first acquisition unit 201 is used to acquire problem information of the target product;

[0079] The input unit 201 is used to input the question information of the target product into a pre-trained search model so that the pre-trained search model outputs answer information corresponding to the question information of the target product.

[0080] In this embodiment, the first acquisition unit 201 acquires the problem information of the target product in the following way: acquiring the problem information of the target product input by the user.

[0081] Furthermore, such as Figure 4 As shown, the apparatus described in this embodiment further includes:

[0082] Storage unit 203 is used to store the user-inputted question information about the target product as input information;

[0083] The second acquisition unit 204 is used to acquire answer information that corresponds to the input information and meets preset conditions as output information;

[0084] The improvement unit 205 is used to improve the pre-trained search model based on the input information and the output information to obtain the improved search model.

[0085] In this embodiment, the input unit 201 is further configured to input the question information of the target product into the improved search model, so that the improved search model outputs the answer information corresponding to the question information of the target product.

[0086] In this embodiment, the user inputs the problem information of the target product using any of the following methods:

[0087] Text input, voice input, or input by clicking preset keywords.

[0088] In this embodiment, the answer information that meets the preset conditions includes: answer information whose matching degree with the input information exceeds a first preset threshold, or answer information whose user satisfaction with the answer information exceeds a second preset threshold.

[0089] In this embodiment, the pre-trained search model outputs the answer information corresponding to the question information of the target product in the following manner:

[0090] Content recognition is performed on the problem information of the target product to obtain keyword information;

[0091] Based on the keyword information, product information matching the keyword information is searched in a pre-established product model database to obtain a product information set; wherein, the product model database stores the association information between each product and other products;

[0092] The product information set is output as the answer information corresponding to the question information of the target product.

[0093] In this embodiment, the pre-trained search model uses the following method to perform content recognition on the question information of the target product to obtain keyword information:

[0094] Hidden Markov Models are used to perform content recognition on the problem information of the target product to obtain the keyword information.

[0095] In this embodiment, the product model database is pre-established in the following manner:

[0096] Obtain basic information and external related information for each existing product;

[0097] Based on the basic information of each product, construct a product model for each product;

[0098] Based on the external association information of each product, a relationship between each product model is constructed; wherein, the relationship between each product model reflects the association information between each product and other products.

[0099] For details regarding the working principle, workflow, and specific implementation methods of the aforementioned device, please refer to the specific implementation methods of the product information acquisition method provided by this invention. The same technical content will not be described in detail here.

[0100] The product information acquisition device provided in this invention acquires question information about a target product and inputs this question information into a pre-trained search model. The pre-trained search model then outputs answer information corresponding to the question information of the target product. This allows for convenient, fast, and efficient acquisition of product information based on the question information and the search model. Because the search model is pre-trained, it can accurately output information based on training samples, that is, accurately output answer information corresponding to the question information of the target product. Therefore, the technical solution provided in this invention, compared with existing information retrieval methods, can acquire product information more accurately and efficiently, thereby meeting user needs.

[0101] Furthermore, the technical solution provided by the embodiments of the present invention can also improve the search model based on the question information of the target product input by the user and the answer information corresponding to the question information and meeting the preset conditions. That is, the search model can be continuously improved during the process of the user inputting question information, thereby further improving the accuracy of the answer information of the target product output by the search model.

[0102] Example 3

[0103] According to an embodiment of the present invention, a computer-readable storage medium is also provided, wherein program code is stored on the computer-readable storage medium, and when the program code is executed by a processor, it implements the product information acquisition method as described in any of the above embodiments.

[0104] Example 4

[0105] According to an embodiment of the present invention, an electronic device is also provided, the electronic device including a memory and a processor, wherein the memory stores program code that can run on the processor, and when the program code is executed by the processor, it implements the product information acquisition method as described in any of the above embodiments.

[0106] The technical solution provided in this invention is applicable to big data products for e-commerce enterprises and is a product model search solution based on massive amounts of data. The big data products for e-commerce enterprises include e-commerce after-sales platforms for various household appliance products and product resource information storage and query platforms.

[0107] The product information acquisition method, apparatus, storage medium, and electronic device provided in this invention acquire question information about a target product and input this question information into a pre-trained search model. The pre-trained search model then outputs answer information corresponding to the question information of the target product. This allows for convenient, fast, and efficient acquisition of product information based on the question information of the target product and the search model. Because the search model is pre-trained, it can accurately output information based on training samples, that is, accurately output answer information corresponding to the question information of the target product. Therefore, the technical solution provided in this invention, compared with existing information retrieval methods, can acquire product information more accurately and efficiently, thereby meeting user needs.

[0108] Furthermore, the technical solution provided by the embodiments of the present invention can also improve the search model based on the question information of the target product input by the user and the answer information corresponding to the question information and meeting the preset conditions. That is, the search model can be continuously improved during the process of the user inputting question information, thereby further improving the accuracy of the answer information of the target product output by the search model.

[0109] This invention solves the following technical problems:

[0110] 1. It solves the problem that the massive amount of statically stored information is disconnected from the actual problems of users, resulting in incomplete information matching during the search process.

[0111] 2. It solved the problem of weak correlation between various products, which failed to meet users' actual search needs.

[0112] 3. It solves the problem of long search times and complex queries for users.

[0113] The present invention also has the following beneficial effects:

[0114] 1. By allowing users to actively input questions, the system achieves matching and association with product information in a massive database, improving the accuracy of searches and continuously refining and evolving into a search method that better meets practical requirements.

[0115] 2. By establishing a relationship between massive amounts of information and user input, the matching rate between products and related issues is improved. The search model will be continuously improved based on user searches, so as to find relevant information and solutions to problems more efficiently and accurately.

[0116] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed.

[0117] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of the present invention, depending on actual needs.

[0118] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0119] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause an electronic device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0120] While the embodiments disclosed in this invention are as described above, the content is merely for the purpose of facilitating understanding of the invention and is not intended to limit the invention. Any person skilled in the art to which this invention pertains may make any modifications and changes in form and detail of the implementation without departing from the spirit and scope disclosed herein; however, the scope of protection of this invention shall still be determined by the scope defined in the appended claims.

Claims

1. A product information acquisition method characterized by comprising: The method comprises: acquiring question information of a target product; inputting the question information of the target product into a pre-trained search model to enable the pre-trained search model to output answer information corresponding to the question information of the target product; the pre-trained search model outputs answer information corresponding to the question information of the target product in the following manner: performing content recognition on the question information of the target product to obtain keyword information; based on the keyword information, searching for product information matching the keyword information in a pre-established product model database to obtain a product information set; wherein the product model database stores association information between each product and other products, the relationship between each product model reflects the association information between each product and other products, the product model, and the relationship network between the product models is continuously updated along with the search content of the user; outputting the product information set as the answer information corresponding to the question information of the target product; the product model database is pre-established in the following manner: acquiring existing basic information and external association information of each product; based on the basic information of each product, constructing a product model of each product; based on the external association information of each product, constructing a relationship between each product model; wherein the relationship between each product model reflects the association information between each product and other products.

2. The product information acquisition method according to claim 1, characterized by, The acquisition of the question information of the target product comprises: acquiring question information of a target product input by a user; The method further comprises: storing the question information of the target product input by the user as input information; acquiring answer information corresponding to the input information and satisfying a preset condition as output information; based on the input information and the output information, perfecting the pre-trained search model to obtain a perfected search model; the inputting of the question information of the target product into the pre-trained search model to enable the pre-trained search model to output answer information corresponding to the question information of the target product comprises: inputting the question information of the target product into the perfected search model to enable the perfected search model to output answer information corresponding to the question information of the target product.

3. The product information acquisition method according to claim 2, characterized by, The user inputs the question information of the target product in any of the following manners: text input, voice input, and clicking on a preset keyword to input.

4. The product information acquisition method according to claim 2, characterized by, The answer information satisfying the preset condition comprises: answer information with a matching degree exceeding a first preset threshold with the input information, or answer information with a satisfaction degree exceeding a second preset threshold with the user.

5. The product information acquisition method according to claim 1, characterized by, The content recognition on the question information of the target product to obtain keyword information comprises: performing content recognition on the question information of the target product using a hidden Markov model to obtain the keyword information.

6. A product information acquisition apparatus characterized by comprising: The device comprises: a first acquisition unit configured to acquire question information of a target product; The input unit is configured to input the question information of the target product into a pre-trained search model, so that the pre-trained search model outputs answer information corresponding to the question information of the target product; The pre-trained search model outputs the answer information corresponding to the question information of the target product in the following manner: Content recognition is performed on the question information of the target product to obtain keyword information; Based on the keyword information, product information set is obtained by searching the product information database for product information matching the keyword information, wherein the product information database stores association information between each product and other products, the relationship between each product model reflects the association information between each product and other products, the product model, and the relationship network between the product models is continuously updated according to the search content of the user; The product information set is output as the answer information corresponding to the question information of the target product; The product information database is pre-established in the following manner: Basic information and external association information of each product are obtained; Based on the basic information of each product, a product model of each product is constructed; Based on the external association information of each product, a relationship between each product model is constructed, wherein the relationship between each product model reflects the association information between each product and other products.

7. A computer-readable storage medium having stored thereon a program code, characterized in that, The program code is executed by the processor to implement the product information acquisition method of any one of claims 1-5.

8. An electronic device, comprising: The electronic device includes a memory and a processor, and the memory stores program code executable on the processor, and the program code is executed by the processor to implement the product information acquisition method of any one of claims 1-5.

Images