Material retrieval method, device and program product

Abstract

The invention belongs to the field of intelligent household appliances, and particularly relates to a material retrieval method and device and a program product. The method comprises the steps of obtaining keywords and a preset attribute value set; the keyword is determined according to a material retrieval request triggered by a user; the preset attribute value set comprises attribute values corresponding to a plurality of preset materials; the number of the keywords is at least one; in the preset attribute value set, obtaining at least one attribute value with the highest association score according to the keyword; the association score is used for representing the association degree between the attribute value and the keyword; determining a total association score corresponding to each preset material according to the attribute value with the highest association score; and sending the at least one preset material with the highest total association score to a user terminal, so that the user terminal displays the preset materials with the highest total association score. The retrieval efficiency is improved, and the user experience is improved.

Description

Technical Field

[0001] This application belongs to the field of smart home appliances, specifically involving a material retrieval method, device, and program product. Background Technology

[0002] With the development of computer technology and the intensification of market competition, many companies usually pre-design a large amount of promotional materials, such as text materials, image materials, audio materials, and video materials, in order to promote their own products. This involves the retrieval of materials.

[0003] Currently, when searching for materials, it is generally done by pre-setting corresponding attribute selection components based on the multiple attributes of the materials. Then, the user specifies the required attribute values ​​in each attribute selection component, and the search is performed on a massive amount of materials based on the user's specified results.

[0004] However, this search method requires users to manually select attribute values ​​in each attribute selection component, which results in low search efficiency and affects user experience. Summary of the Invention

[0005] This application provides a material retrieval method, device, and program product to solve the problems of low retrieval efficiency and negative impact on user experience in existing material retrieval methods.

[0006] Firstly, this application provides a material retrieval method, which includes:

[0007] Obtain a set of keywords and preset attribute values; the keywords are determined based on a user-triggered material retrieval request; the set of preset attribute values ​​includes attribute values ​​corresponding to multiple preset materials; the number of keywords is at least one.

[0008] In the preset set of attribute values, at least one attribute value with the highest association score is obtained based on the keyword; the association score is used to characterize the degree of association between the attribute value and the keyword.

[0009] The total association score for each preset material is determined based on the attribute value with the highest association score.

[0010] Send at least one preset material with the highest total association score to the user terminal so that the user terminal displays the preset material with the highest total association score.

[0011] Optionally, obtaining at least one attribute value with the highest association score from the preset attribute value set based on the keyword includes:

[0012] The keywords and each attribute value in the preset attribute value set are vectorized to obtain vectorized keywords and each vectorized attribute value.

[0013] Calculate the similarity between each vectorized attribute value and the vectorized keyword, and determine the similarity corresponding to each vectorized attribute value as the association score corresponding to each attribute value;

[0014] Sort the attribute values ​​based on their corresponding correlation scores, and obtain at least one attribute value with the highest corresponding correlation score based on the sorting results.

[0015] Optionally, if there are multiple keywords, then calculating the similarity between each vectorized attribute value and the vectorized keyword includes:

[0016] Calculate the initial similarity between each vectorized attribute value and multiple vectorized keywords to obtain multiple initial similarities corresponding to each vectorized attribute value;

[0017] The sum of multiple initial similarities is used to determine the similarity corresponding to each vectorized attribute value.

[0018] Optionally, determining the total association score for each preset material based on the attribute value with the highest association score includes:

[0019] Determine the preset material to which the attribute value with the highest associated score belongs, and obtain the associated scores of the preset materials that belong to the same preset material;

[0020] The association scores of the same preset materials are summed, and the total association score of each preset material is determined based on the processing result.

[0021] Optionally, the attribute values ​​include meta attribute values ​​and custom attribute values;

[0022] The meta-attribute value is used to indicate the form of the preset material; the custom attribute value is used to indicate the content of the preset material.

[0023] Secondly, this application provides a material retrieval method, the method comprising:

[0024] In response to a user-triggered action, a material retrieval request is generated;

[0025] The material retrieval request is sent to the server so that the server can obtain keywords and a set of preset attribute values; the set of preset attribute values ​​includes attribute values ​​corresponding to multiple preset materials; the number of keywords is at least one; in the set of preset attribute values, at least one attribute value with the highest correlation score is obtained according to the keyword; the correlation score is used to characterize the degree of correlation between the attribute value and the keyword; the total correlation score corresponding to each preset material is determined according to the attribute value with the highest correlation score.

[0026] Receive at least one preset material with the highest total association score sent by the server, and display each preset material with the highest total association score.

[0027] Optionally, in response to a user-triggered action, a material retrieval request is generated, specifically including:

[0028] The system receives material description text input by the user and generates a material retrieval request based on the material description text; wherein, the server is used to obtain the material description text based on the material retrieval request and extract at least one keyword from the material description text.

[0029] Thirdly, this application provides a material retrieval device, the device comprising:

[0030] The acquisition module is used to acquire keywords and a set of preset attribute values; the keywords are determined based on the material retrieval request triggered by the user; the set of preset attribute values ​​includes attribute values ​​corresponding to multiple preset materials; the number of keywords is at least one; in the set of preset attribute values, at least one attribute value with the highest correlation score is acquired based on the keywords; the correlation score is used to characterize the degree of correlation between the attribute value and the keyword;

[0031] The processing module is used to determine the total association score for each preset material based on the attribute value with the highest association score.

[0032] The transceiver module is used to send at least one preset material with the highest total association score to the user terminal, so that the user terminal can display the preset material with the highest total association score.

[0033] Optionally, the acquisition module is further configured to, when acquiring at least one attribute value with the highest association score in the preset attribute value set based on the keyword, perform vectorization processing on the keyword and each attribute value in the preset attribute value set to obtain vectorized keywords and each vectorized attribute value; calculate the similarity between each vectorized attribute value and the vectorized keyword respectively, and determine the similarity corresponding to each vectorized attribute value as the association score corresponding to each attribute value; sort each attribute value based on the corresponding association score, and acquire at least one attribute value with the highest corresponding association score based on the sorting result.

[0034] Optionally, the processing module is further configured to, if the number of keywords is multiple, calculate the initial similarity between each vectorized attribute value and multiple vectorized keywords when calculating the similarity between each vectorized attribute value and the vectorized keywords respectively, so as to obtain multiple initial similarities corresponding to each vectorized attribute value; and determine the sum of the multiple initial similarities as the similarity corresponding to each vectorized attribute value.

[0035] Optionally, the processing module is further configured to, when determining the total association score corresponding to each preset material based on the attribute value with the highest association score, determine the preset material to which the attribute value with the highest association score belongs, and obtain the association scores of the same preset material; sum the association scores of the same preset material, and determine the total association score corresponding to each preset material based on the processing result.

[0036] Optionally, the attribute values ​​include meta attribute values ​​and custom attribute values; the meta attribute values ​​are used to indicate the form of the preset material; the custom attribute values ​​are used to indicate the content of the preset material.

[0037] Fourthly, this application provides a material retrieval device, the device comprising:

[0038] The processing module is used to generate material retrieval requests in response to user-triggered operations;

[0039] The transceiver module is used to send the material retrieval request to the server so that the server can obtain keywords and a set of preset attribute values. The set of preset attribute values ​​includes attribute values ​​corresponding to multiple preset materials. The number of keywords is at least one. In the set of preset attribute values, at least one attribute value with the highest correlation score is obtained based on the keyword. The correlation score is used to characterize the degree of correlation between the attribute value and the keyword. The total correlation score corresponding to each preset material is determined based on the attribute value with the highest correlation score.

[0040] The transceiver module is also used to receive at least one preset material with the highest total association score sent by the server.

[0041] The processing module is also used to display the preset material with the highest total correlation score.

[0042] Optionally, the processing module is further configured to receive material description text input by the user and generate a material retrieval request based on the material description text when generating a material retrieval request in response to a user-triggered operation; wherein, the server is configured to obtain the material description text based on the material retrieval request and extract at least one keyword from the material description text.

[0043] Fifthly, this application provides a server-side device, comprising:

[0044] Memory;

[0045] processor;

[0046] The memory stores computer-executed instructions;

[0047] The processor executes computer execution instructions stored in the memory to enable the server-side device to implement the material retrieval method as described in the first aspect and various possible implementations of the first aspect.

[0048] Sixthly, this application provides a user terminal device, comprising:

[0049] Memory;

[0050] processor;

[0051] The memory stores computer-executed instructions;

[0052] The processor executes computer execution instructions stored in the memory to enable the user terminal device to implement the material retrieval method as described in the second aspect and various possible implementations of the second aspect above.

[0053] In a seventh aspect, this application provides a computer storage medium storing computer execution instructions thereon, which are executed by a processor to implement the material retrieval method as described in the first aspect and various possible implementations thereof, as well as the second aspect and various possible implementations thereof.

[0054] Eighthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the material retrieval method described in the first aspect and various possible implementations thereof, as well as the material retrieval method described in the second aspect and various possible implementations thereof.

[0055] The material retrieval method, device, and program product provided in this application obtain keywords and a set of preset attribute values. The keywords are determined based on a material retrieval request triggered by a user. The set of preset attribute values ​​includes attribute values ​​corresponding to multiple preset materials. The number of keywords is at least one. In the set of preset attribute values, at least one attribute value with the highest correlation score is obtained based on the keywords. The correlation score is used to characterize the degree of correlation between the attribute value and the keyword. The total correlation score corresponding to each preset material is determined based on the attribute value with the highest correlation score. At least one preset material with the highest total correlation score is sent to the user terminal so that the user terminal displays the preset materials with the highest total correlation scores. This eliminates the need for users to manually operate in various selection components, improving retrieval efficiency and enhancing user experience. Attached Figure Description

[0056] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0057] Figure 1This is a schematic diagram illustrating a scenario of the material retrieval method provided in this application;

[0058] Figure 2 This is the flow of the material retrieval method provided in this application. Figure 1 ;

[0059] Figure 3 This is the flow of the material retrieval method provided in this application. Figure 2 ;

[0060] Figure 4 This is a signaling interaction diagram of the material retrieval method provided in this application;

[0061] Figure 5 This is a schematic diagram of the material retrieval device provided in this application. Figure 1 ;

[0062] Figure 6 This is a schematic diagram of the material retrieval device provided in this application. Figure 2 ;

[0063] Figure 7 This is a schematic diagram of the server-side device provided in this application;

[0064] Figure 8 This is a schematic diagram of the user terminal device provided in this application.

[0065] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0066] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0067] The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein.

[0068] In this application, the terms "exemplary" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0069] Currently, when searching for materials, users typically manually select attribute values ​​on their devices. Specifically, based on the various attributes of the materials, corresponding attribute values ​​are pre-defined for a vast amount of material, and corresponding selection components are configured for each attribute. Then, in response to a user selecting a corresponding attribute value on their device and triggering a search request based on the selection result, the system searches through the vast amount of material for materials with those selected attribute values ​​and provides these materials as search results to the user.

[0070] The selection components offer a variety of configurable attribute values. Therefore, when a user selects from these components, they are essentially selecting from these configurable attribute values. Once the user has made their selections, the system generates search criteria based on the user-specified configurable attribute values, and the search is then performed based on these criteria.

[0071] This forces users to manually select attribute values ​​in each selection component, reducing search efficiency and impacting user experience.

[0072] To address the aforementioned technical problems, this application proposes the following technical concept: To improve retrieval efficiency and user experience, instead of configuring corresponding selection components for each attribute and having the user manually trigger each selection component for retrieval, a set is pre-formed based on the attribute values ​​corresponding to each material. When a user needs to retrieve materials, they input a text description related to the desired material on their terminal, and a retrieval request is triggered based on the input result. In response to the triggered retrieval request, the aforementioned text description is retrieved, and at least one keyword is determined based on this text description. Then, these keywords are treated as a whole, and the degree of correlation between each attribute value in the aforementioned set and the whole is determined, obtaining N attribute values ​​with high correlation. Finally, based on the aforementioned N attribute values, the overall correlation of each material is determined, and L materials with high overall correlation are provided to the user. This eliminates the need for users to manually operate in each selection component, improving retrieval efficiency and user experience.

[0073] N and L can be pre-configured quantity values ​​by the administrator; they can be the same or different.

[0074] Figure 1 This is a schematic diagram illustrating a scenario for the material retrieval method provided in this application. It should be noted that... Figure 1 The examples shown are merely examples of application scenarios where the material retrieval method of this application can be applied, to help those skilled in the art understand the technical content of this application, but do not mean that the embodiments of this application cannot be used in other devices, systems, environments or scenarios.

[0075] like Figure 1 As shown, the material retrieval system includes: user terminal 1, server 2, and database 3. User terminal 1 is any terminal used by a user with material retrieval needs, including but not limited to mobile terminals and computer terminals. Server 2 is the server that executes the material retrieval method, and database 3 is any database corresponding to server 2. Server 2 is communicatively connected to both user terminal 1 and database 3.

[0076] Based on this, server 2, through its communication connection with user terminal 1, determines and retrieves keywords based on the material retrieval request triggered by user terminal 1. It also retrieves a set of preset attribute values ​​containing attribute values ​​corresponding to multiple preset materials through its communication connection with database 3. Then, based on the correlation between the keywords and each attribute value, at least one attribute value with the highest correlation score is retrieved from the preset attribute value set. The total correlation score for each preset material is determined based on each attribute value with the highest correlation score, and at least one preset material with the highest total correlation score is sent to user terminal 1, whereby user terminal 1 displays the preset materials with the highest total correlation scores.

[0077] The number of keywords must be at least one. The association score is used to characterize the degree of association between the attribute value and the keyword.

[0078] The material retrieval method provided in this application pre-forms a set of preset attribute values ​​based on the attribute values ​​corresponding to multiple preset materials, and determines at least one keyword based on the material retrieval request triggered by the user. Therefore, by obtaining the keyword and the set of preset attribute values, the degree of correlation between each attribute value in the set and the keyword can be determined, forming a correlation score for each attribute value. Furthermore, at least one attribute value with the highest correlation score can be obtained from the set of preset attribute values. Thus, by determining the total correlation score for each preset material based on the attribute value with the highest correlation score, at least one preset material with the highest total correlation score can be sent to the user terminal, allowing the user terminal to display the preset materials with the highest total correlation scores. This eliminates the need for manual operation by the user in various selection components, improving retrieval efficiency and enhancing the user experience.

[0079] The technical solutions of this application and how they solve the aforementioned technical problems are described in detail below with specific embodiments. These specific embodiments may exist independently or in combination with each other. Identical or similar concepts or processes may not be repeated in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0080] Figure 2 The flow of the material retrieval method provided in the embodiments of this application Figure 1 The execution entity in this embodiment can be Figure 1 The server-side corresponding to the Chinese material retrieval system. For example... Figure 2 As shown, the material retrieval method provided in this embodiment includes:

[0081] S201. Obtain the keyword and preset attribute value set; the keywords are determined based on the material retrieval request triggered by the user; the preset attribute value set includes attribute values ​​corresponding to multiple preset materials; the number of keywords is at least one.

[0082] The keywords are words that describe the user's search needs for materials, specifically determined based on the material search request triggered by the user. This material search request is a request that instructs the server to retrieve materials.

[0083] The preset attribute value set is a pre-stored collection containing attribute values ​​corresponding to multiple preset materials. Furthermore, the preset materials are materials designed in advance according to product requirements and can be used to promote the product. In this embodiment, the preset materials can specifically be materials designed according to the requirements of a water heater, i.e., water heater materials.

[0084] The attribute values ​​are configured for the preset materials according to pre-defined attributes and their specific characteristics. These can include meta-attribute values ​​and custom attribute values. Meta-attribute values ​​indicate the format of the preset material, such as the format, duration, and orientation of the water heater material. Custom attribute values ​​indicate the content of the preset material, such as the specific display content and playback content of the water heater material.

[0085] It is understood that the above format can be video, image, audio, document, etc., the above duration can be 2 minutes, 0.5 hours, etc., the above direction can be left flip, right flip, etc., the above specific display content can be the color, shape, water capacity, etc. of the water heater, and the above specific playback content can be the development history of the water heater, instructions for use, etc.

[0086] In this embodiment, the user can input a piece of text on the user terminal in advance according to the required materials and trigger a material retrieval request. In response to the above request being triggered, at least one keyword is determined from the text input by the user.

[0087] Based on this, the aforementioned keywords are obtained, as well as a pre-stored set of preset attribute values.

[0088] S202. In the preset set of attribute values, obtain at least one attribute value with the highest association score based on the keyword; the association score is used to characterize the degree of association between the attribute value and the keyword.

[0089] Among them, the association score is a value that evaluates the similarity between attribute values ​​and user needs, specifically used to characterize the degree of association between attribute values ​​and keywords.

[0090] In this embodiment, after obtaining at least one keyword and a preset set of attribute values, the at least one keyword is treated as a whole, and the correlation score between each attribute value in the preset set of attribute values ​​and the at least one keyword is determined to obtain the correlation score corresponding to each attribute value. Then, according to the size of each correlation score, at least one attribute value with the highest correlation score is obtained.

[0091] It is understandable that each preset material has at least one corresponding attribute value. Therefore, after obtaining the correlation score corresponding to each attribute value, the overall correlation score of each preset material can be determined based on all the attribute values ​​of each preset material.

[0092] S203. Determine the total association score for each preset material based on the attribute value with the highest association score.

[0093] The total association score is the preset overall association score of the material based on the attribute value with the highest association score that has been determined.

[0094] In this embodiment, after obtaining the attribute value with the highest correlation score, the preset material to which the attribute value belongs is determined, and then all attribute values ​​corresponding to the preset material are determined. Then, based on the correlation scores corresponding to all the attribute values, the total correlation score corresponding to the preset material is calculated. Further, other preset materials are obtained, i.e., preset materials whose corresponding attribute values ​​are not among the attribute values ​​with the highest correlation scores, and the total correlation score of these preset materials is set to zero.

[0095] It is understandable that for other preset materials, none of their attribute values ​​match the user's needs and will not be provided to the user later. Therefore, there is no need to calculate their total relevance score, which can save computing resources and further improve retrieval efficiency.

[0096] S204. Send at least one preset material with the highest total association score to the user terminal so that the user terminal can display the preset material with the highest total association score.

[0097] In this embodiment, after obtaining the total association score corresponding to each preset material, at least one preset material with the highest total association score is obtained from the preset materials according to the size of each total association score, and these preset materials are sent to the user terminal in a preset format, so that the user terminal displays the preset material with the highest total association score on the corresponding operation interface, thereby completing the retrieval.

[0098] The preset format is a pre-configured format for displaying preset materials, which can be a list format, a small window format, etc. This application does not limit this.

[0099] The material retrieval method provided in this embodiment pre-forms a set of preset attribute values ​​based on the attribute values ​​corresponding to multiple preset materials, and determines at least one keyword based on the material retrieval request triggered by the user. Therefore, by obtaining the keyword and the set of preset attribute values, the degree of correlation between each attribute value in the set and the keyword can be determined, forming a correlation score for each attribute value. Furthermore, at least one attribute value with the highest correlation score can be obtained from the set of preset attribute values. Thus, by determining the total correlation score for each preset material based on the attribute value with the highest correlation score, at least one preset material with the highest total correlation score can be sent to the user terminal, allowing the user terminal to display the preset materials with the highest total correlation scores. This eliminates the need for manual operation by the user in various selection components, improving retrieval efficiency and enhancing the user experience.

[0100] Figure 3 The flow of the material retrieval method provided in the embodiments of this application Figure 2 The execution entity in this embodiment can be Figure 1 The user terminal corresponding to the material retrieval system. For example... Figure 3 As shown, the material retrieval method provided in this embodiment includes:

[0101] S301. In response to a user-triggered operation, generate a material retrieval request.

[0102] Specifically, the component containing the material retrieval request is triggered by the user through their user terminal, and the user terminal responds by generating the material retrieval request in response to the triggering of the aforementioned component.

[0103] S302. Send a material retrieval request to the server so that the server can obtain keywords and a set of preset attribute values. The set of preset attribute values ​​includes attribute values ​​corresponding to multiple preset materials. The number of keywords is at least one. In the set of preset attribute values, obtain at least one attribute value with the highest association score based on the keyword. The association score is used to characterize the degree of association between the attribute value and the keyword. Determine the total association score corresponding to each preset material based on the attribute value with the highest association score.

[0104] Specifically, after generating a material retrieval request, the user terminal sends the material retrieval request to the server through a communication connection, so that the server executes the corresponding material retrieval process based on the material retrieval request.

[0105] S303. Receive at least one preset material with the highest total association score sent by the server, and display each preset material with the highest total association score.

[0106] Understandably, after the server completes the corresponding material retrieval process, it will send at least one preset material with the highest total relevance score to the user terminal. Based on this, the user terminal receives the preset materials with the highest total relevance scores sent by the server and displays them on the corresponding operation interface.

[0107] The material retrieval method provided in this embodiment generates a material retrieval request when the user triggers a corresponding operation. This request is then sent to the server, enabling the server to successfully execute the subsequent material retrieval process based on the request, thus improving the success rate. Furthermore, by receiving at least one preset material with the highest total relevance score sent by the server, the preset materials with the highest total relevance scores can be displayed. This allows the user to identify the preset material with the highest total relevance score simply by viewing the displayed content, further improving the user experience.

[0108] Figure 4 This is a signaling interaction diagram of the material retrieval method provided in an embodiment of this application. This embodiment is... Figure 2 and Figure 3 Based on the corresponding embodiments, the material retrieval method will be described in detail. For example... Figure 4 As shown, the material retrieval method provided in this embodiment includes:

[0109] S401. The server obtains multiple preset materials and configures corresponding attribute values ​​for each preset material.

[0110] It is understandable that this solution relies on the premise that a set of preset attributes has already been configured and stored. Based on this, multiple pre-designed preset materials can be stored in corresponding storage media in advance. Then, based on configuration requirements, the pre-stored preset materials can be retrieved, and corresponding meta-attribute values ​​and custom attribute values ​​can be configured for each preset material.

[0111] For example, if the preset material is a video introducing model A water heater, the meta attribute values ​​configured for it can be "video", "2 minutes", "1080p", and the custom attribute values ​​can be "electric water heater", "white", "model A", "semi-cylindrical", "100-liter water capacity".

[0112] S402. The server constructs a mapping relationship between each preset material and its corresponding attribute value, and stores the attribute value corresponding to each preset material as a preset attribute value set.

[0113] In this embodiment, after obtaining each preset material and configuring the corresponding attribute value for each preset material, a mapping relationship is constructed between each preset material and the corresponding attribute value, and each attribute value is obtained. The same attribute value is merged into one, and the mapping relationship associated with the attribute value is retained. Then, a set of preset attribute values ​​is generated based on the merged attribute values.

[0114] Since this solution relies on storing a set of preset attribute values, by acquiring multiple preset materials, corresponding attribute values ​​can be configured for each material. Furthermore, by establishing a mapping relationship between each preset material and its corresponding attribute values, the attribute values ​​for each preset material can be stored as a preset attribute value set. This ensures the successful retrieval of the preset attribute value set, increasing the success rate of acquisition, and allows for rapid identification of the preset material to which an attribute value belongs based on this mapping relationship, thus improving the efficiency of identifying the correct preset material.

[0115] S403. The user terminal receives the material description text input by the user and generates a material retrieval request based on the material description text;

[0116] The material description text is the text that describes the material required by the user.

[0117] Specifically, the user prepares a text description based on specific material requirements and inputs this text through their terminal. Based on this, the terminal receives the user-inputted material description and, upon determining the component that triggered the material retrieval request, generates a material retrieval request based on the material description text.

[0118] S404. The user terminal sends a material retrieval request to the server.

[0119] The specific execution steps in this embodiment are similar to those in S302, and will not be repeated here.

[0120] S405. The server receives a material retrieval request sent by the user terminal.

[0121] The material retrieval request includes the material description text.

[0122] S406. The server obtains the material description text based on the material retrieval request and extracts at least one keyword from the material description text.

[0123] In this embodiment, after receiving the material retrieval request, the material description text contained in the material retrieval request is obtained, and the material description text is subjected to a series of preprocessing steps such as text cleaning, word segmentation, and part-of-speech tagging based on natural language processing technology (NLP). Then, at least one keyword is extracted from the preprocessing results.

[0124] For example, if the description text of the material is "Introduction video of a white water heater that can bathe 3-4 people.", then keywords such as "water heater", "80-100 liters of water capacity" and "white" can be extracted.

[0125] Since another prerequisite for implementing this solution is the identification of at least one keyword, the user terminal receives the user-input material description text and generates a material retrieval request based on it. This allows the server to retrieve the material description text from the retrieval request and extract at least one keyword from it. This ensures the subsequent successful retrieval of keywords, increasing the success rate of keyword acquisition.

[0126] S407. The server retrieves the set of keywords and preset attribute values.

[0127] In this embodiment, based on the determined keywords and the stored preset attribute value set, the keywords and the preset attribute value set are obtained. The specific execution steps are similar to S201, and will not be repeated here.

[0128] S408. The server performs vectorization processing on the keywords and each attribute value in the preset attribute value set to obtain vectorized keywords and each vectorized attribute value.

[0129] Vectorization is the process of converting non-numerical data (such as text, images, audio, etc.) into numerical data (i.e., vectors).

[0130] In this embodiment, based on the process of obtaining at least one attribute value with the highest association score from the preset attribute value set according to the keyword, after obtaining the keyword and the preset attribute value set, the keyword and each attribute value in the preset attribute value set are first vectorized, and the keyword and each attribute value are converted into a numerical vector representation to obtain vectorized keyword and each vectorized attribute value.

[0131] Vectorized keywords are keywords represented using numerical vectors. Vectorized attribute values ​​are attribute values ​​represented using numerical vectors.

[0132] S409. When the number of keywords is one, the server calculates the similarity between each vectorized attribute value and the vectorized keyword.

[0133] Understandably, there must be at least one keyword, and each keyword needs to be considered as a whole to evaluate its correlation with each attribute value. Therefore, the execution method differs depending on whether there is one keyword in the whole or multiple keywords in the whole.

[0134] Specifically, after obtaining the vectorized keywords and each vectorized attribute value, if the number of keywords is determined to be one, the cosine value of the angle between each vectorized attribute value and the vectorized keyword in the vector space is calculated. The closer the cosine value is to 1, the more similar the two are; the closer it is to -1, the less similar the two are. Thus, the similarity between each vectorized attribute value and the corresponding vectorized keyword is calculated.

[0135] S410. When there are multiple keywords, the server calculates the initial similarity between each vectorized attribute value and multiple vectorized keywords to obtain multiple initial similarities corresponding to each vectorized attribute value, and determines the sum of the multiple initial similarities as the similarity corresponding to each vectorized attribute value.

[0136] In this case, the initial similarity is determined solely based on the cosine value when there are multiple keywords.

[0137] In this embodiment, if the number of keywords is determined to be multiple, the initial similarity between each vectorized attribute value and the multiple vectorized keywords is calculated respectively. The specific execution method is similar to S307, and will not be repeated here.

[0138] Based on this, the initial similarities determined above are defined as multiple initial similarities corresponding to each vectorized attribute value, and the multiple initial similarities are summed to obtain the similarity corresponding to each vectorized attribute value.

[0139] Since there may be one or more keywords, when the number is determined to be multiple, the initial similarity between each vectorized attribute value and multiple vectorized keywords can be calculated separately to obtain the initial similarity corresponding to each vectorized attribute value. By calculating the sum of the corresponding initial similarities, the sum can be determined as the similarity corresponding to each vectorized attribute value. In this way, the similarity corresponding to each vectorized attribute value can be accurately calculated when there are multiple keywords, thus improving the accuracy of similarity calculation.

[0140] S411. The server determines the similarity corresponding to each vectorized attribute value as the association score corresponding to each attribute value.

[0141] In this embodiment, after obtaining the similarity corresponding to each vectorized attribute value, the corresponding similarity is used as the corresponding association score to obtain the association score corresponding to each attribute value.

[0142] S412. The server sorts each attribute value based on the corresponding association score, and obtains at least one attribute value with the highest corresponding association score based on the sorting result.

[0143] In this embodiment, after obtaining the correlation scores corresponding to each attribute value, the attribute values ​​are sorted in descending order or ascending order according to the corresponding correlation scores. Then, according to a preset number, at least one attribute value ranked first or at least one attribute value ranked last is obtained from the sorted attribute values.

[0144] The preset quantity is the number of attribute values ​​with the highest associated scores that are pre-configured by the administrator.

[0145] By vectorizing keywords and attribute values, vectorized keywords and attribute values ​​are obtained. The similarity between each vectorized attribute value and the vectorized keywords can then be calculated, and the similarity scores corresponding to each vectorized attribute value can be further determined as the association scores for each attribute value. By sorting the attribute values ​​based on their corresponding association scores, the attribute value with the highest association score can be obtained from the sorting results. This accurately determines each association score and systematically retrieves the attribute value with the highest association score, improving the accuracy of determining each association score and the attribute value with the highest association score.

[0146] S413. The server determines the preset material to which the attribute value with the highest associated score belongs, and obtains the associated score of the same preset material.

[0147] In this embodiment, the execution process is based on determining the total associated score of each preset material according to the attribute value with the highest associated score. After obtaining the attribute value with the highest associated score, the preset material to which the attribute value with the highest associated score belongs is determined based on the corresponding mapping relationship.

[0148] It is understandable that preset materials may have multiple attribute values. Therefore, among the attribute values ​​with the highest determined correlation scores, there may be at least two preset materials with the same highest correlation score.

[0149] Based on this, the attribute value with the highest associated score that is the same as the preset material is obtained, and its corresponding associated score is further obtained, that is, the associated score that is the same as the preset material is obtained.

[0150] S414. The server adds up the association scores of the same preset materials and determines the total association score of each preset material based on the processing result.

[0151] In this embodiment, after obtaining the association scores of the same preset materials, these association scores are summed to obtain a sum value, which is then determined as the total association score of the preset materials.

[0152] Understandably, for other preset materials, because their relevance to user needs is relatively low, their corresponding total relevance score is set to zero.

[0153] Since at least two attribute values ​​may belong to the same preset material, by determining the preset material to which the attribute value with the highest association score belongs, the association score for the same preset material can be obtained based on the determination result. By summing the association scores for the same preset material, the total association score for each preset material can be determined based on the sum. This improves the accuracy of calculating the total association score.

[0154] S415. The server sends at least one preset material with the highest total association score to the user terminal.

[0155] The specific execution steps in this embodiment are similar to those in S204, and will not be repeated here.

[0156] S416. The user terminal receives at least one preset material with the highest total association score sent by the server and displays each preset material with the highest total association score.

[0157] The specific execution steps in this embodiment are similar to those in S303, and will not be repeated here.

[0158] Figure 5 Schematic diagram of the material retrieval device provided in this application Figure 1 .like Figure 5 As shown, this application provides a material retrieval device, which is applied to a server. The material retrieval device 50 includes:

[0159] The acquisition module 51 is used to acquire keywords and a set of preset attribute values. Keywords are determined based on the material retrieval request triggered by the user. The set of preset attribute values ​​includes attribute values ​​corresponding to multiple preset materials. There is at least one keyword. In the set of preset attribute values, at least one attribute value with the highest correlation score is acquired based on the keyword. The correlation score is used to characterize the degree of correlation between the attribute value and the keyword.

[0160] Processing module 52 is used to determine the total association score corresponding to each preset material based on the attribute value with the highest association score.

[0161] The transceiver module 53 is used to send at least one preset material with the highest total association score to the user terminal so that the user terminal can display the preset material with the highest total association score.

[0162] Optionally, the acquisition module 41 is further configured to, when acquiring at least one attribute value with the highest association score from the preset attribute value set based on the keyword, perform vectorization processing on the keyword and each attribute value in the preset attribute value set to obtain vectorized keyword and each vectorized attribute value; calculate the similarity between each vectorized attribute value and the vectorized keyword respectively, and determine the similarity corresponding to each vectorized attribute value as the association score corresponding to each attribute value; sort each attribute value based on the corresponding association score, and acquire at least one attribute value with the highest corresponding association score based on the sorting result.

[0163] Optionally, the processing module 42 is further configured to, if there are multiple keywords, calculate the initial similarity between each vectorized attribute value and multiple vectorized keywords when calculating the similarity between each vectorized attribute value and the vectorized keywords, so as to obtain multiple initial similarities corresponding to each vectorized attribute value; and determine the sum of the multiple initial similarities as the similarity corresponding to each vectorized attribute value.

[0164] Optionally, the processing module 42 is further configured to determine the preset material to which the attribute value with the highest association score belongs when determining the total association score corresponding to each preset material based on the attribute value with the highest association score, and obtain the association score with the same attribute value; sum the association scores with the same attribute value belonging to the preset material, and determine the total association score corresponding to each preset material based on the processing result.

[0165] Optionally, attribute values ​​include meta attribute values ​​and custom attribute values; meta attribute values ​​are used to indicate the format of the preset material; custom attribute values ​​are used to indicate the content of the preset material.

[0166] Figure 6 Schematic diagram of the material retrieval device provided in this application Figure 2 .like Figure 6 As shown, this application provides a material retrieval device, which is applied to a user terminal. The material retrieval device 60 includes:

[0167] Processing module 61 is used to generate a material retrieval request in response to a user-triggered operation;

[0168] The transceiver module 62 is used to send a material retrieval request to the server so that the server can obtain keywords and a set of preset attribute values. The set of preset attribute values ​​includes attribute values ​​corresponding to multiple preset materials. There is at least one keyword. In the set of preset attribute values, at least one attribute value with the highest correlation score is obtained based on the keyword. The correlation score is used to characterize the degree of correlation between the attribute value and the keyword. The total correlation score corresponding to each preset material is determined based on the attribute value with the highest correlation score.

[0169] The transceiver module 62 is also used to receive at least one preset material with the highest total association score sent by the server.

[0170] The processing module 61 is also used to display the preset material with the highest total associated score.

[0171] Optionally, the processing module 61 is further configured to receive the material description text input by the user when generating a material retrieval request in response to a user-triggered operation, and generate a material retrieval request based on the material description text; wherein, the server is configured to obtain the material description text based on the material retrieval request, and extract at least one keyword from the material description text.

[0172] Figure 7 A schematic diagram of the server-side device provided in this application. Figure 7 As shown, this application provides a server-side device 70, which includes a receiver 71, a transmitter 72, a processor 73, and a memory 74.

[0173] Receiver 71 is used to receive instructions and data;

[0174] Transmitter 72 is used to send commands and data;

[0175] Memory 74 is used to store instructions executed by the computer;

[0176] The processor 73 is used to execute computer execution instructions stored in the memory 74 to implement the various steps of the material retrieval method in the above embodiments. For details, please refer to the relevant descriptions in the foregoing embodiments of the material retrieval method.

[0177] Alternatively, the memory 74 can be either standalone or integrated with the processor 73.

[0178] When the memory 74 is configured independently, the server-side device also includes a bus for connecting the memory 74 and the processor 73.

[0179] Figure 8 A schematic diagram of the user terminal equipment provided in this application. Figure 8 As shown, this application provides a user terminal device 80, which includes: a receiver 81, a transmitter 82, a processor 83, and a memory 84.

[0180] Receiver 81 is used to receive instructions and data;

[0181] Transmitter 82 is used to send commands and data;

[0182] Memory 84 is used to store instructions executed by the computer;

[0183] The processor 83 is used to execute computer execution instructions stored in the memory 84 to implement the various steps of the material retrieval method in the above embodiments. For details, please refer to the relevant descriptions in the foregoing embodiments of the material retrieval method.

[0184] Alternatively, the memory 84 can be either standalone or integrated with the processor 83.

[0185] When the memory 84 is configured independently, the user terminal device also includes a bus for connecting the memory 84 and the processor 83.

[0186] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the material retrieval method performed by the server-side device and the user terminal device described above.

[0187] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the material retrieval method as described above for server-side devices and user terminal devices.

[0188] It will be understood by those skilled in the art that all or some of the steps, systems, or apparatuses disclosed above, and their functional modules / units, can be implemented as software, firmware, hardware, or suitable combinations thereof. In hardware implementations, the division between functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed collaboratively by several physical components. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit (ASIC). Such software may be distributed on a computer-readable medium, which may include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and can be accessed by a computer. Furthermore, it is well known to those skilled in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0189] The technical solutions of this application have been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it is readily understood by those skilled in the art that the scope of protection of this application is obviously not limited to these specific embodiments. The above embodiments are only used to illustrate the technical solutions of this application and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. These modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A material retrieval method, characterized in that, The method includes: Obtain keywords and a set of preset attribute values; the keywords are determined based on the material retrieval request triggered by the user; the set of preset attribute values ​​includes attribute values ​​corresponding to multiple preset materials; the number of keywords is at least one; In the preset set of attribute values, at least one attribute value with the highest association score is obtained based on the keyword; the association score is used to characterize the degree of association between the attribute value and the keyword. The total association score for each preset material is determined based on the attribute value with the highest association score. Send at least one preset material with the highest total association score to the user terminal so that the user terminal displays the preset material with the highest total association score.

2. The method according to claim 1, characterized in that, The step of obtaining at least one attribute value with the highest association score from the preset attribute value set based on the keyword includes: The keywords and each attribute value in the preset attribute value set are vectorized to obtain vectorized keywords and each vectorized attribute value. Calculate the similarity between each vectorized attribute value and the vectorized keyword, and determine the similarity corresponding to each vectorized attribute value as the association score corresponding to each attribute value; Sort the attribute values ​​based on their corresponding correlation scores, and obtain at least one attribute value with the highest corresponding correlation score based on the sorting results.

3. The method according to claim 2, characterized in that, If there are multiple keywords, then calculating the similarity between each vectorized attribute value and the vectorized keyword includes: Calculate the initial similarity between each vectorized attribute value and multiple vectorized keywords to obtain multiple initial similarities corresponding to each vectorized attribute value; The sum of multiple initial similarities is used to determine the similarity corresponding to each vectorized attribute value.

4. The method according to claim 1, characterized in that, The process of determining the total association score for each preset material based on the attribute value with the highest association score includes: Determine the preset material to which the attribute value with the highest associated score belongs, and obtain the associated scores of the preset materials that belong to the same preset material; The association scores of the same preset materials are summed, and the total association score of each preset material is determined based on the processing result.

5. The method according to any one of claims 1-4, characterized in that, The attribute values ​​include meta attribute values ​​and custom attribute values; The meta-attribute value is used to indicate the form of the preset material; the custom attribute value is used to indicate the content of the preset material.

6. A material retrieval method, characterized in that, The method includes: In response to a user-triggered action, a material retrieval request is generated; The material retrieval request is sent to the server so that the server can obtain keywords and a set of preset attribute values; the set of preset attribute values ​​includes attribute values ​​corresponding to multiple preset materials; the number of keywords is at least one; in the set of preset attribute values, at least one attribute value with the highest correlation score is obtained according to the keyword; the correlation score is used to characterize the degree of correlation between the attribute value and the keyword; the total correlation score corresponding to each preset material is determined according to the attribute value with the highest correlation score. Receive at least one preset material with the highest total association score sent by the server, and display each preset material with the highest total association score.

7. The method according to any one of claims 1-4, characterized in that, In response to a user-triggered action, a material retrieval request is generated, specifically including: The system receives material description text input by the user and generates a material retrieval request based on the material description text; wherein, the server is used to obtain the material description text according to the material retrieval request and extract at least one keyword from the material description text.

8. A server-side device, characterized in that, include: Memory; processor; The memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to enable the server-side device to implement the method as described in any one of claims 1-5.

9. A user terminal device, characterized in that, include: Memory; processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to cause the user terminal device to implement the method as described in any one of claims 6-7.

10. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method as described in any one of claims 1-7.

Images