Product Recommendation Method, Device, Electronic Device, and Storage Medium

Abstract

The present application relates to a commodity recommendation method, device, electronic device, and storage medium, which are applied to the field of intelligent recommendation technology. Among them, the method includes: obtaining the behavior data of a user; when it is determined that the user needs to recommend a commodity, determining the commodity attributes to be recommended according to the user behavior data; determining the target commodity associated with the commodity attributes from a pre-constructed knowledge graph; and recommending the target commodity to the user. This is to solve the problem in the prior art that the accuracy and success rate of commodity recommendation are relatively low, reducing the user experience.

Description

Technical Field

[0001] This application relates to the field of intelligent recommendation technology, and in particular, to a commodity recommendation method, apparatus, electronic device, and storage medium. Background Art

[0002] With the improvement of people's living standards, more and more commodity products are being used. Currently, major manufacturers in the market have launched numerous electrical equipment with different functions and specifications for selection. However, how to select the equipment that meets the user's needs from the dazzling array of commodities often requires the user to spend a lot of time selecting and comparing different products. This process is cumbersome and time-consuming, and sometimes even a large amount of time is spent but still a satisfactory product cannot be selected.

[0003] Generally, the sales of commodities are on public or proprietary e-commerce platforms, and the commodity information in the e-commerce platforms is very messy. In the related art, the recommendation of commodities mostly recommends the same or similar commodities to users. For example, after a user purchases an air conditioner, the e-commerce platform will recommend air conditioners of other categories or brands. However, the possibility of an ordinary user purchasing an air conditioner again is relatively low, resulting in low accuracy and success rate of commodity recommendation and reducing the user experience. Summary of the Invention

[0004] This application provides a commodity recommendation method, apparatus, electronic device, and storage medium to solve the problem in the prior art that the accuracy and success rate of commodity recommendation are relatively low, reducing the user experience.

[0005] In a first aspect, an embodiment of this application provides a commodity recommendation method, including:

[0006] Obtain the user's behavior data;

[0007] When it is determined that the user needs to recommend a commodity, determine the commodity attributes to be recommended according to the user behavior data;

[0008] Determine the target commodity associated with the commodity attributes from a pre-constructed knowledge graph;

[0009] Recommend the target commodity to the user.

[0010] Optionally, the process of constructing the knowledge graph includes:

[0011] Obtain a commodity data set;

[0012] Compare the commodity data in the commodity data set with a preset node set to determine the node information in the commodity data, and the preset node set includes commodity names and commodity attributes;

[0013] Obtain the relationship information among the node information of each node in the preset node set;

[0014] Construct the knowledge graph according to the node information and the relationship information.

[0015] Optionally, determining the target commodity associated with the commodity attribute from the pre-constructed knowledge graph includes:

[0016] Determine the target node information corresponding to the commodity attribute from the knowledge graph;

[0017] Determine the commodity associated with the target node information in the knowledge graph as the target commodity.

[0018] Optionally, the behavior data includes voice information or text information, and determining the commodity attribute to be recommended according to the user behavior data includes:

[0019] Perform semantic analysis on the voice information or text information to determine the control instruction indicated in the voice information or text information;

[0020] Extract the control information in the control instruction and use the control information as the commodity attribute.

[0021] Optionally, determining that the user needs to recommend a commodity includes:

[0022] Perform sentiment analysis on the behavior data to obtain the user's sentiment information;

[0023] If the sentiment information is a positive emotion, it is determined that the user needs to recommend a commodity.

[0024] Optionally, determining that the user needs to recommend a commodity includes:

[0025] When it is determined according to the behavior data that the currently used commodity does not meet the user's usage requirements, it is determined that the user needs to recommend a commodity.

[0026] Optionally, determining that it does not meet the user's usage requirements according to the behavior data includes:

[0027] Determine the control instruction indicated by the behavior data obtained each time;

[0028] Determine the repetition times of the same control instruction within a preset duration;

[0029] If the repetition coefficient is greater than the preset number of times, it is determined that the currently used commodity does not meet the user's usage requirements.

[0030] In a second aspect, an embodiment of the present application provides a commodity recommendation device, including:

[0031] An acquisition module, configured to acquire the behavior data of a user;

[0032] When it is determined that the user needs to recommend a product, determine the product attributes to be recommended according to the user behavior data;

[0033] Determine the target product associated with the product attributes from a pre-constructed knowledge graph;

[0034] Recommend the target product to the user.

[0035] Optionally, the product recommendation device further includes:

[0036] A first acquisition unit, configured to acquire a product data set;

[0037] A first determination unit, configured to compare the product data in the product data set with a preset node set to determine the node information in the product data, where the preset node set includes product categories and product functions;

[0038] A second acquisition unit, configured to acquire the relationship information between the node information in the preset node set;

[0039] A construction unit, configured to construct a knowledge graph according to the node information and the relationship information.

[0040] Optionally, the second determination module includes:

[0041] A second determination unit, configured to determine the target node information corresponding to the product attributes from the knowledge graph;

[0042] A third determination unit, configured to determine the product associated with the target node information in the knowledge graph as the target product.

[0043] Optionally, the behavior data includes voice information or text information, and the first determination module includes:

[0044] A fourth determination unit, configured to perform semantic analysis on the voice information or text information to determine the control instruction indicated in the voice information or text information;

[0045] An extraction unit, configured to extract the control information in the control instruction and use the control information as the product attributes.

[0046] Optionally, the first determination module includes:

[0047] An analysis unit, configured to perform sentiment analysis on the behavior data to obtain the sentiment information of the user;

[0048] A fifth determination unit, configured to determine that the user needs to recommend a product if the sentiment information is a positive emotion.

[0049] Optionally, the first determination module includes:

[0050] A sixth determination unit, configured to determine that the user needs to recommend a product when it is determined according to the behavior data that the user's usage requirements are not met.

[0051] Optionally, the sixth determination unit includes:

[0052] A seventh determination unit, configured to determine the control instruction indicated by the behavior data obtained each time;

[0053] An eighth determination unit, configured to determine the repetition times of the same control instruction within a preset duration;

[0054] A ninth determination unit, configured to determine that the user's usage requirements are not met if the repetition times are greater than a preset number of times.

[0055] In a third aspect, an embodiment of the present application provides an electronic device, including: a processor, a communication interface, a memory, and a communication bus, where the processor, the communication interface, and the memory complete communication with each other through the communication bus;

[0056] The memory is configured to store a computer program;

[0057] The processor is configured to execute the program stored in the memory to implement the product recommendation method in the first aspect.

[0058] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, storing a computer program, where the computer program, when executed by a processor, implements the product recommendation method in the first aspect.

[0059] The above technical solutions provided by the embodiments of the present application have the following advantages compared with the prior art: In the method provided by the embodiments of the present application, by obtaining the user's behavior data; when it is determined that the user needs to recommend a product, determining the product attributes to be recommended according to the user behavior data; determining the target product associated with the product attributes from the pre-constructed knowledge graph; and recommending the target product to the user. In this way, the product attributes to be recommended are determined through the user's behavior data, so that the target product associated with the product attributes can be determined through the pre-constructed knowledge graph, avoiding the way of repeatedly recommending the same type of product to the user, improving the accuracy of product recommendation, and enhancing the user experience. Description of the Drawings

[0060] The drawings here are incorporated into the description and form a part of this description, showing the embodiments consistent with the present invention and used together with the description to explain the principles of the present invention.

[0061] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0062] Figure 1 It is an application scenario diagram of the commodity recommendation method provided by an embodiment of the present application;

[0063] Figure 2 It is a flowchart of the commodity recommendation method provided by an embodiment of the present application;

[0064] Figure 3 It is a structural diagram of the commodity recommendation device provided by an embodiment of the present application;

[0065] Figure 4 It is a structural diagram of the electronic device provided by an embodiment of the present application. Detailed implementation manners

[0066] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some, rather than all, of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts fall within the scope of protection of the present application.

[0067] Before further elaborating on the embodiments of the present invention, the nouns and terms involved in the embodiments of the present invention are explained. The nouns and terms involved in the embodiments of the present invention are applicable to the following explanations.

[0068] A knowledge graph is a modern theory that combines the theories and methods of disciplines such as applied mathematics, graphics, information visualization technology, and information science with methods such as bibliometric citation analysis and co-occurrence analysis, and uses a visual graph to vividly display the core structure, development history, frontier fields, and overall knowledge architecture of a discipline to achieve the purpose of multi-disciplinary integration. The three elements that make up a knowledge graph include: entities, relationships, and attributes. Entities: refer to objectively existing and distinguishable things, which can be specific people, things, or objects, or abstract concepts or connections. Entities are the most basic elements in a knowledge graph. Relationships: In a knowledge graph, edges represent the relationships in the knowledge graph and are used to represent a certain connection between different entities. Attributes: Both entities and relationships in a knowledge graph can have their own attributes.

[0069] According to an embodiment of the present application, a commodity recommendation method is provided. Optionally, in the embodiment of the present application, the above commodity recommendation method can be applied to, for exampleFigure 1 In the hardware environment composed of the terminal 101 and the server 102 as shown. As Figure 1 shown, the server 102 is connected to the terminal 101 through a network and can be used to provide services (such as video services, application services, etc.) for the terminal or the client installed on the terminal. A database can be set up on the server or independently of the server to provide data storage services for the server 102. The above network includes but is not limited to: wide area network, metropolitan area network or local area network. The terminal 101 is not limited to a PC, mobile phone, tablet computer, home appliance device, etc.

[0070] The commodity recommendation method in the embodiment of the present application can be executed by the server 102, or can be executed by the terminal 101, or can also be jointly executed by the server 102 and the terminal 101. Among them, when the terminal 101 executes the commodity recommendation method in the embodiment of the present application, it can also be executed by the client installed on it.

[0071] Taking the server executing the commodity recommendation method in the embodiment of the present application as an example, Figure 2 is a schematic flowchart of an optional commodity recommendation method according to an embodiment of the present application. As Figure 2 shown, the process of this method can include the following steps:

[0072] Step 201, obtain the behavior data of the user.

[0073] In some embodiments, the behavior data of the user can be collected by the home appliance device and sent to the server. For example, a behavior collection unit is set on the home appliance device, and the user behavior is collected through this behavior collection unit.

[0074] Among them, the behavior data of the user can include but is not limited to voice information, text information, video information, and / or image information. The corresponding behavior collection unit on the home appliance device includes a voice collection unit, a text collection unit, a video collection unit, and / or an image collection unit.

[0075] Step 202, when it is determined that the user needs to recommend a commodity, determine the commodity attributes to be recommended according to the user behavior data.

[0076] In some embodiments, after the home appliance device obtains the behavior data of the user, it will perform corresponding analysis on the behavior data of the user, so as to determine the user's intention from the behavior data of the user. Among them, the commodity attribute can be information representing the function of the commodity, or information representing the performance of the commodity (such as energy consumption, power, lifespan).

[0077] Exemplarily, taking the behavior data of the user as voice information as an example, when the voice information is "too hot", it can be known that the actual demand of the user is to cool down. Therefore, "cooling down" is determined as the commodity attribute to be recommended.

[0078] In an alternative embodiment, the behavior data includes voice information or text information. Determining the product attributes to be recommended based on the user behavior data includes:

[0079] Performing semantic analysis on the voice information or text information to determine the control instructions indicated in the voice information or text information; extracting the control information in the control instructions and using the control information as the product attributes.

[0080] Specifically, after the server obtains the voice information or text information, it can first convert the voice information into text information, and determine the control instructions indicated in the voice information or text information by performing semantic analysis on the text information. After obtaining the control instructions, the product attributes are obtained by extracting the control information in the control instructions.

[0081] Among them, there are various ways to convert voice information into text information. For example, the Automatic Speech Recognition technology can be used to convert the user's voice information into text information.

[0082] Exemplarily, taking the voice information as "too hot", the control instruction indicated by this voice information is determined as a cooling instruction through semantic analysis. By extracting "cooling" in the cooling instruction and using it as the product attribute, thus, when recommending products, products that can achieve cooling are recommended.

[0083] Among them, for the way of performing semantic analysis on the text information, the method of text classification can be adopted. First, preprocess the text information (the preprocessing mainly includes word segmentation, removing conjunctions and adverbs), feature extraction (adopting the TF-IDF text feature extraction method), and classification according to the features (by constructing a classification model in advance), so as to determine the control instructions indicated in the voice information or text information. It can be understood that the semantic analysis methods provided in related technologies can also be adopted, which is not limited here.

[0084] In some embodiments, determining that the user needs to recommend products can be obtained after the user triggers relevant buttons on home appliances. A product information search box can be set on the home appliances. After the user inputs relevant product information and triggers the search, it is determined that the user needs to recommend products. Or, it can be determined that the user needs to recommend products through the user's behavior data.

[0085] In an alternative embodiment, determining that the user needs to recommend products includes:

[0086] Performing sentiment analysis on the behavior data to obtain the user's sentiment information; if the sentiment information is a positive emotion, determining that the user needs to recommend products.

[0087] In some embodiments, for sentiment analysis of behavioral data, a multimodal sentiment analysis method can be adopted to determine the user's sentiment information. For example, the behavioral data includes text information and image information. By performing multimodal sentiment analysis on the text information and image information, the user's sentiment information can be obtained. Alternatively, the user's image information can be directly collected, and the user's sentiment information can be determined through image recognition. For example, if the user's expression in the collected image is that the corners of the mouth are turned up, it is determined that the user's sentiment information is positive emotion.

[0088] In an alternative embodiment, determining that the user needs to recommend products includes:

[0089] When it is determined according to the behavioral data that the user's usage requirements are not met, it is determined that the user needs to recommend products.

[0090] In some embodiments, the server determines whether the product currently used by the user meets its usage requirements by obtaining the user's behavioral data. After the user's usage requirements are not met, it is determined that the user needs to recommend products.

[0091] Among them, there are various ways to determine that the currently used product does not meet the user's usage requirements. For example, through the following ways:

[0092] Determine the control instruction indicated by the behavioral data obtained each time; determine the repetition times of the same control instruction within a preset time period; if the repetition times are greater than the preset times, it is determined that the user's usage requirements are not met.

[0093] In some embodiments, after the server obtains the user's behavioral data each time, it determines the control instruction indicated by the user's behavioral data according to the user's behavioral data, and sends the control instruction to the currently used product to perform the operation corresponding to the control instruction. Within a preset time period, the user may generate behavioral data multiple times, thereby controlling the currently used product multiple times. If the user issues more control instructions than the preset times within the preset time period, it can be considered that the currently used product can no longer meet the user's usage requirements.

[0094] Exemplarily, taking the currently used product as an air conditioner, the user said "too hot" 4 times within half an hour through voice, exceeding the preset times three times. After the first three voice messages were sent, a cooling instruction was issued to the air conditioner each time. If the user still issues a cooling instruction after three times, it means that the current air conditioner no longer meets the user's usage requirements.

[0095] Step 203: Determine the target product associated with the product attribute from the pre-constructed knowledge graph.

[0096] In some embodiments, after determining the product attribute, the target product associated with the product attribute can be determined from the pre-constructed knowledge graph.

[0097] Among them, it is constructed in the knowledge graph according to information such as product attributes and product names. Therefore, after determining the product attributes, the target product associated with the product attributes can be determined from the knowledge graph.

[0098] In an alternative embodiment, the process of constructing the knowledge graph includes:

[0099] Obtain a set of product data; compare the product data in the set of product data with a preset set of nodes to determine the node information in the product data, where the preset set of nodes includes product categories and product functions; obtain the relationship information between the node information in the preset set of nodes; construct a knowledge graph based on the node information and the relationship information.

[0100] Among them, the set of product data can be obtained from an e-commerce platform or a database within a product company. Usually, the product data includes product names, product prices, product performances, product attributes, etc.

[0101] Among them, price, after-sales service, and energy efficiency are the attributes of the device itself and can also be used as the attributes for creating nodes. Therefore, the preset nodes in the graph include: device nodes and function nodes. Among them, the device nodes contain three attributes: price, after-sales service, and energy efficiency. For example, the price of an air conditioner is 3000, the after-sales service is 10 years, and the energy efficiency is level 1; the attributes in the function nodes are the explanations of the functions.

[0102] It should be noted that the preset set of nodes in this application is the set of entities in the knowledge graph, the node information is the entity in the knowledge graph, and the relationship information is the relationship between entities.

[0103] After obtaining the set of product data, each product data in it is compared with the preset nodes, and thus, the data in the product data that is the same as the preset nodes is used as the node information of the product data.

[0104] Based on the above related embodiments, after determining the node information in each product data, the relationship information between the nodes in the product data can be determined through the relationship information between the obtained node information, and thus, a knowledge graph can be constructed based on the node information, relationship information, and their attribute information in the product data.

[0105] Among them, the attribute information of the node information and the relationship information can be obtained from the product data. For example, the product data is the product data of an air conditioner, including "5-horsepower cooling and heating floor-standing cabinet type, fixed speed 380V for shops and offices, household and commercial air conditioner cabinets, dual functions of cooling and heating, and first-level energy efficiency". Then, the air conditioner cabinet and cooling and heating in it are the node information, and 5 horsepower, 380V, shop, office, and household are all attribute information. Usually, an air conditioner can achieve cooling and heating. Therefore, "achieve" can be used as the relationship information between the air conditioner cabinet and cooling and heating.

[0106] In an optional embodiment, determining a target product associated with product attributes from a pre-constructed knowledge graph includes:

[0107] Determining, from the knowledge graph, target node information corresponding to the product attributes; and determining, as the target product, the product associated with the target node information in the knowledge graph.

[0108] In some embodiments, after obtaining the product attributes, the target node information consistent with the product attributes can be determined from the pre-constructed knowledge graph according to the product attributes, and then the product associated with the target node information can be found from the knowledge graph.

[0109] Step 204: Recommend the target product to the user.

[0110] In some embodiments, after determining the target product, the target product can be recommended to the user.

[0111] Among them, there are various ways to recommend the target product to the user. For example, the name of the target product can be played through the currently used product by means of voice broadcast; or, the product information of the target product can be sent to the user's home device for display through the home device for the user to select and view.

[0112] Generally, the interaction entry of smart home is voice. After the user purchases a device with a voice entry, recommendations for other devices can be made. In a specific embodiment, the product recommendation method of the present application mainly includes: construction of a smart home knowledge graph; triggering of a recommendation system; and introduction of an interpretable recommendation to provide the user with a suitable reason for recommendation. The following will be introduced in three parts:

[0113] (I) Construction of a smart home knowledge graph

[0114] A complete smart home knowledge graph contains a lot of professional knowledge, but much of this knowledge is unnecessary for users. Generally speaking, the attributes that users care about are as follows: function, price, after-sales service, and energy efficiency. Among them, price, after-sales service, and energy efficiency are the attributes of the device itself and are also used as the attributes for node creation. Therefore, the nodes in the graph include: device nodes (including three attributes: price, after-sales service, and energy efficiency. For example, the price of an air conditioner is 3000 yuan, the after-sales service is 10 years, and it has a first-level energy efficiency); function nodes (the attribute is the explanation of the function. For example, humidification - a humidifier can make the indoor air less dry and protect your respiratory tract). The relationships include the relationships between devices and functions. For example, a humidifier - humidification, a fan - cooling. There are also devices with multiple functions. For example, an air conditioner - refrigeration, heating, and air sweeping. There are also complementary relationships (assistance, energy saving) between functions. For example, "air conditioner refrigeration" will be faster and can also reduce the power consumption of the air conditioner if there are "automatic curtains blocking sunlight" and "fan blowing".

[0115] When building the knowledge graph, function nodes are established first, and then the connections between devices and functions are established. For example, both air conditioners and fans have the function of cooling. Curtains can block sunlight in summer and also assist in cooling. When the user reveals the need for cooling, the corresponding function tags can be found from the knowledge graph, and then the corresponding products can be found according to the functions.

[0116] (2) Trigger condition setting

[0117] The trigger conditions can be considered from the following aspects: The first one: Trigger according to the user's mood. Multimodal (including voice information, text information, and image information) sentiment analysis can be introduced. Only when the user's emotion is detected as positive can the recommendation be triggered. The second one: According to the user's usage needs, corresponding recommendations are made only when the current device does not meet the user's needs. For example, when the user says it's too hot, if the current device can meet the need by lowering the temperature, there is no need to make a recommendation. When the user repeatedly emphasizes that it's too hot and the current device cannot meet the need or the current device can meet the need but the power consumption cost is too high, corresponding solutions can be recommended, such as recommending "automatic curtains + fan".

[0118] Among them, from the perspective of semantic understanding, "it's too hot" can be regarded as a synonym for "lower the temperature". However, when the user repeatedly emphasizes the same instruction, it must be a negative emotion. At this time, it is not very appropriate to emphasize the emotion anymore. It is more appropriate to directly solve the problem and give effective suggestions.

[0119] (3) Explainable recommendation

[0120] An interpretable recommendation provides a reason for the recommendation to the user, which can be achieved by combining the relationships between functions and the attributes of the functions and devices. It is also possible to pre-set a lot of explanatory response messages, which can be triggered when specific scenarios are met. For example, if the user has a need to cool down, and after querying the knowledge graph, there are three devices with a cooling function, namely an air conditioner, a fan, and a curtain. Among them, the air conditioner and the fan have a direct cooling function, and the curtain has an auxiliary cooling function. The response message can be set as follows: Using the A air conditioner and the B fan together can achieve a better cooling effect. You may wish to give it a try. In addition, you can also use the C curtain to assist in cooling.

[0121] It should be noted that the above-mentioned acquisition of the user's behavior data is obtained under the condition that the user has given authorization.

[0122] Based on the same concept, an embodiment of the present application provides a commodity recommendation device. The specific implementation of this device can be referred to the description in the method embodiment part, and the repeated parts will not be elaborated. As Figure 3 shown, this device mainly includes:

[0123] An acquisition module 301, configured to acquire the user's behavior data;

[0124] A first determination module 302, configured to determine the attributes of the commodity to be recommended according to the user's behavior data when it is determined that the user needs to recommend a commodity;

[0125] A second determination module 303, configured to determine the target commodity associated with the commodity attributes from a pre-constructed knowledge graph;

[0126] A recommendation module 304, configured to recommend the target commodity to the user.

[0127] Based on the same concept, an embodiment of the present application also provides an electronic device. As Figure 4 shown, this electronic device mainly includes: a processor 401, a memory 402, and a communication bus 403. Among them, the processor 401 and the memory 402 complete mutual communication through the communication bus 403. Among them, a program executable by the processor 401 is stored in the memory 402, and the processor 401 executes the program stored in the memory 402 to implement the following steps:

[0128] Acquire the user's behavior data;

[0129] When it is determined that the user needs to recommend a commodity, determine the attributes of the commodity to be recommended according to the user's behavior data;

[0130] Determine the target commodity associated with the commodity attributes from a pre-constructed knowledge graph;

[0131] Recommend the target commodity to the user.

[0132] The communication bus 403 mentioned in the above electronic device may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The communication bus 403 can be divided into an address bus, a data bus, a control bus, etc. For the sake of convenience of representation, Figure 4 it is only represented by a thick line in the figure, but it does not mean that there is only one bus or one type of bus.

[0133] The memory 402 may include a Random Access Memory (RAM), or may also include a non-volatile memory, such as at least one disk memory. Optionally, the memory may also be at least one storage device located far from the aforementioned processor 401.

[0134] The aforementioned processor 401 may be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc., or may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0135] In another embodiment of the present application, a computer-readable storage medium is further provided. A computer program is stored in the computer-readable storage medium. When the computer program runs on a computer, the computer is enabled to execute the product recommendation method described in the above embodiment.

[0136] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a computer, the processes or functions according to the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions are transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wire (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, microwave, etc.). The computer-readable storage medium can be any available medium that the computer can access, or a data storage device such as a server or data center that includes one or more integrated available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, magnetic tape, etc.), an optical medium (such as a DVD), or a semiconductor medium (such as a solid-state drive), etc.

[0137] It should be noted that in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise", or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or elements inherent to such process, method, article, or device. Without further limitation, an element defined by the statement "including a..." does not exclude the existence of additional identical elements in the process, method, article, or device including the element.

[0138] The above are only specific embodiments of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather to the broadest scope consistent with the principles and novel features claimed herein.

Claims

1. A commodity recommendation method, characterized in that, it includes: Obtain the user's behavior data; When it is determined that the user needs to recommend a commodity, determine the commodity attributes to be recommended according to the user behavior data; wherein, the behavior data includes voice information or text information, and determining the commodity attributes to be recommended according to the user behavior data includes: performing semantic analysis on the voice information or text information to determine the control instructions indicated in the voice information or text information; extracting the control information in the control instructions and using the control information as the commodity attributes; Determine the target commodity associated with the commodity attributes from a pre-constructed knowledge graph; Recommend the target commodity to the user; Wherein, determining that the user needs to recommend a commodity includes: when it is determined according to the behavior data that the currently used commodity does not meet the user's usage requirements, it is determined that the user needs to recommend a commodity; determining that the currently used commodity does not meet the user's usage requirements according to the behavior data includes: determining the control instructions indicated by the behavior data obtained each time; determining the repetition times of the same control instructions within a preset time period; if the repetition times are greater than the preset times, it is determined that the user's usage requirements are not met.

2. The commodity recommendation method according to claim 1, characterized in that, The process of constructing the knowledge graph includes: Obtain a commodity data set; Compare the commodity data in the commodity data set with a preset node set to determine the node information in the commodity data, and the preset node set includes commodity names and commodity attributes; Obtain the relationship information between the node information in the preset node set; Construct the knowledge graph according to the node information and the relationship information.

3. The commodity recommendation method according to claim 1 or 2, characterized in that, Determining the target commodity associated with the commodity attributes from a pre-constructed knowledge graph includes: Determine the target node information corresponding to the commodity attributes from the knowledge graph; Determine the commodity associated with the target node information in the knowledge graph as the target commodity.

4. The commodity recommendation method according to claim 1, characterized in that, Determining that the user needs to recommend a commodity includes: Performing sentiment analysis on the behavior data to obtain the user's sentiment information; If the sentiment information is a positive emotion, it is determined that the user needs to recommend a commodity.

5. A commodity recommendation device, characterized in that, it includes: An acquisition module for acquiring the user's behavior data; When it is determined that the user needs to recommend a commodity, determine the commodity attributes to be recommended according to the user behavior data; wherein, the behavior data includes voice information or text information, and determining the commodity attributes to be recommended according to the user behavior data includes: performing semantic analysis on the voice information or text information to determine the control instructions indicated in the voice information or text information; extracting the control information in the control instructions and using the control information as the commodity attributes; Determine the target commodity associated with the commodity attributes from a pre-constructed knowledge graph; Recommend the target product to the user; Among them, determining that the user needs to recommend a product includes: when it is determined according to the behavior data that the currently used product does not meet the user's usage requirements, it is determined that the user needs to recommend a product; determining that the currently used product does not meet the user's usage requirements according to the behavior data includes: determining the control instructions indicated by the behavior data obtained each time; determining the repetition times of the same control instructions within a preset time period; if the repetition times are greater than the preset times, it is determined that the user's usage requirements are not met.

6. An electronic device Characterized in that It includes: A processor, a communication interface, a memory, and a communication bus. Among them, the processor, the communication interface, and the memory complete mutual communication through the communication bus; The memory is used to store computer programs; The processor is used to execute the programs stored in the memory to implement the product recommendation method according to any one of claims 1-4.

7. A computer-readable storage medium storing a computer program Characterized in that When the computer program is executed by a processor, it implements the product recommendation method according to any one of claims 1-4.

Images