A method and apparatus for predicting product recommendations based on a dual-model user sentiment

Abstract

This invention provides a user sentiment prediction and product recommendation method, virtual device, computer-readable storage medium, and electronic device based on a dual-model approach, relating to the technical field of shared electric vehicle technology. The method includes: acquiring first image information of a user; performing sentiment information determination processing on the first image information using a preset first model to determine the user's sentiment tag and sentiment level; performing product matching processing on the sentiment tag and sentiment level using a preset second model to obtain target product information; and sending the product information to the user via a preset path. This invention solves the problem of inaccurate product recommendations, thereby improving the accuracy of product recommendations.

Description

Technical Field

[0001] This invention relates to the field of communications, and more specifically, to a method and apparatus for predicting and recommending products based on a dual-model approach for user sentiment. Background Technology

[0002] With the deepening development of commerce based on the Internet and big data, enterprises need to build user interest tags based on user behavior and interest characteristics in order to create more suitable push and consumption solutions for users.

[0003] Currently, interest tags are typically constructed by retrospectively analyzing users' consumption and browsing records. However, these interest tags not only fail to directly reflect users' interest in products (i.e., users may buy a product not because they like it, but because of group buying, matching, etc.), but also fail to reflect users' current emotional characteristics because the construction of interest tags is based on retrospective analysis of post-event behavior. Ultimately, this results in inaccurate and untimely standards for interest tags. Summary of the Invention

[0004] This invention provides a method and apparatus for predicting product recommendations based on a dual-model user sentiment, in order to at least solve the problem of inaccurate product recommendations in related technologies.

[0005] According to an embodiment of the present invention, a user sentiment prediction and product recommendation method based on a dual-model approach is provided, applied to a shared electric vehicle equipped with an in-vehicle camera, comprising:

[0006] Acquire first image information of the user, wherein the first image information is used to indicate the product images browsed by the user within a first time period;

[0007] Using a preset first model, the first image information is processed to determine the emotion information, thereby determining the user's emotion tag and emotion level.

[0008] Using a pre-defined second model, the emotion tags and emotion levels are matched with products to obtain target product information.

[0009] The product information is sent to the user via a preset path.

[0010] In an exemplary embodiment, the step of performing emotion information determination processing on the first image information using a preset first model to determine the user's emotion tag and emotion level includes:

[0011] Based on the first image information, the attribute information of the goods in the first image information is determined, wherein the attribute information includes the color information, type information and browsing frequency information of the first goods, and the first goods are used to indicate the first type of goods included in the goods;

[0012] Based on the color information, type information, and browsing frequency information, the user's emotion tag and emotion level are determined using the first model.

[0013] In an exemplary embodiment, the step of performing emotion information determination processing on the first image information using a preset first model to determine the user's emotion tag and emotion level includes:

[0014] Using the first model, the user's product browsing information included in the first image information is determined, wherein the product browsing information includes the product type and quantity per unit time.

[0015] If the quantity of goods meets the first condition, determine the browsing duration and browsing frequency of the target goods;

[0016] Based on the product type, browsing duration, and browsing frequency, the user's emotion tag and emotion level are determined.

[0017] In one exemplary embodiment, determining the user's emotion tag based on the product type, browsing duration, and browsing frequency includes:

[0018] When the browsing duration is less than a first threshold and the browsing frequency is greater than a second threshold, the first product type in the first time period and the second product type in the second time period are compared, wherein the unit time includes the first time period and the second time period, and the first time period and the second time period are continuous.

[0019] If the first product type and the second product type are inconsistent, the user's emotion tag is determined to be the first tag.

[0020] In an exemplary embodiment, the step of performing product matching processing on the emotion tags and emotion levels using a preset second model to obtain target product information includes:

[0021] Obtain the user's historical information, wherein the historical information includes the user's historical account level;

[0022] Based on the historical account level, the emotion tags and emotion levels are matched with products using a preset second model to obtain the target product information.

[0023] In an exemplary embodiment, the step of performing product matching processing on the emotion tags and emotion levels using a preset second model to obtain target product information further includes:

[0024] Obtain the user's historical information, wherein the historical information includes the user's historical spending amount information;

[0025] Based on the historical consumption amount information, the emotion tags and emotion levels are matched with products using a preset second model to obtain target product information.

[0026] In an exemplary embodiment, the step of performing product matching processing on the emotion tags and emotion levels using a preset second model to obtain target product information further includes:

[0027] Obtain the user's historical information, wherein the historical information includes the user's historical emotion tags and historical emotion levels;

[0028] Based on the historical emotion tags and the historical emotion levels, a preset second model is used to perform product matching processing on the emotion tags and emotion levels to obtain target product information.

[0029] According to another embodiment of the present invention, a user sentiment prediction and product recommendation device based on a dual-model is provided, applied to a shared electric vehicle equipped with an in-vehicle camera, comprising:

[0030] An information acquisition module is used to acquire the user's first image information, wherein the first image information is used to indicate the product images that the user browses within a first time period;

[0031] The emotion information processing module is used to perform emotion information determination processing on the first image information through a preset first model, so as to determine the user's emotion tag and emotion level;

[0032] The product matching module is used to perform product matching processing on the emotion tags and emotion levels through a preset second model to obtain target product information;

[0033] The information sending module is used to send the product information to the user through a preset path.

[0034] According to yet another embodiment of the present invention, a computer-readable storage medium is also provided, wherein a computer program is stored therein, wherein the computer program is configured to perform the steps in any of the above method embodiments when executed.

[0035] According to yet another embodiment of the present invention, an electronic device is also provided, including a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to perform the steps in any of the above method embodiments.

[0036] By labeling and ranking user emotions and further refining the matching of emotional information through a dual-model approach, the accuracy of product recommendations can be improved by further refining the identification of user emotions. Therefore, the problem of low accuracy in product recommendations can be solved, and the effect of improving the accuracy of product recommendations can be achieved. Attached Figure Description

[0037] Figure 1 This is a hardware structure block diagram of a mobile terminal for a user sentiment prediction and product recommendation method based on a dual-model according to an embodiment of the present invention.

[0038] Figure 2 This is a flowchart of a user sentiment prediction and product recommendation method based on a dual model according to an embodiment of the present invention;

[0039] Figure 3 This is a structural block diagram of a user sentiment prediction and product recommendation device based on a dual model according to an embodiment of the present invention. Detailed Implementation

[0040] The embodiments of the present invention will be described in detail below with reference to the accompanying drawings and examples.

[0041] It should be noted that the terms "first," "second," etc., in the specification, claims, and drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0042] The methods and embodiments provided in this application can be executed on a mobile terminal, computer terminal, or similar computing device. Taking running on a mobile terminal as an example, Figure 1 This is a hardware structure block diagram of a mobile terminal for a user sentiment prediction and product recommendation method based on a dual-model according to an embodiment of the present invention. Figure 1 As shown, a mobile terminal may include one or more ( Figure 1 Only one is shown in the diagram. A processor 102 (which may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.) and a memory 104 for storing data are also shown. The mobile terminal may further include a transmission device 106 for communication functions and an input / output device 108. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the mobile terminal described above. For example, the mobile terminal may also include components that are more... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0043] The memory 104 can be used to store computer programs, such as application software programs and modules, like the computer program corresponding to a dual-model-based user sentiment prediction and product recommendation method in this embodiment of the invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, thereby implementing the aforementioned method. The memory 104 may include high-speed random access memory and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the mobile terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0044] The transmission device 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the mobile terminal's communication provider. In one example, the transmission device 106 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 106 may be a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0045] This embodiment provides a user sentiment prediction and product recommendation method based on a dual-model approach, applied to shared electric vehicles equipped with in-vehicle cameras. Figure 2 This is a flowchart of a user sentiment prediction and product recommendation method based on a dual-model according to an embodiment of the present invention, such as... Figure 2 As shown, the process includes the following steps:

[0046] Step S202: Obtain the user's first image information, wherein the first image information is used to indicate the product images that the user browsed within a first time period;

[0047] In this embodiment, the product image is determined to facilitate product classification, accurately obtain product attributes and characteristics, and thus facilitate subsequent prediction and recognition of user emotions based on product attributes and characteristics.

[0048] The first image information includes (but is not limited to) information such as the product's link code. The first time period includes (but is not limited to) 1 minute before and 1 minute after a certain time point, or 30 seconds before and after a certain time point, etc. The specific time period can be adjusted and determined according to the usage environment and needs.

[0049] It should be noted that the determination of the first image information can be achieved through neural networks, data capture via web crawlers, or other methods. When applied to shared electric vehicles, it allows users to conveniently ride shared electric vehicles directly to the corresponding shopping malls to purchase goods, greatly satisfying users' shopping needs and improving the efficiency of goods purchase.

[0050] Step S204: Using a preset first model, perform emotion information determination processing on the first image information to determine the user's emotion tag and emotion level;

[0051] In this embodiment, determining the emotion level and emotion tag is to further categorize the user's emotions, thereby more accurately classifying the relevant emotions.

[0052] The emotion labels include joy, anger, sadness, or positive and negative emotions, and the emotion levels include different levels such as great joy, moderate joy, slight joy, great anger, moderate anger, and slight anger. The emotion recognition model can be (but is not limited to) a CNN model, or other models that can recognize emotions.

[0053] Step S206: Using a preset second model, perform product matching processing on the emotion tags and emotion levels to obtain target product information;

[0054] In this embodiment, since users need different types of items under different emotions, and under the same emotion tag, different degrees of emotion will also affect the user's product selection. For example, when they are overjoyed (celebrating), they may choose beer, while when they are slightly happy, they may choose red wine, etc. Or, if it is determined that they are overjoyed, they may push cola, betel nuts, etc. to the user, and so on; if they are slightly sad, they may recommend sweets, cakes, and other snacks.

[0055] The second model can be a Transformer matching model or other types of models, which will not be elaborated here.

[0056] Step S208: Send the product information to the user via a preset path.

[0057] In this embodiment, sending product information along a preset path is to enable the product information to be transmitted quickly, reduce the time for gateway identification and judgment, and improve information transmission efficiency.

[0058] Through the above steps, by labeling and ranking user emotions, and further refining the matching of emotional information through a dual-model approach, user emotions are identified in a more detailed manner, thereby improving the accuracy of product recommendations and solving the problem of low accuracy in product recommendations.

[0059] The entities that perform the above steps can be base stations, terminals, etc., but are not limited to these.

[0060] In an optional embodiment, the step of performing emotion information determination processing on the first image information using a preset first model to determine the user's emotion tag and emotion level includes:

[0061] Step S2042: Based on the first image information, determine the attribute information of the goods in the first image information, wherein the attribute information includes the color information, type information and browsing frequency information of the goods, and the first goods are used to indicate the first type of goods included in the goods;

[0062] Step S2044: Based on the color information, type information, and browsing frequency information, the user's emotion tag and emotion level are determined using the first model.

[0063] In this embodiment, user emotions are the manifestation of user psychological reactions, which can be represented by the color of product images, product type, and browsing frequency. Therefore, user emotions can be determined by determining color information, type information, and frequency information.

[0064] For example, if the products that users browse most frequently are in warm colors, it can be determined that the user's mood is positive; if the products that users browse most frequently are in cool colors, it indicates that the user's mood is low, and so on.

[0065] The color information includes (but is not limited to) the color of the product image, such as red, black, blue, green, etc., and the product type information includes (but is not limited to) the product's browsing frequency information, including the number of products viewed within a time period, the number of product types, and the number of times the same type of product was viewed.

[0066] In an optional embodiment, the step of performing emotion information determination processing on the first image information using a preset first model to determine the user's emotion tag and emotion level includes:

[0067] Step S20464: Using the first model, determine the user's product browsing information included in the first image information, wherein the product browsing information includes the product type and product quantity per unit time.

[0068] Step S2048: If the quantity of goods meets the first condition, determine the browsing duration and browsing frequency of the target goods;

[0069] Step S20410: Determine the user's emotion tag and emotion level based on the product type, browsing duration, and browsing frequency.

[0070] In this embodiment, to further improve the accuracy of product recommendations, in addition to determining the user's emotional level, the user's emotional tags and / or emotional levels can be further determined by analyzing the user's recent browsing history. For example, if the user browses the same type of product for a long time and spends a short time on each product, it can be judged that the user is in a more anxious mood; while if the user browses the same type of product for a long time and spends a long time on each product, it can be judged that the user is in a more calm mood.

[0071] In an optional embodiment, determining the user's emotion tag based on the product type, browsing duration, and browsing frequency includes:

[0072] Step S204102: When the browsing duration is less than a first threshold and the browsing frequency is greater than a second threshold, the first product type in the first time period and the second product type in the second time period are compared. The unit time includes the first time period and the second time period, and the first time period and the second time period are continuous.

[0073] Step S204104: If the first product type and the second product type are inconsistent, determine the user's emotion tag as the first tag.

[0074] In this embodiment, the browsing duration can be 5 seconds, 3 seconds, or 1 minute or 2 minutes, and can be adjusted according to actual needs; the first time period and the second time period can be two consecutive time periods before sending product information to the user, or two consecutive time periods from the most recent browsing record; the first tag can be an unstable emotion, or other types of emotion tags.

[0075] For example, if a user browses clothing and food for two consecutive half-hour periods before and after sending product information, and browses more than ten items in each category, with each item's browsing time being less than 3 seconds, it can be determined that the user's emotions are unstable, and emotional instability can be used as the primary label.

[0076] In an optional embodiment, determining the user's emotion tag based on the product type, browsing duration, and browsing frequency includes:

[0077] Step S204106: When the browsing duration is greater than a first threshold and the browsing frequency is less than a second threshold, the first product type in the first time period and the second product type in the second time period are compared. The unit time includes the first time period and the second time period, and the first time period and the second time period are continuous.

[0078] Step S204108: If the first product type and the second product type are the same, determine the user's emotion tag as the second tag.

[0079] In this embodiment, for example, if the user browsed clothing for two consecutive half-hour periods before and after sending product information to the user within one hour, and browsed more than ten products, and the browsing time for each product was greater than 3 seconds, it can be determined that the user's emotions are stable and the intention to buy is strong. At this time, the user's emotions are determined to be stable, and emotional stability is used as the second label.

[0080] In an optional embodiment, the step of performing product matching processing on the emotion tags and emotion levels using a preset second model to obtain target product information includes:

[0081] Step S2062: Obtain the user's historical information, wherein the historical information includes the user's historical account level;

[0082] Step S2064: Based on the historical account level, perform product matching processing on the emotion tag and emotion level through a preset second model to obtain the target product information.

[0083] In this embodiment, historical account levels can be used to quickly determine users' consumption habits and emotional change patterns, thereby providing users with more accurate product information.

[0084] The historical account rating includes (but is not limited to) information that can be used to determine the user's emotional rating and emotional rating, such as historical emotional tags, historical emotional rating, and historical product browsing frequency; the product matching process includes (but is not limited to) matching products that have been viewed in the past based on emotional tags and / or emotional rating, and may also include other types of matching.

[0085] In an optional embodiment, the step of performing product matching processing on the emotion tags and emotion levels using a preset second model to obtain target product information further includes:

[0086] Step S2066: Obtain the user's historical information, wherein the historical information includes the user's historical consumption amount information;

[0087] Step S2068: Based on the historical consumption amount information, the emotion tags and emotion levels are matched with products using a preset second model to obtain target product information.

[0088] In this embodiment, obtaining historical consumption amount information is to further determine the user's consumption habits, thereby narrowing down the range of products to be matched and improving product matching efficiency.

[0089] In an optional embodiment, the step of performing product matching processing on the emotion tags and emotion levels using a preset second model to obtain target product information further includes:

[0090] Step S20610: Obtain the user's historical information, wherein the historical information includes the user's historical emotion tags and historical emotion levels;

[0091] Step S20612: Based on the historical emotion tags and the historical emotion levels, the emotion tags and emotion levels are matched with products using a preset second model to obtain target product information.

[0092] In this embodiment, obtaining historical emotion levels and historical emotion tags is to determine the pattern of user emotion changes, thereby providing a reference and adjustment basis for the timing of product push notifications.

[0093] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0094] This embodiment also provides a user sentiment prediction and product recommendation device based on a dual-model approach. This device is used to implement the above embodiments and preferred embodiments, and details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0095] Figure 3 This is a structural block diagram of a user sentiment prediction and product recommendation device based on a dual-model according to an embodiment of the present invention, such as... Figure 3 As shown, the device is applied to shared electric vehicles equipped with in-vehicle cameras, and includes:

[0096] Information acquisition module 32 is used to acquire the user's first image information, wherein the first image information is used to indicate the product images browsed by the user in a first time period;

[0097] The emotion information processing module 34 is used to perform emotion information determination processing on the first image information through a preset first model, so as to determine the user's emotion tag and emotion level.

[0098] The product matching module 36 is used to perform product matching processing on the emotion tags and emotion levels through a preset second model to obtain target product information;

[0099] The information sending module 38 is used to send the product information to the user through a preset path.

[0100] In an optional embodiment, the emotion information processing module 34 includes:

[0101] The attribute information unit 342 is used to determine the attribute information of the product in the first image information based on the first image information, wherein the attribute information includes the color information, type information and browsing frequency information of the product, and the first product is used to indicate the first type of product included in the product;

[0102] Information determination unit 344 is used to determine the user's emotion tag and emotion level based on the color information, type information and browsing frequency information, through the first model.

[0103] In an optional embodiment, the apparatus further includes:

[0104] The browsing information collection module 310 is used to obtain the user's product browsing information after performing product matching processing on the user based on the emotion tag information and the emotion level information to obtain the matching result, wherein the product browsing information includes the product type and product quantity per unit time.

[0105] The information determination module 312 is used to determine the browsing duration and browsing frequency of the target product when the quantity of the product meets the first condition.

[0106] The tag determination module 314 is used to determine the user's emotion tag based on the product type, browsing duration, and browsing frequency.

[0107] The second information sending module 316 is used to send product information of the target product type to the user based on the emotion tag and the product type.

[0108] In an optional embodiment, the label determination module 314 includes:

[0109] The first comparison unit 3142 is used to compare the first product type in the first time period with the second product type in the second time period when the browsing duration is less than the first threshold and the browsing frequency is greater than the second threshold. The unit time includes the first time period and the second time period, and the first time period and the second time period are continuous.

[0110] The first tag unit 3144 is used to determine the user's emotion tag as the first tag when the first product type and the second product type are inconsistent.

[0111] In an optional embodiment, the label determination module 314 further includes:

[0112] The second comparison unit 3146 is used to compare the first product type in the first time period with the second product type in the second time period when the browsing duration is greater than the first threshold and the browsing frequency is less than the second threshold. The unit time includes the first time period and the second time period, and the first time period and the second time period are continuous.

[0113] The second tag unit 3148 is used to determine the user's emotion tag as the second tag when the first product type and the second product type are the same.

[0114] In an optional embodiment, the product matching module 36 includes:

[0115] Historical information unit 362 is used to obtain the user's historical information, wherein the historical information includes the user's historical account level;

[0116] The first product matching unit 364 is used to perform product matching processing on the emotion tag and emotion level based on the historical account level and through a preset second model to obtain the target product information.

[0117] In an optional embodiment, the product matching module 36 further includes:

[0118] The historical amount determination unit 366 is used to obtain the user's historical information, wherein the historical information includes the user's historical consumption amount information;

[0119] The second product matching unit 368 is used to perform product matching processing on the emotion tags and emotion levels based on the historical consumption amount information and through a preset second model to obtain target product information.

[0120] In an optional embodiment, the product matching module 36 further includes:

[0121] The historical emotion determination unit 3610 is used to obtain the user's historical information, wherein the historical information includes the user's historical emotion tags and historical emotion levels;

[0122] The third product matching unit 3612 is used to perform product matching processing on the historical emotion tags and the historical emotion levels through a preset second model to obtain target product information.

[0123] It should be noted that the above modules can be implemented by software or hardware. For the latter, they can be implemented in the following ways, but are not limited to: all the above modules are located in the same processor; or, the above modules are located in different processors in any combination.

[0124] Embodiments of the present invention also provide a computer-readable storage medium storing a computer program, wherein the computer program is configured to perform the steps in any of the above method embodiments when executed.

[0125] In one exemplary embodiment, the aforementioned computer-readable storage medium may include, but is not limited to, various media capable of storing computer programs, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard disk, magnetic disk, or optical disk.

[0126] Embodiments of the present invention also provide an electronic device including a memory and a processor, the memory storing a computer program and the processor being configured to run the computer program to perform the steps in any of the above method embodiments.

[0127] In one exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor and the input / output device is connected to the processor.

[0128] Specific examples in this embodiment can be found in the examples described in the above embodiments and exemplary implementations, and will not be repeated here.

[0129] It is obvious to those skilled in the art that the modules or steps of the present invention described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. They can be implemented using computer-executable program code, and thus can be stored in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those described herein, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, the present invention is not limited to any particular combination of hardware and software.

[0130] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, or improvements made within the principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for predicting product recommendations based on a dual-model user sentiment, characterized in that, Applications include shared electric vehicles equipped with onboard cameras, including: Acquire first image information of the user, wherein the first image information is used to indicate the product images browsed by the user within a first time period; Using a preset first model, the first image information is processed to determine the emotion information, thereby determining the user's emotion tag and emotion level. Using a pre-defined second model, the emotion tags and emotion levels are matched with products to obtain target product information. The product information is sent to the user via a preset path; The step of performing emotion information determination processing on the first image information using a preset first model to determine the user's emotion tag and emotion level includes: determining the attribute information of the goods in the first image information based on the first image information, wherein the attribute information includes the color information, type information, and browsing frequency information of the first goods, the first goods being used to indicate a first type of goods included in the goods; and determining the user's emotion tag and emotion level based on the color information, type information, and browsing frequency information using the first model. Alternatively, using the first model, determine the user's product browsing information included in the first image information, wherein the product browsing information includes the product type and quantity per unit time; if the quantity of products meets a first condition, determine the browsing duration and browsing frequency of the target product; and determine the user's emotion tag and emotion level based on the product type, browsing duration, and browsing frequency. The step of determining the user's emotion tag based on the product type, browsing duration, and browsing frequency includes: when the browsing duration is less than a first threshold and the browsing frequency is greater than a second threshold, comparing a first product type within a first time period with a second product type within a second time period, wherein the unit time includes the first time period and the second time period, and the first time period and the second time period are consecutive; when the first product type and the second product type are inconsistent, determining the user's emotion tag as a first tag.

2. The method according to claim 1, characterized in that, The step of performing product matching processing on the emotion tags and emotion levels using a preset second model to obtain target product information includes: Obtain the user's historical information, wherein the historical information includes the user's historical account level; Based on the historical account level, the emotion tags and emotion levels are matched with products using a preset second model to obtain the target product information.

3. The method according to claim 2, characterized in that, The step of performing product matching processing on the emotion tags and emotion levels using a preset second model to obtain target product information also includes: Obtain the user's historical information, wherein the historical information includes the user's historical spending amount information; Based on the historical consumption amount information, the emotion tags and emotion levels are matched with products using a preset second model to obtain target product information.

4. The method according to claim 2, characterized in that, The step of performing product matching processing on the emotion tags and emotion levels using a preset second model to obtain target product information also includes: Obtain the user's historical information, wherein the historical information includes the user's historical emotion tags and historical emotion levels; Based on the historical emotion tags and the historical emotion levels, a preset second model is used to perform product matching processing on the emotion tags and emotion levels to obtain target product information.

5. A user sentiment prediction and product recommendation device based on a dual-model approach, characterized in that, Applications include shared electric vehicles equipped with onboard cameras, including: An information acquisition module is used to acquire the user's first image information, wherein the first image information is used to indicate the product images that the user browses within a first time period; The emotion information processing module is used to perform emotion information determination processing on the first image information through a preset first model, so as to determine the user's emotion tag and emotion level; The product matching module is used to perform product matching processing on the emotion tags and emotion levels through a preset second model to obtain target product information; The information sending module is used to send the product information to the user via a preset path; The step of performing emotion information determination processing on the first image information using a preset first model to determine the user's emotion tag and emotion level includes: determining the attribute information of the goods in the first image information based on the first image information, wherein the attribute information includes the color information, type information, and browsing frequency information of the first goods, the first goods being used to indicate a first type of goods included in the goods; and determining the user's emotion tag and emotion level based on the color information, type information, and browsing frequency information using the first model. Alternatively, using the first model, determine the user's product browsing information included in the first image information, wherein the product browsing information includes the product type and quantity per unit time; if the quantity of products meets a first condition, determine the browsing duration and browsing frequency of the target product; and determine the user's emotion tag and emotion level based on the product type, browsing duration, and browsing frequency. The step of determining the user's emotion tag based on the product type, browsing duration, and browsing frequency includes: when the browsing duration is less than a first threshold and the browsing frequency is greater than a second threshold, comparing a first product type within a first time period with a second product type within a second time period, wherein the unit time includes the first time period and the second time period, and the first time period and the second time period are consecutive; when the first product type and the second product type are inconsistent, determining the user's emotion tag as a first tag.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein the computer program is configured to perform the method described in any one of claims 1 to 4 when executed.

7. An electronic device comprising a memory and a processor, characterized in that, The memory stores a computer program, and the processor is configured to run the computer program to perform the method described in any one of claims 1 to 4.

Images