A Method, Device and Terminal Device for Linking Gas Station User Information with Local Services

Abstract

An embodiment of the present invention provides a method, device and terminal device for linking gas station user information with local services, belonging to the technical field of data processing. The method includes: obtaining first consumption information of historical users in a target gas station corresponding to a target area and second consumption information of the historical users in a target store corresponding to the target area, and obtaining initial evaluation information corresponding to the target store; performing false detection on the initial evaluation information to obtain target evaluation information corresponding to the target store; determining a target consumption model corresponding to historical users in the target area according to the first consumption information, the second consumption information and the target evaluation information; obtaining third consumption information of a target user corresponding to a target gas station and fourth consumption information corresponding to relevant stores in the target area; and determining a target consumption route corresponding to the target user according to the third consumption information and the fourth consumption information by using the target consumption model.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to a method, device and terminal device for linking gas station user information with local services. Background Art

[0002] Traditionally, the service data of gas stations and local merchants are usually stored and managed independently, lacking effective integration and sharing, resulting in serious data island problems. Such data islands not only hinder the comprehensive analysis of user consumption behaviors, but also prevent merchants from accurately grasping the consumption routes and preferences of target users in specific areas. Therefore, merchants cannot make preparations in advance to provide personalized service experiences, ultimately leading to a significant reduction in the satisfaction and consumption experience of consumer users. For example, after a user refuels at a gas station, they may need to go to a nearby restaurant or supermarket. However, due to the lack of a unified data platform, merchants cannot understand the user's needs in advance, thus missing the opportunity to provide customized services and causing inconvenience to users during the consumption process. Summary of the Invention

[0003] The main objective of the embodiments of the present invention is to provide a method, device and terminal device for linking gas station user information with local services, aiming to solve the problem in the related art that merchants cannot accurately grasp the consumption routes and preferences of target users in specific areas, so merchants cannot make preparations in advance to provide personalized service experiences, ultimately leading to a significant reduction in the satisfaction and consumption experience of consumer users.

[0004] In a first aspect, the embodiments of the present invention provide a method for linking gas station user information with local services, including:

[0005] Obtaining first consumption information of historical users in a target gas station corresponding to a target area and second consumption information of the historical users in a target store corresponding to the target area, and obtaining initial evaluation information corresponding to the target store;

[0006] Performing false detection on the initial evaluation information to obtain target evaluation information corresponding to the target store;

[0007] Determining a target consumption model corresponding to the historical users in the target area according to the first consumption information, the second consumption information and the target evaluation information;

[0008] Obtaining third consumption information of a target user in the target gas station and fourth consumption information of the target user in related stores corresponding to the target area;

[0009] Determining a target consumption route corresponding to the target user according to the third consumption information and the fourth consumption information by using the target consumption model.

[0010] Second aspect, an embodiment of the present invention provides a linkage device for gas station user information and local services, including:

[0011] A data acquisition module, configured to obtain first consumption information of historical users in a target gas station corresponding to a target area and second consumption information of the historical users in a target store corresponding to the target area, and obtain initial evaluation information corresponding to the target store;

[0012] A false detection module, configured to perform false detection on the initial evaluation information to obtain target evaluation information corresponding to the target store;

[0013] A model determination module, configured to determine a target consumption model corresponding to the historical users in the target area according to the first consumption information, the second consumption information, and the target evaluation information;

[0014] A data collection module, configured to obtain third consumption information of a target user corresponding to the target gas station and fourth consumption information of the target user corresponding to relevant stores in the target area;

[0015] A route determination module, configured to determine a target consumption route corresponding to the target user by using the target consumption model according to the third consumption information and the fourth consumption information.

[0016] Third aspect, an embodiment of the present invention further provides a terminal device, where the terminal device includes a processor, a memory, a computer program stored on the memory and executable by the processor, and a data bus for realizing connection communication between the processor and the memory. When the computer program is executed by the processor, the steps of any one of the gas station user information and local service linkage methods provided in the specification of the present invention are implemented.

[0017] An embodiment of the present invention provides a method, apparatus, and terminal device for linking gas station user information with local services. The method includes: obtaining first consumption information of historical users in a target gas station corresponding to a target area and second consumption information of historical users in a target store corresponding to the target area, and obtaining initial evaluation information corresponding to the target store; performing false detection on the initial evaluation information to obtain target evaluation information corresponding to the target store; determining a target consumption model corresponding to historical users in the target area according to the first consumption information, the second consumption information, and the target evaluation information; obtaining third consumption information of a target user corresponding to the target gas station and fourth consumption information corresponding to relevant stores in the target area; and determining a target consumption route corresponding to the target user according to the third consumption information and the fourth consumption information using the target consumption model. By obtaining the consumption information of historical users in the target gas station and the target store in the target area, as well as the evaluation information of the target store, this method realizes the integration of multi-source data. In this way, it is possible to more comprehensively understand the user's behavior patterns and preferences, laying a foundation for establishing an accurate consumption model. Performing false detection on the initial evaluation information ensures the authenticity and reliability of the target store evaluation. This helps to improve the accuracy of the consumption model and avoid being misled by false evaluations. Based on the consumption information of historical users and the evaluation information of the target store, a target consumption model corresponding to historical users in the target area is constructed. This model can capture the consumption habits and potential needs of users in different scenarios. Therefore, by obtaining the consumption information of the target user in the target gas station and relevant stores and using the previously established target consumption model, it is possible to determine a personalized target consumption route for the target user. This not only improves the user experience but also may promote the sales of merchants, achieving a win-win situation. Furthermore, it realizes the accurate prediction of user consumption behavior and route optimization, thereby improving the overall service quality and user experience. It also solves the problem in the related art that merchants cannot accurately grasp the consumption route and preferences of target users in a specific area, so merchants cannot make preparations in advance and provide a personalized service experience, ultimately resulting in a significant reduction in the satisfaction and consumption experience of consumer users. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0019] Figure 1 It is a schematic flowchart of a method for linking gas station user information with local services provided by an embodiment of the present invention;

[0020] Figure 2Schematic diagram of the module structure of a device for linking gas station user information with local services provided by an embodiment of the present invention;

[0021] Figure 3 Schematic block diagram of the structure of a terminal device provided by an embodiment of the present invention. Detailed implementation manners

[0022] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0023] The flowchart shown in the accompanying drawings is only an example illustration, and does not necessarily include all contents and operations / steps, nor does it necessarily need to be executed in the described order. For example, some operations / steps can be decomposed, combined, or partially merged, so the actual execution order may be changed according to the actual situation.

[0024] It should be understood that the terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in this specification of the present invention and the appended claims, unless otherwise clearly specified in the context, the singular forms "a", "an", and "the" are intended to include the plural forms.

[0025] An embodiment of the present invention provides a method, a device, and a terminal device for linking gas station user information with local services. Among them, the method for linking gas station user information with local services can be applied to a terminal device, and the terminal device can be an electronic device such as a tablet computer, a notebook computer, a desktop computer, a personal digital assistant, and a wearable device. The terminal device can be a server or a server cluster.

[0026] Next, some embodiments of the present invention will be described in detail in conjunction with the accompanying drawings. Without conflict, the following embodiments and the features in the embodiments can be combined with each other.

[0027] Please refer to Figure 1 , Figure 1 which is a schematic flowchart of a method for linking gas station user information with local services provided by an embodiment of the present invention.

[0028] As Figure 1 shown, the method for linking gas station user information with local services includes steps S101 to S105.

[0029] Step S101: Obtain the first consumption information of historical users at the target gas station and the second consumption information of the historical users at the target store in the target area, and obtain the initial evaluation information corresponding to the target store.

[0030] Exemplarily, obtain the first consumption information of historical users at the target gas station from the data management system of the gas station, including but not limited to information such as user ID, consumption time, consumption amount, consumption items (such as refueling, car washing, etc.). And obtain the second consumption information of historical users from the data management system of the target store, including but not limited to information such as user ID, consumption time, consumption amount, consumption items (such as purchased goods or services), and obtain the initial evaluation information of historical users from an online evaluation platform (such as Dianping) or the internal evaluation system of the target store, including but not limited to information such as user ID, evaluation time, evaluation content, score, etc. Among them, the target store and the target gas station are distributed in the target area.

[0031] Step S102: Perform false detection on the initial evaluation information to obtain the target evaluation information corresponding to the target store.

[0032] Exemplarily, determine the false evaluation criteria, such as the evaluation content being too exaggerated, containing a large amount of repetitive or templated language, a large number of similar evaluations appearing in a short period of time, etc. Then, perform data annotation on the historical evaluation information according to the false evaluation criteria to obtain the labeled training data, and thus use a machine learning model, such as a classification model (such as logistic regression, support vector machine, random forest, etc.) or a deep learning model to train the model with the labeled training data, so as to obtain the corresponding false classification model.

[0033] Exemplarily, perform false detection on the initial evaluation information according to the false classification model to obtain the target evaluation information corresponding to the target store.

[0034] In some embodiments, the false detection of the initial evaluation information to obtain the target evaluation information corresponding to the target store includes: obtaining the corresponding first evaluation text and evaluation image in the initial evaluation information, and obtaining the first evaluation user corresponding to the first evaluation text from the initial evaluation information; obtaining the corresponding second evaluation text when evaluating the first evaluation text, and the second evaluation user corresponding to the second evaluation text; establishing a user network connection map according to the first evaluation text, the first evaluation user, the second evaluation text and the second evaluation user, wherein the first evaluation text, the first evaluation user, the second evaluation text and the second evaluation user are respectively determined as the nodes corresponding to the user network connection map, and the connections between the first evaluation text and the first evaluation user, the second evaluation text and the second evaluation user, and the first evaluation text and the second evaluation text are determined as the map relationships corresponding to the user network connection map; obtaining the user social feature vector corresponding to the first evaluation text under the involved map relationships from the user network connection map by using the map feature extraction layer of the false recognition model; obtaining a first text vector by performing text representation on the first evaluation text by using the text feature extraction layer of the false recognition model; obtaining an image feature vector by performing image information extraction on the evaluation image by using the image feature extraction layer of the false recognition model; obtaining a description image text corresponding to the evaluation image by performing text description on the evaluation image by using the text generation layer of the false recognition model; obtaining a second text vector by performing text representation on the description image text by using the text feature extraction layer of the false recognition model; obtaining a plurality of initial fusion feature vectors by performing feature fusion on the first text vector, the second text vector, the image feature vector and the user social feature vector by using the feature fusion layer of the false recognition model; obtaining a target fusion feature vector by performing feature splicing on the user social feature vector, the first text vector, the image feature vector, the second text vector and the plurality of initial fusion feature vectors by using the feature splicing layer of the false recognition model; obtaining the classification category corresponding to the first evaluation text by performing data classification on the target fusion feature vector by using the information classification layer of the false recognition model; and performing false detection on the initial evaluation information according to the classification category to obtain the target evaluation information corresponding to the target store.

[0035] Exemplarily, a first evaluation text and an evaluation image are obtained from the initial evaluation information, and the first evaluation user corresponding to the first evaluation text is obtained, and the corresponding second evaluation text when evaluating the first evaluation text and the second evaluation user corresponding to the second evaluation text are obtained from the evaluation system.

[0036] Exemplarily, the first evaluation text, the first evaluating user, the second evaluation text, and the second evaluating user are respectively used as nodes of the user network connection graph. The following connections are established in the user network connection graph: the connection between the first evaluation text and the first evaluating user, the connection between the second evaluation text and the second evaluating user, and the connection between the first evaluation text and the second evaluation text.

[0037] Exemplarily, through the graph feature extraction layer of the false recognition model, user social feature vectors related to the existence of a graph relationship with the first evaluating user are extracted from the user network connection graph.

[0038] Exemplarily, the first evaluation text is processed by using the text feature extraction layer of the false recognition model, such as BERT, to obtain a first text vector. The evaluation image is processed by using the image feature extraction layer of the false recognition model to obtain an image feature vector, and the text generation layer of the false recognition model performs text description on the evaluation image to obtain the description image text corresponding to the evaluation image. Then, the text feature extraction layer is used to process the description image text to obtain a second text vector.

[0039] Exemplarily, the first text vector, the second text vector, the image feature vector, and the user social feature vector are fused by using the feature fusion layer of the false recognition model, such as an attention network, to generate multiple initial fusion feature vectors. For example, an initial fusion feature vector is obtained by the attention network based on the first text vector, the second text vector, and the image feature vector, and another initial fusion feature vector is obtained by the attention network based on the second text vector, the image feature vector, and the user social feature vector.

[0040] Exemplarily, by using the feature splicing layer of the false recognition model, the user social feature vector, the first text vector, the image feature vector, the second text vector, and multiple initial fusion feature vectors are vector-spliced to generate a target fusion feature vector. Then, the information classification layer of the false recognition model is used to classify the target fusion feature vector to determine the classification category corresponding to the first evaluation text (such as a true evaluation or a false evaluation). Finally, false detection is performed on the initial evaluation information according to the classification category, false evaluations are removed, and true evaluation information is retained to form the target evaluation information of the target store.

[0041] In some embodiments, the graph feature extraction layer using the false recognition model obtains the user social feature vector corresponding to the first evaluation text under the graph relationship involved in the user network connection graph, including: using the text processing network of the graph feature extraction layer to perform text representation on the first evaluation text to obtain the first node representation corresponding to the first evaluation text; using the text processing network of the graph feature extraction layer to perform text representation on the second evaluation text to obtain the second node representation corresponding to the second evaluation text; using the node initialization network of the graph feature extraction layer to perform vector mean processing on the first evaluation text corresponding to the first evaluation user and the third evaluation text when the first evaluation user evaluates other users' texts to obtain the third node representation corresponding to the first evaluation user; using the node initialization network of the graph feature extraction layer to perform vector mean processing on the second evaluation text corresponding to the second evaluation user and the fourth evaluation text published by the second evaluation user for other stores to obtain the fourth node representation corresponding to the second evaluation user; using the node relationship network of the graph feature extraction layer to perform adjacent node calculation according to the graph relationship and the first node representation in combination with the second node representation, the third node representation and the fourth node representation to obtain the node similarity between any adjacent nodes in the graph relationship; using the path determination network of the graph feature extraction layer to determine the node path corresponding to the first evaluation text according to the node similarity and a preset threshold; using the feature splicing network of the graph feature extraction layer to determine the user social feature vector corresponding to the first evaluation text according to the node path.

[0042] Exemplarily, the text processing network of the graph feature extraction layer is used to process the first evaluation text to generate the first node representation, and then the text processing network of the graph feature extraction layer is used to process the second evaluation text to generate the second node representation.

[0043] Exemplarily, the node initialization network of the graph feature extraction layer is used to perform vector mean processing on the first evaluation text corresponding to the first evaluation user and the text vector corresponding to the third evaluation text of the first evaluation user for other users, so as to generate the third node representation corresponding to the first evaluation user. Through the mean processing, the overall characteristics of the first evaluation user in the evaluation behavior are captured.

[0044] Exemplarily, the node initialization network of the graph feature extraction layer is used to perform vector mean processing on the second evaluation text corresponding to the second evaluation user and the text vector corresponding to the fourth evaluation text of the second evaluation user for other stores, so as to generate the fourth node representation corresponding to the second evaluation user.

[0045] Exemplarily, according to the graph relationships in the user network connection graph (such as the connections between the first evaluation text and the second evaluation text, and between the first evaluating user and the second evaluating user), the calculation of adjacent nodes is performed by combining the first node representation, the second node representation, the third node representation, and the fourth node representation, and the node similarity between any adjacent nodes in the graph relationship is obtained. Thus, through the similarity calculation, the relevance and interaction intensity between nodes are measured.

[0046] Exemplarily, when the node similarity is greater than or equal to a preset threshold, the node associated with the first evaluation text is determined as the node path corresponding to the first evaluation text. The node path represents the propagation path or association path of the first evaluation text in the user network graph. By determining the path, the propagation pattern and key path of the first evaluation text in the user network are captured.

[0047] Exemplarily, according to the determined node path, the user social feature vector related to the first evaluation text is extracted. The user social feature vector integrates the node representations and node relationship information on the node path. Thus, through the node representations and node relationship information on the path, a comprehensive user social feature vector is formed for subsequent fraud detection.

[0048] Step S103: Determine the target consumption model corresponding to the historical user in the target area according to the first consumption information, the second consumption information, and the target evaluation information.

[0049] Exemplarily, feature extraction is performed on the first consumption information and the second consumption information to obtain the consumption amount, consumption frequency, consumption time, and consumption location.

[0050] Exemplarily, a sentiment analysis tool is used to perform sentiment classification on the target evaluation text, such as positive, neutral, or negative, so as to combine the consumption features and evaluation features to form a comprehensive feature vector. For example, features such as consumption amount, consumption frequency, consumption time, consumption location, evaluation sentiment, evaluation time, and evaluation object can be combined to form a high-dimensional feature vector.

[0051] Exemplarily, a suitable machine learning model is selected, such as linear regression, decision tree, random forest, gradient boosting tree, neural network, etc. The selected model is trained using the high-dimensional feature vector, and the model parameters are adjusted to achieve the best performance, thereby obtaining the target consumption model. The target consumption model can be used to predict user consumption behavior, identify high-value users, optimize marketing strategies, etc.

[0052] In some embodiments, determining the target consumption model corresponding to the historical user in the target area according to the first consumption information, the second consumption information, and the target evaluation information includes: determining the consumption category corresponding to the historical user and the consumption frequency corresponding to the consumption category according to the first consumption information and the second consumption information; determining a first association matrix between the historical user and the consumption category according to the consumption category and the consumption frequency; determining a second association matrix between the historical user and the consumption category according to the consumption category, the consumption frequency, and the target evaluation information; determining the similarity corresponding to the historical users according to the first association matrix and the second association matrix; determining the consumption score corresponding to the historical user for the consumption category according to the similarity, and determining the target consumption model corresponding to the historical user according to the similarity and the consumption score.

[0053] Exemplarily, based on the first consumption information and the second consumption information, the consumption categories of historical users are extracted. For example, according to the product categories, service types, etc. in the consumption records. The consumption times of each historical user in each consumption category are counted. For example, user A consumed 10 times in the food and beverage category and 5 times in the shopping category, etc.

[0054] Exemplarily, the first association matrix represents the association relationship between the historical user and the consumption category, based on the consumption frequency. The rows of the first association matrix represent historical users, the columns represent consumption categories, and the matrix elements are the consumption frequencies of the user in that category. The second association matrix combines the consumption category, the consumption frequency, and the target evaluation information to construct a more comprehensive association matrix between the user and the consumption category, combines the target evaluation information (such as evaluation scores, positive / negative evaluations, etc.) with the consumption frequency to form a weighted association degree.

[0055] Exemplarily, the cosine similarity, Pearson correlation coefficient are used to combine the first association matrix and the second association matrix to obtain the similarity between any two historical users.

[0056] Exemplarily, based on the similarity, the scores of historical users for each consumption category are predicted. For example, if two users are similar in consumption habits and the score of one user for a certain category is known, the score of the other user for that category can be predicted. Thus, comprehensively considering the similarity and the consumption score, a target consumption model is constructed. This model can be used to predict the future consumption tendency and score of users for different consumption categories, thereby guiding personalized recommendations and marketing strategies.

[0057] In some embodiments, the historical users at least include a first user and a second user. Determining the corresponding similarity between the historical users according to the first association matrix and the second association matrix includes: obtaining, from the first association matrix, a first relevant consumption category corresponding to the first user and a first relevant consumption frequency corresponding to the first relevant consumption category; obtaining, from the first association matrix, a second relevant consumption category corresponding to the second user and a second relevant consumption frequency corresponding to the second relevant consumption category; performing an intersection process on the first relevant consumption category and the second relevant consumption category to obtain a target relevant consumption category corresponding to the first user and the second user; determining a first consumption average value corresponding to the first user according to the first relevant consumption frequency and determining a second consumption average value corresponding to the second user according to the second relevant consumption frequency; obtaining, from the first relevant consumption frequency, a third relevant consumption frequency corresponding to the first user under the target relevant consumption category and obtaining, from the second relevant consumption frequency, a fourth relevant consumption frequency corresponding to the second user under the target relevant consumption category; fusing the first consumption average value, the second consumption average value, the third relevant consumption frequency, and the fourth relevant consumption frequency to determine a first relevant value corresponding to the first user and the second user; obtaining, from the second association matrix, a first consumption preference corresponding to the first user and a first consumption access value corresponding to the first consumption preference; obtaining, from the second association matrix, a second consumption preference corresponding to the second user and a second consumption access value corresponding to the second consumption preference; performing an intersection process on the first consumption preference and the second consumption preference to obtain a target consumption preference corresponding to the first user and the second user; determining a first evaluation score corresponding to the target consumption preference of the first user according to the first effective evaluation corresponding to the first user, and determining a first attenuation factor corresponding to the target consumption preference of the first user according to the first evaluation score; determining a second evaluation score corresponding to the target consumption preference of the second user according to the second effective evaluation corresponding to the second user, and determining a second attenuation factor corresponding to the target consumption preference of the second user according to the second evaluation score; determining a first access average value corresponding to the first user according to the first consumption access value and determining a second access average value corresponding to the second user according to the second consumption access value; obtaining, from the first consumption access value, a third consumption access value corresponding to the first user under the target consumption preference and obtaining, from the second consumption access value, a fourth consumption access value corresponding to the second user under the target consumption preference; using the first attenuation factor and the second attenuation factor to fuse the first access average value, the second access average value, the third consumption access value, and the fourth consumption access value to determine a second relevant value corresponding to the first user and the second user;Determine the corresponding similarity between the first user and the second user by integrating the first correlation value and the second correlation value.

[0058] Exemplarily, extract the consumption frequencies of the first user in each consumption category from the first association matrix to form the first relevant consumption categories and the corresponding first relevant consumption frequencies. Extract the consumption frequencies of the second user in each consumption category from the first association matrix to form the second relevant consumption categories and the corresponding second relevant consumption frequencies, and then perform an intersection process on the first relevant consumption categories and the second relevant consumption categories to obtain the common target relevant consumption categories of the first user and the second user.

[0059] Exemplarily, calculate the first consumption average value of the first user in all relevant consumption categories according to the first relevant consumption frequencies. Calculate the second consumption average value of the second user in all relevant consumption categories according to the second relevant consumption frequencies.

[0060] Exemplarily, obtain the third relevant consumption frequency of the first user in the target relevant consumption category from the first relevant consumption frequencies and obtain the fourth relevant consumption frequency of the second user in the target relevant consumption category from the second relevant consumption frequencies.

[0061] Exemplarily, integrate the first consumption average value, the second consumption average value, the third relevant consumption frequency, and the fourth relevant consumption frequency according to the Pearson correlation coefficient to calculate the first correlation value corresponding to the first user and the second user.

[0062] Exemplarily, extract the consumption access values of the first user in each consumption preference from the second association matrix to form the first consumption preferences and the corresponding first consumption access values. Extract the consumption access values of the second user in each consumption preference from the second association matrix to form the second consumption preferences and the corresponding second consumption access values. Perform an intersection process on the first consumption preferences and the second consumption preferences to obtain the common target consumption preferences of the first user and the second user.

[0063] Exemplarily, obtain the first valid evaluations corresponding to the first user from the initial evaluation information, then perform score conversion on each first valid evaluation to obtain the first evaluation scores of the first user for the target consumption preferences, and then compare the first evaluation scores with the preset scores to obtain the first quantity corresponding to when the first evaluation scores are greater than the preset scores. Then calculate the ratio of the first quantity to the total number of users to obtain the first ratio. Then take the logarithm base 10 after adding 1 to the first ratio, and then take the reciprocal after adding 1 to the logarithm value to obtain the first decay factor.

[0064] Exemplarily, a second valid evaluation corresponding to the second user is obtained from the initial evaluation information, and then each second valid evaluation is subjected to score conversion to obtain a second evaluation score of the second user for the target consumption preference. Then, the second evaluation score is compared with a preset score to obtain a second quantity corresponding to the case where the second evaluation score is greater than the preset score. Then, the second quantity is divided by the total number of users to obtain a second ratio. Then, 1 is added to the second ratio and the logarithm base 10 is taken. Then, 1 is added to the logarithm value and the reciprocal is taken to obtain the second attenuation factor.

[0065] Exemplarily, according to the first consumption access value, the first access average value of the first user under all target consumption preferences is calculated. And according to the second consumption access value, the second access average value of the second user under all target consumption preferences is calculated.

[0066] Exemplarily, a third consumption access value of the first user under the target consumption preference is obtained from the first consumption access value. A fourth consumption access value of the second user under the target consumption preference is obtained from the second consumption access value.

[0067] Exemplarily, the first attenuation factor is used to fuse the first difference between the first access average value and the third consumption access value, and the second attenuation factor is used to fuse the second difference between the second access average value and the fourth consumption access value. Then, the second correlation value corresponding to the first user and the second user is calculated by using the first difference and the second difference in combination with the Pearson correlation coefficient.

[0068] Exemplarily, the first weight corresponding to the first correlation value and the second weight corresponding to the second correlation value are determined. Then, the first correlation value and the second correlation value are fused according to the first weight and the second weight to determine the similarity between the first user and the second user.

[0069] In some embodiments, determining the consumption score of the historical user corresponding to the consumption category according to the similarity includes: determining the relevant users corresponding to the historical user according to the similarity, and obtaining the first relevant information corresponding to the historical user and the second relevant information corresponding to the relevant users from the target evaluation information; using a score prediction model to perform score prediction on the first relevant information to obtain a first score value corresponding to the first relevant information and using the score prediction model to perform score prediction on the second relevant information to obtain a second score value corresponding to the second relevant information; calculating the mean value according to the first score value to obtain a first mean value corresponding to the historical user and calculating the mean value according to the second score value to obtain a second mean value corresponding to the relevant users; determining the consumption category, and obtaining a third score value corresponding to the consumption category by the relevant users from the second score values; determining the consumption score of the historical user corresponding to the consumption category according to the similarity, the third score value, the first mean value and the second mean value; wherein, the consumption score is obtained according to the following formula:

[0070]

[0071] Wherein, represents the consumption score of the u-th historical user corresponding to the i-th consumption category, represents the first mean value corresponding to the u-th historical user, and n represents the number of users corresponding to the relevant users, represents the similarity between the u-th historical user and the j-th relevant user, represents the third score value corresponding to the i-th consumption category by the j-th relevant user, represents the second mean value corresponding to the j-th relevant user.

[0072] Exemplarily, based on the previously calculated similarity, relevant users similar to the historical user are determined. Several users with the highest similarity can be selected as relevant users. Extract the first relevant information corresponding to the historical user from the target evaluation information, such as the consumption records, evaluations, etc. of the historical user. And extract the second relevant information corresponding to the relevant users from the target evaluation information, which also includes consumption records, evaluations, etc.

[0073] Exemplarily, use a score prediction model to perform score prediction on the first relevant information to obtain a first score value corresponding to the first relevant information. Use the score prediction model to perform score prediction on the second relevant information to obtain a second score value corresponding to the second relevant information.

[0074] Exemplarily, the first mean value corresponding to the historical user is obtained by calculating the mean value according to the first scoring value. And the second mean value corresponding to the relevant user is obtained by calculating the mean value according to the second scoring value.

[0075] Exemplarily, the consumption category to be evaluated is determined. The third scoring value corresponding to the specified consumption category of the relevant user is extracted from the second scoring value. Then, using the similarity, the third scoring value, the first mean value, and the second mean value, the consumption score corresponding to the specified consumption category of the historical user is calculated through the following formula:

[0076]

[0077] where, represents the consumption score corresponding to the i-th consumption category of the u-th historical user, represents the first mean value corresponding to the u-th historical user, n represents the number of users corresponding to the relevant users, represents the similarity between the u-th historical user and the j-th relevant user, represents the third scoring value corresponding to the i-th consumption category of the j-th relevant user, represents the second mean value corresponding to the j-th relevant user.

[0078] Step S104, obtain the third consumption information of the target user corresponding to the target gas station and the fourth consumption information of the relevant stores corresponding to the target area.

[0079] Exemplarily, the third consumption information of the target user at the target gas station is obtained from the data management system of the gas station, including but not limited to information such as user ID, consumption time, consumption amount, consumption items (such as refueling, car washing, etc.). And the fourth consumption information of the target user is obtained from the data management system of the relevant stores, including but not limited to information such as user ID, consumption time, consumption amount, consumption items (such as purchased goods or services).

[0080] Step S105, determine the target consumption route corresponding to the target user by using the target consumption model according to the third consumption information and the fourth consumption information.

[0081] Exemplarily, the similarity between the target user and the historical user is obtained from the third consumption information and the fourth consumption information. Then, the relevant users corresponding to the target user are screened out according to the similarity. Then, the scores of the relevant users for each consumption category are determined according to the target consumption model. Furthermore, the predicted consumption stores corresponding to the target user are obtained according to the scores. Then, the store location corresponding to the predicted consumption store and the current location of the target user are obtained according to the Beidou positioning. Furthermore, route planning is performed according to the current location and the store location by using the consumption score corresponding to the predicted consumption store, so as to obtain the target consumption route corresponding to the target user.

[0082] In some embodiments, determining the target consumption route corresponding to the target user by using the target consumption model according to the third consumption information and the fourth consumption information includes: obtaining the associated user corresponding to the target user from the historical users according to the third consumption information, the fourth consumption information, the first consumption information, and the second consumption information; determining the initial consumption category corresponding to the target user and the initial consumption score corresponding to the initial consumption category from the target consumption model according to the associated user; obtaining the current time, and adjusting the initial consumption score corresponding to the initial consumption category according to the current time to obtain the target consumption score corresponding to the initial consumption category; adjusting the initial consumption category according to the target consumption score to obtain the target consumption category; and determining the target consumption route corresponding to the target user according to the target consumption category.

[0083] Exemplarily, according to the third consumption information and the fourth consumption information, in combination with the first consumption information and the second consumption information, the associated user most similar to the target user is screened out from the historical users by calculating the similarity (such as cosine similarity, Pearson correlation coefficient, etc.).

[0084] Exemplarily, the initial consumption category corresponding to the target user is determined from the target consumption model according to the information of the associated user. And the consumption score of the associated user under the initial consumption category is obtained from the target consumption model as the initial consumption score corresponding to the initial consumption category for the target user.

[0085] Exemplarily, the current time such as hours and minutes is obtained, and then the initial consumption score corresponding to the initial consumption category is adjusted according to the current time to obtain the target consumption score corresponding to the target consumption category. For example, the initial consumption score can be adjusted according to the time period (such as weekdays, weekends, holidays, etc.) or a specific time period (such as morning rush hour, evening rush hour, etc.) to obtain the target consumption score.

[0086] Exemplarily, the initial consumption category is adjusted according to the adjusted target consumption score to obtain the final target consumption category. For example, if the score of a certain consumption category is significantly increased at the current time, it can be used as the target consumption category.

[0087] Exemplarily, according to the finally determined target consumption category, a corresponding consumption route is recommended for the target user. For example, if the target consumption category is dining, high-rated restaurants nearby can be recommended; if the target consumption category is shopping, popular shopping centers can be recommended.

[0088] In some embodiments, adjusting the initial consumption score corresponding to the initial consumption category according to the current time to obtain the target consumption score corresponding to the initial consumption category includes: obtaining a first time corresponding to the third consumption information and a second time corresponding to the fourth consumption information; calculating a consumption similarity value between the target user and the associated user at the first time and the second time; obtaining the current consumption category corresponding to the associated user at the current time from the initial consumption category; obtaining the relevant consumption score corresponding to the current consumption category from the initial consumption score, and adjusting the relevant consumption score according to the consumption similarity value to obtain the target consumption score corresponding to the relevant consumption score.

[0089] Exemplarily, the specific time point when the third consumption information occurs is determined as the first time, and the specific time point when the fourth consumption information occurs is determined as the second time. According to the consumption behaviors of the target user and the associated user at the first time and the second time, the consumption similarity value between the two is calculated. For example, if the target user and the associated user have carried out similar types of consumption within the same time period, the similarity value will be higher.

[0090] Exemplarily, from the initial consumption category, the current consumption category corresponding to the associated user at the current time is extracted, and then from the initial consumption score, the relevant consumption score corresponding to the current consumption category is obtained. Then, according to the calculated consumption similarity value, the relevant consumption score is adjusted to obtain the target consumption score. For example, if the consumption similarity value is high, the relevant consumption score can be appropriately increased; if the consumption similarity value is low, the relevant consumption score can be appropriately decreased. By calculating the consumption similarity value at a specific time, the consumption behavior of the user within a specific time period can be more accurately reflected, the accuracy of the recommendation can be improved, and then by dynamically adjusting the consumption score according to the consumption similarity value, the recommendation result can be made more in line with the actual needs and behavior habits of the user.

[0091] Please refer to Figure 2 , Figure 2A linkage device 200 for gas station user information and local services provided by an embodiment of the present application. The linkage device 200 for gas station user information and local services includes a data acquisition module 201, a false detection module 202, a model determination module 203, a data collection module 204, and a route determination module 205. Among them, the data acquisition module 201 is used to obtain the first consumption information of historical users in the target gas station corresponding to the target area and the second consumption information of the historical users in the target store corresponding to the target area, and obtain the initial evaluation information corresponding to the target store. The false detection module 202 is used to perform false detection on the initial evaluation information to obtain the target evaluation information corresponding to the target store. The model determination module 203 is used to determine the target consumption model corresponding to the historical users in the target area according to the first consumption information, the second consumption information, and the target evaluation information. The data collection module 204 is used to obtain the third consumption information of the target user in the target gas station and the fourth consumption information of the target user in the relevant stores corresponding to the target area. The route determination module 205 is used to determine the target consumption route corresponding to the target user according to the third consumption information and the fourth consumption information by using the target consumption model.

[0092] In some embodiments, the linkage device 200 for gas station user information and local services can be applied to a terminal device.

[0093] It should be noted that those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the above-described linkage device 200 for gas station user information and local services can refer to the corresponding process in the foregoing embodiment of the gas station user information and local services linkage method, and will not be repeated here.

[0094] Please refer to Figure 3 , Figure 3 A schematic block diagram of the structure of a terminal device provided by an embodiment of the present invention.

[0095] As Figure 3 shown, the terminal device 300 includes a processor 301 and a memory 302. The processor 301 and the memory 302 are connected through a bus 303, and this bus is, for example, an I2C (Inter-integrated Circuit) bus.

[0096] Specifically, the processor 301 is used to provide computing and control capabilities to support the operation of the entire terminal device. The processor 301 may be a central processing unit (CPU), and the processor 301 may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0097] Specifically, the memory 302 may be a Flash chip, read-only memory (ROM), magnetic disk, optical disc, USB flash drive, or mobile hard disk, etc.

[0098] Those skilled in the art can understand that Figure 3 the structure shown in is only a block diagram of some structures related to the solution of the embodiment of the present invention, and does not constitute a limitation on the terminal device to which the solution of the embodiment of the present invention is applied. A specific server may include more or fewer components than those shown in the figure, or combine some components, or have a different component layout.

[0099] Among them, the processor is used to run the computer program stored in the memory and implement any one of the gas station user information and local service linkage methods provided by the embodiments of the present invention when executing the computer program.

[0100] In one embodiment, the processor is used to run the computer program stored in the memory and implement the following steps when executing the computer program:

[0101] Obtain the first consumption information of historical users in the target gas station corresponding to the target area and the second consumption information of the historical users in the target store corresponding to the target area, and obtain the initial evaluation information corresponding to the target store;

[0102] Perform false detection on the initial evaluation information to obtain the target evaluation information corresponding to the target store;

[0103] Determine the target consumption model corresponding to the historical users in the target area according to the first consumption information, the second consumption information, and the target evaluation information;

[0104] Obtain the third consumption information corresponding to the target user at the target gas station and the fourth consumption information corresponding to relevant stores in the target area.

[0105] Use the target consumption model to determine the target consumption route corresponding to the target user according to the third consumption information and the fourth consumption information.

[0106] It should be noted that those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working process of the above-described terminal device can refer to the corresponding process in the embodiment of the gas station user information and local service linkage method described above, and will not be repeated here.

[0107] The embodiment of the present invention also provides a storage medium for computer-readable storage. The storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of any gas station user information and local service linkage method provided in the specification of the embodiment of the present invention.

[0108] Among them, the storage medium can be an internal storage unit of the terminal device described in the foregoing embodiment, such as the hard disk or memory of the terminal device. The storage medium can also be an external storage device of the terminal device, such as a plug-in hard disk equipped on the terminal device, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc.

[0109] Those of ordinary skill in the art can understand that all or some of the steps in the methods disclosed above, and the functional modules / units in systems and devices, can be implemented as software, firmware, hardware, and their appropriate combinations. In a hardware embodiment, the division of functional modules / units mentioned above does not necessarily correspond to the division of physical components; for example, one physical component can have multiple functions, or one function or step can be executed by several physical components in cooperation. Some or all physical components can be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include a computer storage medium (or non-transitory medium) and a communication medium (or transitory medium). As is well known to those of ordinary skill in the art, the term computer storage medium 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 includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disk (DVD) or other optical disk storage, magnetic cassettes, tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, it is well known to those of ordinary skill in the art that a communication medium typically contains computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transmission mechanism, and can include any information delivery medium.

[0110] It should be understood that the term "and / or" used in the specification and appended claims of the present invention refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations. It should be noted that in this article, the term "comprises," "comprising," or any other variation thereof is intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a series of elements includes not only those elements but also other elements not expressly listed, or elements that are inherent to such process, method, article, or system. Without further limitation, an element defined by the statement "comprising an..." does not exclude the existence of additional identical elements in the process, method, article, or system that comprises the element.

[0111] The serial numbers of the embodiments of the present invention above are only for description and do not represent the superiority or inferiority of the embodiments. The above is only the specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of various equivalent modifications or substitutions, and these modifications or substitutions should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A method for linking gas station user information with local services, characterized in that: The method comprises: Obtaining first consumption information corresponding to a historical user at a target gas station in a target area and second consumption information corresponding to the historical user at a target store in the target area, and obtaining initial evaluation information corresponding to the target store; Performing false detection on the initial evaluation information to obtain target evaluation information corresponding to the target store; Determine a target consumption model corresponding to the historical user in the target area according to the first consumption information, the second consumption information and the target evaluation information; Obtain third consumption information corresponding to the target user at the target gas station and fourth consumption information corresponding to relevant stores in the target area; Determine a target consumption route corresponding to the target user by using the target consumption model according to the third consumption information and the fourth consumption information; The step of determining the target consumption model corresponding to the historical user in the target area according to the first consumption information, the second consumption information and the target evaluation information includes: Determine the consumption category corresponding to the historical user and the consumption frequency corresponding to the consumption category according to the first consumption information and the second consumption information; Determine a first association matrix between the historical users and the consumption categories according to the consumption categories and the consumption frequencies; Determine a second association matrix between the historical users and the consumption categories according to the consumption categories, the consumption frequencies and the target evaluation information; Determine the corresponding similarities between the historical users according to the first association matrix and the second association matrix; The consumption score of the historical user corresponding to the consumption category is determined according to the similarity, and the target consumption model corresponding to the historical user is determined according to the similarity and the consumption score.

2. The method according to claim 1, characterized in that The performing false detection on the initial evaluation information to obtain target evaluation information corresponding to the target store includes: Obtaining a first evaluation text and an evaluation image corresponding to the initial evaluation information, and obtaining a first evaluation user corresponding to the first evaluation text from the initial evaluation information; Obtaining a second evaluation text corresponding to the evaluation of the first evaluation text, and a second evaluation user corresponding to the second evaluation text; Establish a user network connection graph according to the first evaluation text, the first evaluation user, the second evaluation text and the second evaluation user, wherein the first evaluation text, the first evaluation user, the second evaluation text and the second evaluation user are respectively determined as nodes corresponding to the user network connection graph, and the connections between the first evaluation text and the first evaluation user, the second evaluation text and the second evaluation user, and the first evaluation text and the second evaluation text are determined as graph relationships corresponding to the user network connection graph; Obtaining a user social feature vector corresponding to the first evaluation text under the involved graph relationship from the user network connection graph using the graph feature extraction layer of the false recognition model; Using the text feature extraction layer of the false recognition model to perform text representation on the first evaluation text to obtain a first text vector; Using the image feature extraction layer of the false recognition model to extract image information from the evaluation image to obtain an image feature vector; Using the text generation layer of the false recognition model to perform text description on the evaluation image to obtain a description image text corresponding to the evaluation image; Using the text feature extraction layer of the false recognition model to perform text representation on the description image text to obtain a second text vector; Using the feature fusion layer of the false recognition model, the first text vector, the second text vector, the image feature vector and the user social feature vector are subjected to feature fusion to obtain a plurality of initial fused feature vectors; Using the feature concatenation layer of the false recognition model, concatenating the user social feature vector, the first text vector, the image feature vector, the second text vector, and a plurality of the initial fused feature vectors to obtain a target fused feature vector; Using the information classification layer of the false recognition model to perform data classification on the target fusion feature vector to obtain the classification category corresponding to the first evaluation text; The initial evaluation information is subjected to false detection according to the classification category to obtain the target evaluation information corresponding to the target store.

3. The method according to claim 2, characterized in that The graph feature extraction layer using the false recognition model obtains the user social feature vector corresponding to the first evaluation text under the graph relationship involved from the user network connection graph, including: Using the text processing network of the graph feature extraction layer to perform text representation on the first evaluation text to obtain a first node representation corresponding to the first evaluation text; Using the text processing network of the graph feature extraction layer to perform text representation on the second evaluation text to obtain a second node representation corresponding to the second evaluation text; Using the node initialization network of the graph feature extraction layer, vector mean processing is performed on the first evaluation text corresponding to the first evaluation user and the third evaluation text when the first evaluation user evaluates the evaluation text of other users to obtain a third node representation corresponding to the first evaluation user; Using the node initialization network of the graph feature extraction layer, vector mean processing is performed on the second evaluation text corresponding to the second evaluation user and the fourth evaluation text published by the second evaluation user on other stores to obtain a fourth node representation corresponding to the second evaluation user; Utilizing the node relationship network of the graph feature extraction layer to perform adjacent node calculation according to the graph relationship and the first node representation in combination with the second node representation, the third node representation and the fourth node representation, to obtain node similarity between any adjacent nodes in the graph relationship; Determine the node path corresponding to the first evaluation text according to the node similarity and a preset threshold value using the path determination network of the graph feature extraction layer; The user social feature vector corresponding to the first evaluation text is determined according to the node path using the feature concatenation network of the graph feature extraction layer.

4. The method according to claim 1, characterized in that: The historical users include at least a first user and a second user, and determining the corresponding similarities between the historical users according to the first association matrix and the second association matrix includes: Obtaining a first relevant consumption category corresponding to the first user and a first relevant consumption frequency corresponding to the first relevant consumption category from the first association matrix; Obtaining from the first association matrix a second relevant consumption category corresponding to the second user and a second relevant consumption frequency corresponding to the second relevant consumption category; Performing intersection processing on the first related consumption category and the second related consumption category to obtain target related consumption categories corresponding to the first user and the second user; Determine a first consumption average value corresponding to the first user according to the first relevant consumption frequency and determine a second consumption average value corresponding to the second user according to the second relevant consumption frequency; Obtaining a third relevant consumption frequency corresponding to the first user under the target relevant consumption category from the first relevant consumption frequency and obtaining a fourth relevant consumption frequency corresponding to the second user under the target relevant consumption category from the second relevant consumption frequency; Determine first correlation values ​​corresponding to the first user and the second user by fusing the first consumption average value, the second consumption average value, the third relevant consumption frequency, and the fourth relevant consumption frequency; Obtaining a first consumption preference corresponding to the first user and a first consumption access value corresponding to the first consumption preference from the second association matrix; Obtaining, from the second association matrix, a second consumption preference corresponding to the second user and a second consumption access value corresponding to the second consumption preference; Performing intersection processing on the first consumption preference and the second consumption preference to obtain target consumption preferences corresponding to the first user and the second user; Determine a first evaluation score corresponding to the target consumption preference of the first user according to a first valid evaluation corresponding to the first user, and determine a first attenuation factor corresponding to the target consumption preference of the first user according to the first evaluation score; Determining a second evaluation score corresponding to the target consumption preference of the second user according to the second valid evaluation corresponding to the second user, and determining a second attenuation factor corresponding to the target consumption preference of the second user according to the second evaluation score; Determine a first access average value corresponding to the first user according to the first consumption access value and determine a second access average value corresponding to the second user according to the second consumption access value; Obtaining a third consumption access value corresponding to the first user under the target consumption preference from the first consumption access value and obtaining a fourth consumption access value corresponding to the second user under the target consumption preference from the second consumption access value; Determine second correlation values ​​corresponding to the first user and the second user by fusing the first access average value, the second access average value, the third consumption access value, and the fourth consumption access value using the first attenuation factor and the second attenuation factor; The first correlation value and the second correlation value are combined to determine the corresponding similarity between the first user and the second user.

5. The method according to claim 1, characterized in that Determining the consumption score corresponding to the consumption category by the historical user according to the similarity includes: Determine the related user corresponding to the historical user according to the similarity, and obtain the first related information corresponding to the historical user and the second related information corresponding to the related user from the target evaluation information; Using the scoring prediction model to score the first relevant information to obtain a first scoring value corresponding to the first relevant information, and using the scoring prediction model to score the second relevant information to obtain a second scoring value corresponding to the second relevant information; Performing mean calculation according to the first scoring value to obtain a first mean value corresponding to the historical user, and performing mean calculation according to the second scoring value to obtain a second mean value corresponding to the related user; Determine the consumption category, and obtain a third scoring value corresponding to the consumption category by the relevant user from the second scoring value; Determine the consumption score corresponding to the consumption category by the historical user according to the similarity, the third score value, the first mean value, and the second mean value; The consumption score is obtained according to the following formula: ； in, represents the consumption score corresponding to the i-th consumption category by the u-th historical user, represents the first mean value corresponding to the u-th historical user, n represents the number of users corresponding to the relevant user, represents the similarity between the u-th historical user and the j-th related user, represents the third rating value corresponding to the i-th consumption category by the j-th related user, represents the second mean corresponding to the j-th related user.

6. The method according to claim 1, characterized in that The determining the target consumption route corresponding to the target user by using the target consumption model according to the third consumption information and the fourth consumption information includes: Obtaining associated users corresponding to the target user from the historical users according to the third consumption information and the fourth consumption information combined with the first consumption information and the second consumption information; Determining, according to the associated user, from the target consumption model, an initial consumption category corresponding to the target user and an initial consumption score corresponding to the initial consumption category; Obtaining the current time, and adjusting the initial consumption score corresponding to the initial consumption category according to the current time to obtain a target consumption score corresponding to the initial consumption category; Adjusting the initial consumption category according to the target consumption score to obtain a target consumption category; The target consumption route corresponding to the target user is determined according to the target consumption category.

7. The method according to claim 6, characterized in that The adjusting the initial consumption score corresponding to the initial consumption category according to the current time to obtain the target consumption score corresponding to the initial consumption category includes: Obtaining a first time corresponding to the third consumption information and a second time corresponding to the fourth consumption information; Calculate the consumption similarity value between the target user and the associated user at the first time and the second time; Obtaining a current consumption category corresponding to the associated user at the current time from the initial consumption category; The relevant consumption score corresponding to the current consumption category is obtained from the initial consumption score, and the relevant consumption score is adjusted according to the consumption similarity value to obtain the target consumption score corresponding to the relevant consumption score.

8. A gas station user information and local service linkage device, characterized in that: include: A data acquisition module, used to obtain first consumption information corresponding to a historical user at a target gas station in a target area and second consumption information corresponding to the historical user at a target store in the target area, and to obtain initial evaluation information corresponding to the target store; A false detection module, used to perform false detection on the initial evaluation information to obtain target evaluation information corresponding to the target store; A model determination module, used to determine the target consumption model corresponding to the historical user in the target area according to the first consumption information, the second consumption information and the target evaluation information, wherein the determination of the target consumption model corresponding to the historical user in the target area according to the first consumption information, the second consumption information and the target evaluation information includes: determining the consumption category corresponding to the historical user and the consumption frequency corresponding to the consumption category according to the first consumption information and the second consumption information; determining a first association matrix between the historical user and the consumption category according to the consumption category and the consumption frequency; determining a second association matrix between the historical user and the consumption category according to the consumption category, the consumption frequency and the target evaluation information; determining the corresponding similarity between the historical users according to the first association matrix and the second association matrix; determining the consumption score of the historical user for the consumption category according to the similarity, and determining the target consumption model corresponding to the historical user according to the similarity and the consumption score; A data collection module, used to obtain third consumption information corresponding to the target user at the target gas station and fourth consumption information corresponding to the relevant stores in the target area; A route determination module is used to determine a target consumption route corresponding to the target user using the target consumption model according to the third consumption information and the fourth consumption information.

9. A terminal device, characterized in that: The terminal device includes a processor and a memory; The memory is used to store computer programs; The processor is used to execute the computer program and implement the gas station user information and local service linkage method according to any one of claims 1 to 7 when executing the computer program.

Images