Method and device for analyzing user consumption behavior based on oil product order data

Abstract

The present invention relates to data analysis technology, and proposes a method, device, computer equipment and storage medium for analyzing user consumption behavior based on oil product order data, wherein the method comprises: obtaining oil product order data of multiple users, the oil product order data including membership attributes and refueling attributes of the users; extracting index label data of each attribute from each oil product order data, and marking the corresponding user according to the index label data; analyzing the user consumption behavior of the corresponding user according to the index label data of each attribute, the user consumption behavior is characterized by refueling trend data, frequency trend data and price sensitive data; screening out users who need to be recalled from multiple users according to the user consumption behavior; selecting preset coupons and sending them to the users who need to be recalled, so as to send coupons in a targeted manner to achieve the purpose of attracting users with low stickiness to gas station enterprises, greatly narrowing the user range and realizing accurate delivery.

Description

Technical Field

[0001] The present invention relates to the field of data analysis technology, and in particular to a method, device, computer equipment and storage medium for analyzing user consumption behavior based on oil product order data. Background Art

[0002] With the development of information technology, many gas station companies have adopted customer management systems to manage customers. In order to provide better services and improve customer stickiness, it is usually necessary to analyze customers and then launch preferential strategies. However, the current customer analysis cannot mark specific customers, and the strategy adopted is only a single strategy. For example, only consistent coupons can be issued to all customers, but different customers have inconsistent usage habits, and it is impossible to recall them purposefully based on user characteristics, resulting in a waste of resources and a low conversion rate of the strategy. Summary of the invention

[0003] The main purpose of the present invention is to provide a method, device, computer equipment and storage medium for analyzing user consumption behavior based on oil product order data, aiming to solve the technical problem in the prior art that corresponding strategies cannot be purposefully adopted according to user characteristics.

[0004] Based on the above invention objectives, the present invention proposes a method for analyzing user consumption behavior based on oil product order data, comprising:

[0005] Obtaining oil product order data of multiple users, wherein the oil product order data includes membership attributes and refueling attributes of the users;

[0006] Extracting index label data of each attribute from each of the oil product order data, and marking the corresponding user according to the index label data;

[0007] Analyze the user consumption behavior corresponding to the user according to the indicator label data of each attribute, and the user consumption behavior is characterized by refueling trend data, frequency trend data and price sensitivity data;

[0008] Screening out users who need to be recalled from the multiple users according to the consumption behavior of the users;

[0009] Select a preset coupon and send it to the user who needs to be recalled.

[0010] Furthermore, the step of selecting users to be recalled from the plurality of users according to the user consumption behavior includes:

[0011] Determining whether the user is a low-dependence user based on the user's consumption behavior;

[0012] If so, the user is determined to be a user who needs to be recalled, otherwise the user is determined to be a user who does not need to be recalled.

[0013] Furthermore, after the step of extracting the index label data of each attribute from each of the oil product order data and marking the corresponding user according to the index label data, the following steps are included:

[0014] The user, the indicator tag data, and the association relationship between the user and the indicator tag data are stored in a designated storage database and a query database respectively.

[0015] Furthermore, after the step of storing the user, the indicator tag data, and the association relationship between the user and the indicator tag data in a designated storage database and a query database respectively, the following steps are included:

[0016] Receive an input parameter corresponding to the query rule, where the input parameter is used to characterize the query condition of the user required for the query, and different input parameters correspond to different indicator label data;

[0017] Users with indicator tag data corresponding to the input parameters are filtered out according to the input parameters, and the filtered users are displayed.

[0018] Furthermore, after the step of selecting a preset coupon and sending it to the user who needs to be recalled, the following steps are included:

[0019] Classifying each of the users according to the indicator label data to obtain a plurality of classified users with a common label, wherein different common labels correspond to different specific coupons;

[0020] The corresponding specific coupon is obtained according to the common label, and the specific coupon is sent to the user of the corresponding category.

[0021] The present invention also provides a device for analyzing user consumption behavior based on oil product order data, comprising:

[0022] An order acquisition unit, used to acquire oil product order data of multiple users, wherein the oil product order data includes membership attributes and refueling attributes of the users;

[0023] An index extraction unit, used to extract index label data of each attribute from each of the oil product order data, and mark the corresponding user according to the index label data;

[0024] A behavior analysis unit, configured to analyze the user consumption behavior corresponding to the user according to the indicator tag data of each attribute, wherein the user consumption behavior is characterized by refueling trend data, frequency trend data, and price-sensitive data;

[0025] A user screening unit, used for screening out users who need to be recalled from a plurality of users according to the consumption behavior of the users;

[0026] The discount sending unit is used to select a preset coupon and send it to the user who needs to be recalled.

[0027] Further, the user screening unit includes:

[0028] A user determination subunit, configured to determine whether the user is a low-dependence user based on the user's consumption behavior;

[0029] The user recall subunit is used to determine that the user is a low-dependence user, and then determine that the user is a user who needs to be recalled, otherwise determine that the user is a user who does not need to be recalled.

[0030] Further, including:

[0031] The data storage unit is used to store the user, the indicator label data, and the association relationship between the user and the indicator label data in a designated storage database and a query database respectively.

[0032] The present invention also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the method for analyzing user consumption behavior based on oil product order data are implemented.

[0033] The present invention also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the above-mentioned method for analyzing user consumption behavior based on oil product order data are implemented.

[0034] The beneficial effects of the present invention are: obtaining the user's indicator label data through the user's order data, and then analyzing the user's consumption behavior based on the indicator label, so as to send coupons in a targeted manner to achieve the purpose of attracting users with low stickiness to the gas station enterprise, greatly narrowing the user range and achieving precise delivery. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 A schematic diagram of the steps of a method for analyzing user consumption behavior based on oil product order data in one embodiment of the present invention;

[0036] Figure 2 It is a schematic block diagram of the structure of an apparatus for analyzing user consumption behavior based on oil product order data in one embodiment of the present invention;

[0037] Figure 3 is a schematic structural block diagram of an embodiment of a storage medium of the present application;

[0038] Figure 4It is a schematic structural block diagram of an embodiment of a computer device of the present application.

[0039] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0040] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0041] Reference Figure 1 The method for analyzing user consumption behavior based on oil product order data in this embodiment includes:

[0042] Step S1: Obtaining oil product order data of multiple users, wherein the oil product order data includes membership attributes and refueling attributes of the users;

[0043] Step S2: extracting the index label data of each attribute from each of the oil product order data, and marking the corresponding user according to the index label data;

[0044] Step S3: analyzing the user consumption behavior corresponding to the user according to the indicator label data of each attribute, wherein the user consumption behavior is characterized by refueling trend data, frequency trend data and price sensitivity data;

[0045] Step S4: selecting users to be recalled from the plurality of users according to the user's consumption behavior;

[0046] Step S5: Select a preset coupon and send it to the user who needs to be recalled;

[0047] The above-mentioned method for analyzing user consumption behavior based on oil product order data is implemented based on an order system. The system includes three service subsystems, namely, order service, label service and recall service system. Among them, the order service system is used to generate corresponding order data according to user operations. The order is the oil product order data for the user to refuel, specifically including user account, refueling amount, refueling price, gas station and other information, and then the order data is stored in HBase; the above-mentioned label service system includes timed collection of indicator label data, specifically extracting different indicator label data from the order data, adding it to the user label, and then storing it in the query database Solr; the above-mentioned recall service system includes sending recall coupons to specific users according to business needs, and counting the number of recall responses after issuing the recall coupons, and recording the number of recall responses in the user label, as well as other customized services.

[0048] As in step S1 above, the oil product order data of multiple users is obtained through the order system. The oil product order data includes the user's membership attributes, refueling consumption attributes and predicted attributes, wherein the membership attributes include the user's basic information, such as gender, address, contact information, and general information, such as vehicle type, first refueling time, maximum liters of refueling, number of recalls and number of valid coupons, and gas stations consumed, etc. The refueling consumption attributes include commonly used oil products, commonly used gas stations, number of refueling times / liters in the past XX days, etc. Predicted attributes can be used to recall or guide continued consumption based on user consumption habits, such as predicted user consumption behavior data.

[0049] As described in the above step S2, the indicator label data of each attribute is extracted from the above oil order data, such as attribute information such as female gender, Toyota vehicle type, maximum refueling capacity of 50 liters, and commonly used gas stations as gas stations in xx district. At this time, the indicator label data is "female, Toyota, maximum refueling capacity of 50 liters, gas station in xx district", and then the user is marked according to these indicator label data, and these label data are bound and associated with the user.

[0050] Preferably, the user, indicator tag data and the association between the user and indicator tag data are stored in a designated storage database and a query database, for example, distributed storage in a MySQL database and Solr, so that they can be retrieved from the storage database at any time later, or queried based on the query database.

[0051] As described in the above step S3, the user consumption behavior of the corresponding user is analyzed based on the indicator label data of each attribute. Specifically, the above user consumption behavior can be characterized by refueling trend data, frequency trend data and price-sensitive data. The above refueling trend data is data predicted based on the user's recent (such as within the last three months) refueling situation. For example, the user's membership attributes can be input into a preset prediction model for prediction to obtain a prediction result. The above prediction result includes the user's consumption behavior, such as refueling time, refueling amount, etc. The prediction model can be a hidden Markov model, based on the user's membership attributes and the user's recent refueling history data as training data, to obtain a prediction model. The above prediction model is an existing mature technology, and the training process and the prediction calculation process will not be repeated here. At this time, the refueling trend data can be directly used to characterize the above user consumption behavior.

[0052] In addition, a refueling trend curve can be obtained based on the refueling interval and refueling amount in the recent period, thereby obtaining a corresponding refueling trend index, that is, the above-mentioned refueling trend data. From the above-mentioned refueling trend curve, the predicted refueling data for a period of time in the future can be obtained, including future refueling time, refueling amount, etc. The above-mentioned frequency trend data is obtained from the user's recent refueling frequency. For example, a frequency curve is obtained based on the refueling time and number of refueling in historical data, thereby obtaining a frequency trend index. The frequency trend index, that is, the above-mentioned refueling trend data, is used to obtain the user's refueling frequency in the future period. The above-mentioned price sensitive data represents the impact of oil price changes on user behavior. Specifically, it can be obtained through changes in historical oil prices and changes in the corresponding user's refueling behavior. For example, the price sensitivity index is obtained by the change rate of oil prices and changes in the user's refueling amount in the same period. The user's consumption behavior is determined by the above-mentioned refueling trend index, frequency trend index and price sensitivity index.

[0053] As described in the above step S4, users who need to be recalled are screened out from multiple users based on their consumption behaviors. For example, users who have not refueled for a long time and whose refueling frequency is too low are screened out from the above refueling trend data, frequency trend data and price-sensitive data, such as users whose refueling trend index value is lower or higher than a preset range, and users whose frequency trend index value is lower than a preset range.

[0054] As described in step S5 above, preset coupons are then selected and sent to those users who need to be recalled, such as sending coupons of specified amounts, or coupons for discounts on purchases above a certain amount. Targeted strategies are adopted to send coupons to attract users with low stickiness to gas station companies, greatly narrowing the user range and achieving precise delivery.

[0055] In one embodiment, the above step S4 includes:

[0056] Determining whether the user is a low-dependence user based on the user's consumption behavior;

[0057] If so, the user is determined to be a user who needs to be recalled, otherwise the user is determined to be a user who does not need to be recalled.

[0058] In this embodiment, the user is judged as a low-dependence user or a high-dependence user based on the user's consumption behavior. The above-mentioned low-dependence user is a user who has not refueled at the gas station of the gas station enterprise for a long time or has a low refueling frequency, and the above-mentioned high-dependence user is a user who often refuels at the gas station of the gas station enterprise and has a relatively high refueling frequency. For example, the refueling trend data obtained through the prediction model can be used to determine whether the user is a low-dependence user based on the refueling trend data, or specifically, it can be determined based on the above-mentioned refueling trend data, frequency trend data and price sensitivity data to determine whether the user is a low-dependence user. For example, a user whose refueling trend index value, frequency trend index value and price sensitivity index are all lower than the preset range can be determined as a user who needs to be recalled, otherwise it can be determined as a user who does not need to be recalled.

[0059] In one embodiment, after the above step S2, the following steps are included:

[0060] Step S22: receiving an input parameter corresponding to the query rule, wherein the input parameter is used to characterize the query condition of the user required for the query, and different input parameters correspond to different indicator label data;

[0061] Step S23: filtering out users with indicator tag data corresponding to the input parameters according to the input parameters, and displaying the filtered users.

[0062] In this embodiment, the query rules are rules for determining query conditions based on the input parameters, and filtering out corresponding users based on the query conditions. The input parameters are used to characterize the query conditions of the users required for the query, and different input parameters correspond to different index label data. For example, for users whose query conditions are male and whose first refueling time is November 2021, the input parameters are male and whose first refueling time is November 2021. The corresponding index label data is male and whose first refueling time is 2021 / 11. The corresponding index label data is matched to the corresponding index label data through the input parameters, for example, "male" is filtered out from the gender label, and then "first refueling time is 2021 / 11" is filtered out from the "male" label to obtain the corresponding users, and then display these users.

[0063] Parameters are passed into the tag service system according to the input query rules, so that query conditions can be freely spliced ​​to screen out users who meet the conditions.

[0064] In one embodiment, after the above step S5, the following steps are included:

[0065] Step S6: Classify each of the users according to the indicator label data to obtain a plurality of classified users with common labels, where different common labels correspond to different specific coupons;

[0066] Step S7: Obtain corresponding specific coupons according to the shared tags, and send the specific coupons to the users of the corresponding categories.

[0067] In this embodiment, each user is classified according to the above-mentioned indicator label data, for example, they are divided into male and female categories according to gender, for example, they are classified according to car type, etc., so that users with the same attributes are users belonging to the same category and have common labels. In order to improve personalized services, different specific coupons can be designed for different common labels, such as gift coupons, shopping coupons, coupons, etc. Then, the corresponding specific coupons are obtained according to the common labels. For example, for users classified as female, the corresponding specific coupons are gift coupons, for users classified as male, the corresponding specific coupons are coupons, and for users of different car types, the corresponding specific coupons are gift coupons of the corresponding car types, and then these specific coupons are sent to the user accounts of the corresponding users. In this way, under different labels, gas stations can provide targeted services to different users, and the operability and feasibility have been greatly improved.

[0068] Reference Figure 2 In this embodiment, a device for analyzing user consumption behavior based on oil product order data is provided. The device corresponds to the above-mentioned method for analyzing user consumption behavior based on oil product order data. The device includes:

[0069] The order acquisition unit 1 is used to acquire oil product order data of multiple users, wherein the oil product order data includes membership attributes and refueling attributes of the users;

[0070] An index extraction unit 2 is used to extract index label data of each attribute from each of the oil product order data, and mark the corresponding user according to the index label data;

[0071] A behavior analysis unit 3 is used to analyze the user consumption behavior corresponding to the user according to the indicator label data of each attribute, and the user consumption behavior is characterized by refueling trend data, frequency trend data and price sensitivity data;

[0072] A user screening unit 4 is used to screen out users who need to be recalled from a plurality of users according to the user consumption behavior;

[0073] The discount sending unit 5 is used to select a preset coupon and send it to the user who needs to be recalled.

[0074] The above-mentioned device for analyzing user consumption behavior based on oil product order data is implemented based on an order system. The system includes three service subsystems, namely order service, label service and recall service system. Among them, the order service system is used to generate corresponding order data according to user operations. The order is the oil product order data for the user to refuel, specifically including user account, refueling amount, refueling price, gas station and other information, and then the order data is stored in HBase; the above-mentioned label service system includes timed collection of indicator label data, specifically extracting different indicator label data from the order data, adding it to the user label, and then storing it in the query database Solr; the above-mentioned recall service system includes sending recall coupons to specific users according to business needs, and counting the number of recall responses after issuing the recall coupons, and recording the number of recall responses in the user label, as well as other customized services.

[0075] As described above, the order acquisition unit 1 acquires the oil product order data of multiple users through the order system. The oil product order data includes the user's membership attributes, refueling consumption attributes, and predicted attributes, wherein the membership attributes include the user's basic information, such as gender, address, contact information, and general information, such as vehicle type, first refueling time, maximum number of liters of refueling, number of recalls, number of valid coupons, and gas stations consumed, etc. The refueling consumption attributes include commonly used oil products, commonly used gas stations, number of refueling times / liters in the past XX days, etc. The predicted attributes can be used to recall or guide continued consumption based on the user's consumption habits, such as predicted user consumption behavior data.

[0076] As described in the above-mentioned index extraction unit 2, the index label data of each attribute is extracted from the above-mentioned oil order data, such as attribute information such as female gender, Toyota vehicle type, maximum refueling capacity of 50 liters, and commonly used gas stations as xx district gas stations. At this time, the index label data is "female, Toyota, maximum refueling capacity of 50 liters, xx district gas station", and then the user is marked according to these indicator label data, and these label data are bound and associated with the user.

[0077] Preferably, it also includes a storage data unit, which is used to store users, indicator label data and the association between users and indicator label data in a designated storage database and a query database, for example, distributed storage in a MySQL database and Solr, so that it can be retrieved from the storage database at any time later, or queried based on the query database.

[0078] As described in the above-mentioned behavior analysis unit 3, the user consumption behavior of the corresponding user is analyzed based on the indicator label data of each attribute. Specifically, the above-mentioned user consumption behavior can be characterized by refueling trend data, frequency trend data and price-sensitive data. The above-mentioned refueling trend data is data predicted based on the user's recent (such as within the past three months) refueling situation. For example, the user's membership attributes can be input into a preset prediction model for prediction to obtain a prediction result. The above-mentioned prediction result includes the user's consumption behavior, such as refueling time, refueling amount, etc. The prediction model can be a hidden Markov model, based on the user's membership attributes and the user's recent refueling history data as training data, to obtain a prediction model. The above-mentioned prediction model is an existing mature technology, and the training process and the prediction calculation process will not be repeated here. At this time, the refueling trend data can be directly used to characterize the above-mentioned user consumption behavior.

[0079] In addition, a refueling trend curve can be obtained based on the refueling interval and refueling amount in the recent period, thereby obtaining a corresponding refueling trend index, that is, the above-mentioned refueling trend data. From the above-mentioned refueling trend curve, the predicted refueling data for a period of time in the future can be obtained, including future refueling time, refueling amount, etc. The above-mentioned frequency trend data is obtained from the user's recent refueling frequency. For example, a frequency curve is obtained based on the refueling time and number of refueling in historical data, thereby obtaining a frequency trend index. The frequency trend index, that is, the above-mentioned refueling trend data, is used to obtain the user's refueling frequency in the future period. The above-mentioned price sensitive data represents the impact of oil price changes on user behavior. Specifically, it can be obtained through changes in historical oil prices and changes in the corresponding user's refueling behavior. For example, the price sensitivity index is obtained by the change rate of oil prices and changes in the user's refueling amount in the same period. The user's consumption behavior is determined by the above-mentioned refueling trend index, frequency trend index and price sensitivity index.

[0080] As described in the above-mentioned user screening unit 4, users who need to be recalled are screened out from multiple users based on their user consumption behaviors. For example, users who have not refueled for a long time and whose refueling frequency is too low are screened out from the above-mentioned refueling trend data, frequency trend data and price-sensitive data, such as users whose refueling trend index value is lower or higher than a preset range, and users whose frequency trend index value is lower than a preset range.

[0081] As described in the above-mentioned discount sending unit 5, preset coupons are then selected and sent to those users who need to be recalled, such as sending coupons of specified amounts, or coupons for discounts on purchases above a certain amount. Corresponding strategies are taken in a targeted manner to send coupons to attract users with low stickiness to gas station companies, greatly narrowing the user range and achieving precise delivery.

[0082] In one embodiment, the above-mentioned user screening unit 4 includes:

[0083] A user determination subunit, configured to determine whether the user is a low-dependence user based on the user's consumption behavior;

[0084] The user recall subunit is used to determine that the user is a low-dependence user, and then determine that the user is a user who needs to be recalled, otherwise determine that the user is a user who does not need to be recalled.

[0085] In this embodiment, the user is judged as a low-dependence user or a high-dependence user based on the user's consumption behavior. The above-mentioned low-dependence user is a user who has not refueled at the gas station of the gas station enterprise for a long time or has a low refueling frequency, and the above-mentioned high-dependence user is a user who often refuels at the gas station of the gas station enterprise and has a relatively high refueling frequency. For example, the refueling trend data obtained through the prediction model can be used to determine whether the user is a low-dependence user based on the refueling trend data, or specifically, it can be determined based on the above-mentioned refueling trend data, frequency trend data and price sensitivity data to determine whether the user is a low-dependence user. For example, a user whose refueling trend index value, frequency trend index value and price sensitivity index are all lower than the preset range can be determined as a user who needs to be recalled, otherwise it can be determined as a user who does not need to be recalled.

[0086] In one embodiment, the above device comprises:

[0087] A parameter receiving unit, used to receive an input parameter corresponding to a query rule, wherein the input parameter is used to characterize the query condition of the user required for the query, and different input parameters correspond to different indicator label data;

[0088] The user display unit is used to filter out users with indicator tag data corresponding to the input parameters according to the input parameters, and display the filtered users.

[0089] In this embodiment, the query rules are rules for determining query conditions based on the input parameters, and filtering out corresponding users based on the query conditions. The input parameters are used to characterize the query conditions of the users required for the query, and different input parameters correspond to different index label data. For example, for users whose query conditions are male and whose first refueling time is November 2021, the input parameters are male and whose first refueling time is November 2021. The corresponding index label data is male and whose first refueling time is 2021 / 11. The corresponding index label data is matched to the corresponding index label data through the input parameters, for example, "male" is filtered out from the gender label, and then "first refueling time is 2021 / 11" is filtered out from the "male" label to obtain the corresponding users, and then display these users.

[0090] Parameters are passed into the tag service system according to the input query rules, so that query conditions can be freely spliced ​​to screen out users who meet the conditions.

[0091] In one embodiment, the above device comprises:

[0092] A classification label unit, used to classify each of the users according to the indicator label data, and obtain a plurality of classified users with a common label, where different common labels correspond to different specific coupons;

[0093] The specific acquisition unit is used to acquire the corresponding specific coupon according to the common label, and send the specific coupon to the user of the corresponding category.

[0094] In this embodiment, each user is classified according to the above-mentioned indicator label data, for example, they are divided into male and female categories according to gender, for example, they are classified according to car type, etc., so that users with the same attributes are users belonging to the same category and have common labels. In order to improve personalized services, different specific coupons can be designed for different common labels, such as gift coupons, shopping coupons, coupons, etc. Then, the corresponding specific coupons are obtained according to the common labels. For example, for users classified as female, the corresponding specific coupons are gift coupons, for users classified as male, the corresponding specific coupons are coupons, and for users of different car types, the corresponding specific coupons are gift coupons of the corresponding car types, and then these specific coupons are sent to the user accounts of the corresponding users. In this way, under different labels, gas stations can provide targeted services to different users, and the operability and feasibility have been greatly improved.

[0095] refer to Figure 3 The present application also provides a computer-readable storage medium 10, in which a computer program 20 is stored. When the computer program 20 is executed on a computer, the computer executes the method for analyzing user consumption behavior based on oil product order data described in the above embodiment.

[0096] refer to Figure 4 The present application also provides a computer device 40 including instructions, the computer device includes a memory 30 and a processor 50, the memory 30 stores a computer program 20, and when the processor 30 executes the computer program 20, the method for analyzing user consumption behavior based on oil product order data described in the above embodiment is implemented.

[0097] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of a computer program product.

[0098] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from a website site, a computer, a server, or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server, or data center. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a server or a data center that includes one or more available media integrations. The available medium may be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state hard disk Solid State Disk (SSD)), etc.

[0099] The above description is only a preferred embodiment of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A method for analyzing user consumption behavior based on oil product order data, characterized in that: include: Obtaining oil product order data of multiple users, wherein the oil product order data includes membership attributes and refueling attributes of the users; Extracting index label data of each attribute from each of the oil product order data, and marking the corresponding user according to the index label data; Analyze the user consumption behavior corresponding to the user according to the indicator label data of each attribute, and the user consumption behavior is characterized by refueling trend data, frequency trend data and price sensitivity data; Screening out users who need to be recalled from the multiple users according to the consumption behavior of the users; Select a preset coupon and send it to the user who needs to be recalled; The step of selecting users to be recalled from a plurality of users according to the user consumption behavior comprises: Determining whether the user is a low-dependence user based on the user's consumption behavior; If so, the user is determined to be a user who needs to be recalled, otherwise the user is determined to be a user who does not need to be recalled; After the step of extracting the index label data of each attribute from each of the oil product order data and marking the corresponding user according to the index label data, the method further includes: The user, the indicator tag data, and the association relationship between the user and the indicator tag data are stored in a designated storage database and a query database respectively; After the steps of storing the user, the indicator tag data, and the association relationship between the user and the indicator tag data in a designated storage database and a query database, the method further includes: Receive an input parameter corresponding to the query rule, where the input parameter is used to characterize the query condition of the user required for the query, and different input parameters correspond to different indicator label data; Filter out users with indicator tag data corresponding to the input parameters according to the input parameters, and display the filtered users; After the step of selecting a preset coupon and sending it to the user who needs to be recalled, the method includes: Classifying each of the users according to the indicator label data to obtain a plurality of classified users with a common label, wherein different common labels correspond to different specific coupons; The corresponding specific coupon is obtained according to the common label, and the specific coupon is sent to the user of the corresponding category.

2. A device for analyzing user consumption behavior based on oil product order data, used in the method for analyzing user consumption behavior based on oil product order data as claimed in claim 1, characterized in that: include: An order acquisition unit, used to acquire oil product order data of multiple users, wherein the oil product order data includes membership attributes and refueling attributes of the users; An index extraction unit, used to extract index label data of each attribute from each of the oil product order data, and mark the corresponding user according to the index label data; A behavior analysis unit, configured to analyze the user consumption behavior corresponding to the user according to the indicator tag data of each attribute, wherein the user consumption behavior is characterized by refueling trend data, frequency trend data, and price-sensitive data; A user screening unit, used for screening out users who need to be recalled from a plurality of users according to the consumption behavior of the users; The discount sending unit is used to select a preset coupon and send it to the user who needs to be recalled.

3. The device for analyzing user consumption behavior based on oil product order data according to claim 2 is characterized in that: The screening user unit comprises: A user determination subunit, configured to determine whether the user is a low-dependence user based on the user's consumption behavior; The user recall subunit is used to determine that the user is a low-dependence user, and then determine that the user is a user who needs to be recalled, otherwise determine that the user is a user who does not need to be recalled.

4. The device for analyzing user consumption behavior based on oil product order data according to claim 2 is characterized in that: include: The data storage unit is used to store the user, the indicator label data, and the association relationship between the user and the indicator label data in a designated storage database and a query database respectively.

5. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for analyzing user consumption behavior based on oil product order data described in claim 1 are implemented.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for analyzing user consumption behavior based on oil product order data described in claim 1 are implemented.

Images