Suggested decision-making method and device based on front-end application system user portrait model

Abstract

The invention belongs to the technical field of data processing, and particularly relates to a suggestion decision-making method and device based on a front-end application system user portrait model, and the method comprises the steps: collecting the user behavior data of each user according to a pre-constructed user access behavior monitoring index system; constructing a user portrait model based on the user behavior data; acquiring a user portrait report by using the user portrait model; based on the user portrait report, utilizing a pre-established analysis and prediction model to obtain a predicted suggested decision scheme; the analysis and prediction model is constructed by using a historical user portrait report and a historical suggestion decision scheme. According to the technical scheme provided by the invention, the user demand can be efficiently and quickly mined, and the investigation work of the user is greatly reduced; and core users can be quickly distinguished in a labor-saving manner, which are key marketing groups and which are key construction directions are determined, and the problem of long demand confirmation time caused by multiple internal levels of an enterprise is relieved.

Description

Technical Field

[0001] The present invention belongs to the field of data processing technology, and specifically relates to a suggestion decision method and device based on a user portrait model of a front-end application system. Background Art

[0002] The informatization construction of some enterprises has always been inefficient due to the nature of the enterprise. On the one hand, the design and development of the front-end system generally has the core problems of vague enterprise-level user needs, multiple levels of demand docking channels, and short time, making it difficult to distinguish between core functional needs and construction value. On the other hand, the front-end system evaluation lacks scientific basis, and the system construction and system evaluation links are separated, and a benign operation and expansion chain of system construction-system application-system evaluation feedback-system optimization construction has not been formed.

[0003] Therefore, it is necessary to carry out user portrait analysis to help users clarify their functional requirements. By building a general user portrait indicator model for the State Grid's information system, we can deeply explore the value and behavioral characteristics of various types of users, anchor the key units for the promotion and application of middle-office functions, and help the middle-office operation side of the project to achieve precise marketing, product evaluation and optimization, user experience improvement and scientific decision-making support.

[0004] However, there are differences between the current user portrait theory and the actual application of the information construction system of some enterprises, and there is a lack of targeted scientific user portrait analysis system research methods. The current problems with the information reduction system are: First, short-term and high-frequency usage scenarios of internal business data of enterprises often occur. Such problems are often dealt with after the fact, and cannot be prevented in advance based on user portrait prediction and evaluation. Second, the implemented functions cannot test the application effectiveness, and the user experience lacks a clear feedback channel, and feedback also requires time and manpower costs. Third, it is impossible to make accurate recommendations and marketing based on user preferences. Fourth, in order to expand the market and explore business opportunities, the construction party only decides the direction of function construction based on personal experience, which is theoretical and divorced from the actual needs of users.

[0005] The most common methodology for approximating user portrait analysis is the RFM model. The RFM model is an important tool and means for outlets to measure the current user value and customer potential value. It is a method theory specifically used to study user structure. The model is divided into three dimensions: R, F, and M. They are: R (Recency) refers to the most recent consumption. This dimension measures the time of the user's last consumption. The closer the time of the last consumption, the greater the user value; F (Frequency) refers to the consumption frequency. This dimension measures the frequency of user consumption over a period of time. The higher the user's consumption frequency, the greater the user value; M (Monetary) refers to the total consumption amount. The higher the user's consumption amount, the greater the user value. The RFM model uses the above three dimensions as coordinate axes to establish a three-dimensional coordinate system. Users will be subdivided into 8 different categories to achieve user value differentiation. However, the three dimensions of the RFM model have limitations for B-side information products with parent-subsidiary relationships. On the one hand, the indicators of the three dimensions are too thin to fully and deeply explore the user value; on the other hand, the model focuses on evaluating user value and does not explore the value of the objects operated by users, which is often the key to marketing, that is, it is impossible to evaluate the value of marketing items. Summary of the invention

[0006] In order to overcome the problems existing in the above-mentioned related technologies, the present application provides a recommendation decision method and device based on a user portrait model of a front-end application system.

[0007] According to a first aspect of an embodiment of the present application, a suggestion decision method based on a user portrait model of a front-end application system is provided, comprising:

[0008] Collect user behavior data of each user based on the pre-built user access behavior monitoring indicator system;

[0009] Building a user portrait model based on the user behavior data;

[0010] Obtaining a user portrait report using the user portrait model;

[0011] Based on the user portrait report, using a pre-established analysis and prediction model, obtain a predicted recommended decision plan;

[0012] The analysis and prediction model is constructed by using historical user portrait reports and historical recommended decision plans.

[0013] Preferably, collecting user behavior data of each user according to a pre-built user access behavior monitoring indicator system includes:

[0014] According to the pre-built user access behavior monitoring indicator system, the user behavior data of each user is automatically captured by the front-end embedding point, and the user behavior data of each user is recorded in the pre-built user footprint log ledger.

[0015] Preferably, the user behavior data includes at least the following nine types:

[0016] The proportion of access functions, the proportion of access time, the proportion of demand, the proportion of feasibility study amount, the number of visiting users, the total number of unit users, the rate of new users, the rate of user churn and the frequency of access.

[0017] Preferably, the constructing of a user portrait model based on the user behavior data includes:

[0018] Utilizing the user behavior data, the user stickiness, user activity and user voice are calculated respectively;

[0019] The user portrait model is constructed by utilizing the user stickiness, the user activity and the user voice.

[0020] Preferably, the calculating user stickiness by using the user behavior data includes:

[0021] The user access ratio is calculated using the number of access users and the total number of unit users;

[0022] Based on a first preset rule, determining the user access ratio score according to the user access ratio;

[0023] The user change rate is calculated using the user addition rate and user churn rate;

[0024] Based on a second preset rule, determining the user change rate score according to the user change rate;

[0025] Based on a third preset rule, determining the access frequency score according to the user's access frequency;

[0026] The user stickiness is calculated using the user access ratio score, the user change rate score and the access frequency score.

[0027] Preferably, the calculating the user activity by using the user behavior data includes:

[0028] The user activity is calculated using the proportion of the number of access functions and the proportion of the access duration.

[0029] Preferably, the calculating the user's voice rights by using the user behavior data includes:

[0030] The user's voice is calculated by using the proportion of the number of demands and the proportion of the feasibility study amount.

[0031] Preferably, the obtaining of a user portrait report by using the user portrait model includes:

[0032] Determine the user type of each user by using the user portrait model;

[0033] Based on a preset report template, a user portrait report is generated according to the user type of each user.

[0034] Preferably, the determining the user type of each user by using the user portrait model includes:

[0035] The users corresponding to the user stickiness, user activity and user voice in the user portrait model that meet the first preset condition are important development users;

[0036] The users corresponding to the user stickiness, user activity and user voice in the user portrait model that meet the second preset condition are core value users;

[0037] The users corresponding to the user stickiness, user activity and user voice in the user portrait model that meet the third preset condition are general development users;

[0038] The users corresponding to the user stickiness, user activity and user voice in the user portrait model that meet the fourth preset condition are the users to be mined;

[0039] The users corresponding to the user stickiness, user activity and user voice in the user portrait model that meet the fifth preset condition are general retention users;

[0040] The users corresponding to the user stickiness, user activity and user voice in the user portrait model that meet the sixth preset condition are generally retained users;

[0041] The users who meet the seventh preset condition in the user portrait model, such as user stickiness, user activity and user voice, are considered important users to be retained;

[0042] The users whose user stickiness, user activity and user voice meet the eighth preset condition in the user portrait model are important retained users.

[0043] Preferably, the process of establishing the analysis and prediction model includes:

[0044] Collect historical user profile reports and historical recommended decision plans;

[0045] Preprocessing the historical user portrait reports and the historical recommended decision solutions to obtain processed historical user portrait reports and processed historical recommended decision solutions;

[0046] constructing a data set using the processed historical user profile reports and the processed historical recommended decision solutions;

[0047] The large model is trained using the data set to obtain a trained large model, and the trained large model is the analysis and prediction model.

[0048] Preferably, the calculation formula for the user access ratio includes:

[0049] A1=uv / uvt

[0050] The calculation formula of the user change rate includes:

[0051] A2=uar-ucr

[0052] The calculation formula of the user stickiness includes:

[0053] A=W1*a1+W2*a2+W3*a3

[0054] In the above formula, A1 is the user access ratio, uv is the number of visiting users, uvt is the total number of unit users, A2 is the user change rate, uar is the user addition rate, ucr is the user churn rate, A is the user stickiness, W1 is the weight of the user access ratio score, a1 is the user access ratio score, W2 is the weight of the user change rate score, a2 is the user change rate score, W3 is the weight of the access frequency score, and a3 is the access frequency score.

[0055] Preferably, the calculation formula for the user activity includes:

[0056] B=W4*b1+W5*b2

[0057] In the above formula, B is the user activity, W4 is the weight of the proportion of the number of access functions, b1 is the proportion of the number of access functions, W5 is the weight of the proportion of access time, and b2 is the proportion of access time.

[0058] Preferably, the calculation formula of the user's voice rights includes:

[0059] C=W6*c1+W7*c2

[0060] In the above formula, C is the user's voice, W6 is the weight of the proportion of demand number, c1 is the proportion of demand number, W7 is the weight of the proportion of feasibility study amount, and c2 is the proportion of feasibility study amount.

[0061] According to a second aspect of an embodiment of the present application, a front-end application system user portrait construction device is provided, comprising:

[0062] A collection unit, used to collect user behavior data of each user according to a pre-built user access behavior monitoring indicator system;

[0063] A construction unit, used to construct a user portrait model based on the user behavior data;

[0064] An acquisition unit, used for acquiring a user portrait report using the user portrait model;

[0065] A prediction unit, configured to obtain a predicted recommended decision plan based on the user portrait report and using a pre-established analysis and prediction model;

[0066] The analysis and prediction model is constructed by using historical user portrait reports and historical recommended decision plans.

[0067] According to a third aspect of an embodiment of the present application, there is provided an electronic device, comprising: at least one processor and a memory; the memory and the processor are connected via a bus;

[0068] The memory is used to store one or more programs;

[0069] When the one or more programs are executed by the at least one processor, the recommendation decision method based on the user portrait model of the front-end application system is implemented.

[0070] According to the fourth aspect of the embodiment of the present application, a readable storage medium is provided, on which an execution program is stored. When the execution program is executed, the recommendation decision method based on the user portrait model of the front-end application system is implemented.

[0071] The technical solution provided by the present invention has the following beneficial effects:

[0072] The present invention provides a method and device for making recommendations based on a user portrait model of a front-end application system, including: collecting user behavior data of each user according to a pre-constructed user access behavior monitoring indicator system; constructing a user portrait model based on user behavior data; obtaining a user portrait report using the user portrait model; obtaining a predicted recommendation decision plan based on the user portrait report using a pre-established analysis and prediction model; the analysis and prediction model is constructed using historical user portrait reports and historical recommendation decision plans. The technical solution provided by the present invention can not only efficiently and quickly mine user needs and greatly reduce user research work; it can also save effort and quickly distinguish core users, clarify which are the key marketing groups, clarify which are the key construction directions, and alleviate the problem of long demand confirmation time caused by multiple levels within the enterprise. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0074] Figure 1 It is a flow chart of a suggestion decision method based on a user portrait model of a front-end application system provided by an embodiment of the present invention;

[0075] Figure 2 is a schematic diagram of a user portrait model provided by an embodiment of the present invention;

[0076] Figure 3 It is a structural block diagram of a suggestion decision device based on a user portrait model of a front-end application system provided by an embodiment of the present invention;

[0077] Figure 4 It is a structural block diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0078] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the following embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0079] Embodiment 1

[0080] The present invention provides a suggestion decision method based on the user portrait model of the front-end application system, such as Figure 1 As shown, the following steps are included:

[0081] Step 11: Collect user behavior data of each user based on the pre-built user access behavior monitoring indicator system;

[0082] Step 12: Build a user portrait model based on user behavior data;

[0083] Step 13: Use the user portrait model to obtain a user portrait report;

[0084] Step 14: Based on the user portrait report, use the pre-established analysis and prediction model to obtain the predicted recommended decision-making plan;

[0085] The analytical prediction model is built by utilizing historical user profile reports and historical recommended decision solutions.

[0086] The present invention is based on the RFM model. Aiming at the user portrait analysis scenario of the information-based B-end products with parent-subsidiary relationships, the RFM model is adaptively changed to build a complete indicator system for user portrait analysis, and a more adaptive user portrait analysis model and functional value evaluation model are constructed. For example, the RFM model only focuses on three indicators, has certain limitations, and lacks an analyzable user access behavior monitoring indicator system. In addition, the focus of previous user portrait research has been on user value classification, and there is a lack of product value evaluation. This model will pay attention to marketing objects and marketing items at the same time, and when evaluating the value of marketing items, costs are not considered due to the particularity of energy industry enterprises.

[0087] The present invention optimizes the underlying access log association rules and improves the quality of user data. At present, user portrait analysis usually relies on a large amount of user data for training and labeling. However, in practical applications, there are often problems such as low data quality and inaccurate labeling, which will lead to a decrease in the effect of the model and thus affect the accuracy of the user portrait. This method presets function changes in advance, builds a functional architecture, and constructs a version trace function.

[0088] It should be noted that the present invention does not limit the "pre-built user access behavior monitoring indicator system", which can be set by those skilled in the art based on experimental data or expert experience. In some embodiments, the user access behavior monitoring indicator system is a six-level indicator system; the first-level indicator includes: analysis content and analysis report template chapter settings;

[0089] The secondary indicators corresponding to the analysis content include: analysis direction, analysis indicators and analysis methods;

[0090] The secondary indicators corresponding to the chapter settings of the analysis report template include: user information portrait, user behavior portrait and user group portrait;

[0091] The three-level indicators corresponding to the analysis direction include: analysis direction establishment logic, user attributes, behavior attributes, resource attributes and time attributes;

[0092] The three-level indicators corresponding to the analysis indicators include: activity (i.e., the process of each visit, examining the user's participation in the visit; its purpose is to identify the resources / functions that users pay high attention to), stickiness (i.e., the situation of users continuously visiting and using the website over a period of time; its purpose is to explore resources / functions that enhance user stickiness and reduce user churn) and voice (i.e., measuring the value output created by users based on the website's business; its purpose is to identify the extent to which user demand creates revenue for the project platform);

[0093] The three-level indicators corresponding to the analysis methods include: RFM model expansion (i.e. establishing coordinates, distinguishing user levels, and giving priority to the needs of important development customers and important retention customers), comparative analysis (i.e. mining resources and functions with high user attention through statistics and comparison of analysis indicator data) and other common analysis methods;

[0094] The three-level indicators corresponding to the user information profile include: unit distribution, department distribution, occupation distribution and special distribution; among them, the statistical indicators of unit distribution, department distribution, occupation distribution and special distribution are the number of users, and the statistical time of unit distribution, department distribution, occupation distribution and special distribution is the cumulative time;

[0095] The three-level indicators corresponding to user behavior portraits include: user activity analysis, user stickiness analysis, and user voice analysis;

[0096] The three-level indicators corresponding to the user grouping portrait include: user persistence grouping, user importance grouping and user purchasing power grouping;

[0097] The four-level indicators corresponding to the logic of setting up the analysis direction include: the initiator of the behavior, the initiating action, the initiating object and the initiating time;

[0098] The four-level indicators corresponding to user attributes include: unit, department, position, project involved, gender, age, education and personality;

[0099] The four-level indicators corresponding to the behavior attributes include: query, browse, favorite, and data modification;

[0100] The four-level indicators corresponding to resource attributes include: application class, service class, document class and function class;

[0101] The four-level indicators corresponding to the time attribute include: cumulative, this month, and today;

[0102] The four-level indicators corresponding to user activity include: average number of pages visited and average dwell time;

[0103] The four-level indicators corresponding to user stickiness include: number of users, user access rate, new user ratio (for example: new user ratio> user churn rate: the product is in the development and growth stage; new user ratio=user churn rate: it is stable and in the mature and stable stage; new user ratio<user churn rate: the product is in the decline and decline stage), user churn rate, active user ratio, number of visits, visit frequency (to identify old users), visit duration, visit interval time (to identify the number of active users), first visit time (to identify new users) and last visit time (to identify lost users);

[0104] The four-level indicators corresponding to user voice include: number of demands (identifying purchasing users and silent users) and feasibility study amount;

[0105] The four-level indicators corresponding to the comparative analysis method include: resource access ranking, unit access ranking, department access ranking and access peak period;

[0106] The four-level indicators corresponding to other common analysis methods include: questionnaire survey method and TF-IDF algorithm;

[0107] The four-level indicators corresponding to user activity analysis include: statistical indicators (see the activity in the analysis indicators), statistical dimensions (such as unit / department / position / special project, resource, and operation behavior (see the lower-level indicators corresponding to the analysis direction)) and statistical time (for example, cumulative time and this month);

[0108] The four-level indicators corresponding to user stickiness analysis include: statistical indicators (see stickiness in analysis indicators), statistical dimensions (such as by unit / department / position / special project, by resource, by operation behavior), and statistical time (for example, cumulative time and this month);

[0109] The four-level indicators corresponding to the user voice analysis include: statistical indicators (see output in the analysis indicators), statistical dimensions (such as unit / department / position / special project) and statistical time (for example, cumulative time and this month);

[0110] The four-level indicators corresponding to user persistence grouping include: new users, old users, and lost users;

[0111] The four-level indicators corresponding to user importance clustering include: see RFM model;

[0112] The four-level indicators corresponding to the user purchasing power grouping include: purchasing users and silent users;

[0113] The five-level indicators corresponding to data modification include: addition (such as demand submission and service subscription), deletion and modification;

[0114] The five-level indicators corresponding to the number of users include: classification by usage continuity, classification by purchase behavior, and classification by importance;

[0115] The six-level indicators corresponding to the persistence classification include: new users (the corresponding seven-level indicators include: general users and active users), old users (the corresponding seven-level indicators include: general users and active users) and lost users:

[0116] The six-level indicators corresponding to the purchase behavior classification include: purchasing users and silent users;

[0117] The six-level indicators corresponding to the importance classification include: important development users, important value users, important retention users, important retained users, general retention users, general retained users, general development users and general value users.

[0118] It can be understood that by establishing a user access behavior monitoring indicator system, the user access behavior monitoring indicator logic can be determined, as shown in Table 1.

[0119] Table 1 User access behavior monitoring indicator logic

[0120]

[0121]

[0122] The present invention determines the user behavior data that needs to be collected by establishing user access behavior monitoring indicator rules, and can also support user behavior analysis applications such as user value and behavior characteristics; then based on the user access behavior monitoring indicators, the RFM model is expanded to build a user portrait model suitable for state-owned enterprises and B-end products.

[0123] This paper is based on the RFM customer value analysis model theory and the user access behavior monitoring index system, and constructs a user portrait model from three aspects: user stickiness, user activity, and user voice. This is the main idea for building the research model. From a technical perspective, research and development builds functional identity ledgers and access footprint ledgers to collect user information and support the practical application and implementation of user portrait models.

[0124] Further, step 11 includes:

[0125] Based on the pre-built user access behavior monitoring indicator system, the user behavior data of each user is automatically captured by the front-end embedding point, and the user behavior data of each user is recorded in the pre-established user footprint log ledger.

[0126] In some embodiments, a function account can also be established to identify the function identity through the function code, and prepare the version statistics basis for the later function iteration and name change. The function code setting rule is that the function code is 7 digits in total. If it is less than 7, it is padded with 0. The first digit identifies the first-level function, and the next 2-3, 4-5, and 5-7 digits identify the second-level, third-level, and fourth-level functions respectively. For example, the function ID coding example shown in Table 2 and the user footprint log account field table shown in Table 3.

[0127] Table 2 Example of function ID coding

[0128]

[0129] Table 3 User footprint log account field table

[0130]

[0131] The present invention optimizes the underlying access log association rules and improves the quality of user data by building a functional identity ledger and an access footprint ledger.

[0132] Furthermore, user behavior data includes at least the following nine types:

[0133] The proportion of access functions, the proportion of access time, the proportion of demand, the proportion of feasibility study amount, the number of visiting users, the total number of unit users, the rate of new users, the rate of user churn and the frequency of access.

[0134] It is understandable that user behavior generally refers to the behavior of users clicking on function pages on the system. Once the conditions for data indicators exist, it is subdivided into four types of click behaviors: querying data, adding data, modifying data, and deleting data.

[0135] Further, step 12 includes:

[0136] Step 121: Calculate user stickiness, user activity, and user voice using user behavior data;

[0137] like Figure 2 As shown, step 122: construct a user portrait model using user stickiness, user activity and user voice.

[0138] It is understandable that the user portrait model constructed through the three dimensions of user stickiness, user activity, and user voice is a three-dimensional model.

[0139] Further, step 121 includes: step 1211: using user behavior data to calculate user stickiness; step 1211, including:

[0140] Step 1211a: Calculate the user access ratio using the number of access users and the total number of unit users;

[0141] Specifically, the calculation formula for user access ratio includes:

[0142] A1=uv / uvt

[0143] In the above formula, A1 is the user access ratio, uv is the number of access users, and uvt is the total number of unit users;

[0144] Step 1211b: Based on the first preset rule, determine the user access ratio score according to the user access ratio;

[0145] Step 1211c: Calculate the user change rate using the user addition rate and the user churn rate;

[0146] Specifically, the calculation formula for user change rate includes:

[0147] A2=uar-ucr

[0148] In the above formula, A2 is the user change rate, uar is the user addition rate, and ucr is the user churn rate;

[0149] Step 1211d: Based on the second preset rule, determine the user change rate score according to the user change rate;

[0150] Step 1211e: Based on the third preset rule, determine the access frequency score according to the user's access frequency;

[0151] Step 1211f: Calculate user stickiness using the user access ratio score, user change rate score, and access frequency score;

[0152] Specifically, the calculation formula for user stickiness includes:

[0153] A=W1*a1+W2*a2+W3*a3

[0154] A is user stickiness, W1 is the weight of the user access ratio score, a1 is the user access ratio score, W2 is the weight of the user change rate score, a2 is the user change rate score, W3 is the weight of the access frequency score, and a3 is the access frequency score.

[0155] It is understandable that the access frequency generally refers to the interval between visits / clicks.

[0156] Further, step 121 includes: step 1212: calculating user activity using user behavior data;

[0157] Step 1212 includes: calculating the user activity by using the percentage of access functions and the percentage of access duration;

[0158] Specifically, the calculation formula for user activity includes:

[0159] B=W4*b1+W5*b2

[0160] In the above formula, B is the user activity, W4 is the weight of the proportion of the number of access functions, b1 is the proportion of the number of access functions, W5 is the weight of the proportion of access time, and b2 is the proportion of access time.

[0161] It can be understood that the number of accessed functions refers to the number of functions clicked by user X within the statistical time interval. The percentage of the number of accessed functions can be, but is not limited to: the number of functions clicked by user X within the statistical time interval / the number of all available functions. The access duration refers to the time a user stays on a certain function page. The percentage of access duration can be, but is not limited to: the percentage of access duration of user A accessing function X = user A's access duration of function X / the sum of function X's access durations.

[0162] Further, step 121 includes: step 1213: calculating the user's voice right by using the user behavior data;

[0163] Step 1213 includes: using the demand number ratio and the feasibility study amount ratio to calculate the user's voice;

[0164] Specifically, the calculation formula for user voice rights includes:

[0165] C=W6*c1+W7*c2

[0166] In the above formula, C is the user's voice, W6 is the weight of the proportion of demand number, c1 is the proportion of demand number, W7 is the weight of the proportion of feasibility study amount, and c2 is the proportion of feasibility study amount.

[0167] It is understandable that user demand generally refers to the business request or management requirement proposed by the user side to realize a certain business scenario through the construction of X product functions, which is finally realized in the Internet product function. The proportion of demand number is the number of demands submitted by a certain user / the sum of the number of demands submitted by all users. The feasibility study amount is the investment amount on the user side of the Internet project, which is used to evaluate the user's investment capacity. For example, the proportion of the feasibility study amount is the investment amount of a certain user in project A / the investment amount of all users in project A.

[0168] It should be noted that the present invention does not limit the "weight of user access ratio score", "weight of user change rate score", "weight of access frequency score", "weight of access function number proportion", "weight of access duration proportion", "weight of demand number proportion" and "weight of feasibility study amount proportion", which can be set by technical personnel in this field according to experimental data or expert experience. In some embodiments, the hierarchical analysis method based on qualitative and quantitative analysis methods can be used in combination with fuzzy comprehensive evaluation method to set each weight, so as to more easily and quickly construct the weight of the data health evaluation index system with multi-objective, multi-criteria and multi-structural characteristics.

[0169] For example, through the hierarchical analysis method, the weight of the proportion of demand number is 0.2, the weight of the proportion of feasibility study amount is 0.8, the weight of the proportion of access functions is 0.3, the weight of the proportion of access time is 0.7, the weight of household access ratio score is 0.15, the weight of user change rate score is 0.15, and the weight of access frequency score is 0.7.

[0170] It should be noted that the embodiments of the present invention do not limit the "first preset rule", "second preset rule" and "third preset rule", which can be set by those skilled in the art according to engineering needs, experimental data or expert experience. In some embodiments, the first preset rule includes:

[0171] If the user access ratio falls within the first threshold range, the user access ratio score is the first score;

[0172] If the user access ratio falls within the second threshold range, the user access ratio score is the second score;

[0173] If the user access ratio falls within the third threshold range, the user access ratio score is the third score;

[0174] If the user access ratio falls within the fourth threshold range, the user access ratio score is the fourth score;

[0175] If the user access ratio falls within the fifth threshold range, the user access ratio score is the fifth score.

[0176] For example, if the user access ratio = 0, 0 points are recorded; if the user access ratio ∈(0, 0.2], 1 point is recorded; if the user access ratio ∈(0.2, 0.4], 2 points are recorded; if the user access ratio ∈(0.4, 0.6], 3 points are recorded; if the user access ratio ∈(0.6, 0.8], 4 points are recorded; if the user access ratio ∈(0.8, 1], 5 points are recorded.

[0177] In some embodiments, the second preset rule includes:

[0178] If the user change rate falls within the sixth threshold range, the user change rate score is the sixth score;

[0179] If the user change rate falls within the seventh threshold range, the user change rate score is the seventh score;

[0180] If the user change rate falls within the eighth threshold range, the user change rate score is the eighth score;

[0181] If the user change rate falls within the ninth threshold range, the user change rate score is the ninth score;

[0182] If the user change rate falls within the tenth threshold range, the user change rate score is the tenth value;

[0183] If the user change rate falls within the eleventh threshold range, the user change rate score is the eleventh score;

[0184] If the user change rate falls within the twelfth threshold range, the user change rate score is the twelfth score;

[0185] If the user change rate falls within the thirteenth threshold range, the user change rate score is the thirteenth value;

[0186] If the user change rate falls within the fourteenth threshold range, the user change rate score is the fourteenth score.

[0187] For example, user change rate ∈(∞, -0.2], counted as -5 points; user change rate ∈(-0.2, -0.1], counted as -3 points; user change rate ∈(-0.1, -0.05], counted as -2 points; user change rate ∈(-0.05, 0], counted as -1 point; user change rate = 0, counted as 0 points; user change rate ∈(0, 0.05], counted as 1 point; user change rate ∈(0.05, 0.1], counted as 2 points; user change rate ∈(0.1, 0.2], counted as 3 points; user change rate ∈(0.1, 0.2], counted as 5 points.

[0188] In some embodiments, the third preset rule includes:

[0189] If the user's access frequency falls within the fifteenth threshold range, the access frequency score is the fifteenth score;

[0190] If the user's access frequency falls within the sixteenth threshold range, the access frequency score is the sixteenth score;

[0191] If the user's access frequency falls within the seventeenth threshold range, the access frequency score is the seventeenth score;

[0192] If the user's access frequency falls within the eighteenth threshold range, the access frequency score is the eighteenth score;

[0193] If the user's access frequency falls within the nineteenth threshold range, the access frequency score is the nineteenth score;

[0194] If the user's access frequency falls within the twentieth threshold range, the access frequency score is the twentieth tenth value;

[0195] If the user's access frequency falls within the 21st threshold range, the access frequency score is the 21st score;

[0196] If the user's access frequency falls within the 22nd threshold range, the access frequency score is the 22nd score;

[0197] If the user's access frequency falls within the 23rd threshold range, the access frequency score is the 23rd score;

[0198] If the user's access frequency falls within the 24th threshold range, the access frequency score is the 24th score;

[0199] If the user's access frequency falls within the 25th threshold range, the access frequency score is the 25th score.

[0200] For example, if the access frequency ∈(365 days / time, ∞), -5 points are given; if the access frequency ∈(180 days / time, 365 days / time], -4 points are given; if the access frequency ∈(150 days / time, 180 days / time], -3 points are given; if the access frequency ∈(120 days / time, 150 days / time], -2 points are given; if the access frequency ∈(90 days / time, 120 days / time], -1 point is given; if the access frequency = 90 days / time, 0 points are given; if the access frequency ∈(30 days / time, 90 days / time], 1 point is given; if the access frequency ∈(1 day / time, 30 days / time], 2 points are given; if the access frequency ∈(1 hour / time, 1 day / time], 3 points are given; if the access frequency ∈(1 minute / time, 1 hour / time], 4 points are given; if the access frequency ∈(0, 1 minute / time], 5 points are given.

[0201] Further, step 13 includes:

[0202] Step 131: Determine the user type of each user using the user portrait model;

[0203] Step 132: Generate a user portrait report based on a preset report template and according to the user type of each user.

[0204] It should be noted that the present invention does not limit the "preset report template", which can be set by those skilled in the art based on experimental data or expert experience.

[0205] Further, step 131 includes:

[0206] In the user portrait model, users who meet the first preset conditions of user stickiness, user activity and user voice are considered important development users;

[0207] In the user portrait model, users who meet the second preset conditions of user stickiness, user activity and user voice are core value users;

[0208] In the user portrait model, users who meet the third preset condition of user stickiness, user activity and user voice are general development users;

[0209] In the user portrait model, users who meet the fourth preset condition of user stickiness, user activity and user voice are the users to be mined;

[0210] In the user portrait model, users who meet the fifth preset condition of user stickiness, user activity and user voice are generally retained users;

[0211] In the user portrait model, users who meet the sixth preset condition of user stickiness, user activity and user voice are generally retained users;

[0212] In the user portrait model, users who meet the seventh preset condition, such as user stickiness, user activity, and user voice, are considered important users to be retained.

[0213] Users who meet the eighth preset condition in the user portrait model in terms of user stickiness, user activity and user voice are considered important retained users.

[0214] It should be noted that the present invention does not limit the "first preset condition, the second preset condition, the third preset condition, the fourth preset condition, the fifth preset condition, the sixth preset condition, the seventh preset condition and the eighth preset condition", which can be limited by those skilled in the art based on experimental data or expert experience. For example, the first preset condition, the second preset condition, the third preset condition, the fourth preset condition, the fifth preset condition, the sixth preset condition, the seventh preset condition and the eighth preset condition can be set to, but not limited to, the threshold range corresponding to user stickiness, user activity and user voice, so as to further limit the user type.

[0215] Furthermore, the method further comprises: Step 10: establishing an analysis and prediction model; Step 10 comprises:

[0216] Step 101: Collect historical user profile reports and historical recommended decision plans;

[0217] Step 102: pre-processing the historical user portrait reports and the historical recommended decision solutions to obtain processed historical user portrait reports and processed historical recommended decision solutions;

[0218] In some embodiments, "preprocessing historical user profile reports and historical recommended decision solutions" means preprocessing the collected data before model training, which may include but is not limited to cleaning, annotating, and formatting the collected historical user profile reports and historical recommended decision solutions, so as to ensure that the large model can learn useful information from them;

[0219] Step 103: construct a data set using the processed historical user profile reports and the processed historical recommended decision solutions;

[0220] Step 104: Use the data set to train the large model to obtain a trained large model, where the trained large model is an analysis and prediction model.

[0221] In some embodiments, the data set can also be used to train deep learning models such as deep neural network models.

[0222] It should be noted that the method of “using a data set to train a large model” involved in the present invention is well known to those skilled in the art, and therefore, its specific implementation method will not be described in detail.

[0223] In practical applications, the proposed decision-making method based on the user portrait model of the front-end application system proposed by the present invention can be used to obtain the predicted recommended decision-making plan, which can be directly used to guide the actual construction work. The predicted recommended decision-making plan can also be adjusted according to actual needs or expert experience. By comparing user data from different periods, the effectiveness of the management means of the evaluation report and the recommended decision-making plan can be evaluated. If it deviates from the actual situation, the weights of the demand number proportion, feasibility study amount proportion, access function number proportion, access time proportion, user access ratio score, user change rate score and access frequency score can be adjusted for applicability, or the statistical indicators can be adjusted and corrected, so as to re-collect user behavior data and construct a new user portrait model.

[0224] In some embodiments, after using the predicted recommended decision-making plan to guide the actual construction work, user behavior will change, that is, the user footprint will change. This requires re-collecting user behavior data and building a new user portrait model to form a new user portrait report.

[0225] The present invention proposes a recommendation decision-making method based on the user portrait model of the front-end application system. Based on the RFM customer value analysis model theory, the user access behavior data of the front-end system is statistically analyzed, and the user behavior portrait analysis technology research of the middle-office operation portal system of the project is carried out. The value and behavior characteristics of various users are analyzed, which can support the promotion and application of key users of anchoring middle-office functions, and clarify the direction for the planning and construction of hot functions and the optimization and improvement of low-visit functions, laying the foundation for further improving user experience, optimizing product functions, and achieving business growth.

[0226] The present invention aims at the shortcomings existing in the prior art: lack of a complete user access behavior monitoring indicator system, lack of adaptability to user portrait analysis of B-end products, lack of evaluation of functional value, the analysis is limited to users, and the existing functional value evaluation technology generally adopts cost-benefit analysis, but this situation is not used in state-owned enterprises in the energy industry that attach importance to information security. A recommendation decision method based on the user portrait model of the front-end application system is proposed. By collecting user behavior data of each user according to a pre-constructed user access behavior monitoring indicator system, a user portrait model is constructed based on user behavior data, and a user portrait report is obtained by using the user portrait model. Based on the user portrait report, a pre-established analysis and prediction model is used to obtain a predicted recommendation decision plan, wherein the analysis and prediction model is constructed using historical user portrait reports and historical recommendation decision plans, which realizes efficient and rapid mining of user needs and greatly reduces user research work; saves effort and quickly distinguishes core users, and clarifies which are key marketing groups; saves effort and quickly distinguishes hot functions, and clarifies which are key construction directions. Alleviate the problem of long demand confirmation time caused by internal enterprise levels.

[0227] Embodiment 2

[0228] The present invention also provides a suggestion decision device based on the user portrait model of the front-end application system, such as Figure 3 As shown, including:

[0229] Collect user behavior data of each user based on the pre-built user access behavior monitoring indicator system;

[0230] A collection unit, used to collect user behavior data of each user according to a pre-built user access behavior monitoring indicator system;

[0231] A construction unit, used to construct a user portrait model based on user behavior data;

[0232] An acquisition unit, used for acquiring a user portrait report using a user portrait model;

[0233] The prediction unit is used to obtain the predicted recommended decision-making plan based on the user portrait report and using the pre-established analysis and prediction model;

[0234] The analytical prediction model is built by utilizing historical user profile reports and historical recommended decision solutions.

[0235] Furthermore, the collection unit is specifically used for:

[0236] Based on the pre-built user access behavior monitoring indicator system, the user behavior data of each user is automatically captured by the front-end embedding point, and the user behavior data of each user is recorded in the pre-established user footprint log ledger.

[0237] Furthermore, user behavior data includes at least the following nine types:

[0238] The proportion of access functions, the proportion of access time, the proportion of demand, the proportion of feasibility study amount, the number of visiting users, the total number of unit users, the rate of new users, the rate of user churn and the frequency of access.

[0239] Furthermore, the construction unit comprises:

[0240] The calculation module is used to calculate user stickiness, user activity and user voice using user behavior data;

[0241] The construction module is used to build a user portrait model by utilizing user stickiness, user activity and user voice.

[0242] Furthermore, the computing module comprises:

[0243] The first calculation submodule is used to calculate the user access ratio by using the number of access users and the total number of unit users;

[0244] A first determination submodule, configured to determine a user access ratio score according to the user access ratio based on a first preset rule;

[0245] The second calculation submodule is used to calculate the user change rate by using the user addition rate and the user churn rate;

[0246] A second determination submodule, configured to determine a user change rate score according to the user change rate based on a second preset rule;

[0247] A third determination submodule, configured to determine an access frequency score according to a user's access frequency based on a third preset rule;

[0248] The third calculation submodule is used to calculate the user stickiness by using the user access ratio score, the user change rate score and the access frequency score.

[0249] Furthermore, the calculation module further includes:

[0250] The fourth calculation submodule is used to calculate the user activity by using the proportion of the number of access functions and the proportion of access duration.

[0251] Furthermore, the calculation module further includes:

[0252] The fifth calculation submodule is used to calculate the user's voice by using the proportion of demand numbers and the proportion of feasibility study amounts.

[0253] Furthermore, the acquisition unit includes:

[0254] A fourth determination submodule is used to determine the user type of each user using the user portrait model;

[0255] The generation submodule is used to generate a user portrait report based on a preset report template and according to the user type of each user.

[0256] Furthermore, the fourth determination submodule is specifically used for:

[0257] In the user portrait model, users who meet the first preset conditions of user stickiness, user activity and user voice are considered important development users;

[0258] In the user portrait model, users who meet the second preset conditions of user stickiness, user activity and user voice are core value users;

[0259] In the user portrait model, users who meet the third preset condition of user stickiness, user activity and user voice are general development users;

[0260] In the user portrait model, users who meet the fourth preset condition of user stickiness, user activity and user voice are the users to be mined;

[0261] In the user portrait model, users who meet the fifth preset condition of user stickiness, user activity and user voice are generally retained users;

[0262] In the user portrait model, users who meet the sixth preset condition of user stickiness, user activity and user voice are generally retained users;

[0263] In the user portrait model, users who meet the seventh preset condition, such as user stickiness, user activity, and user voice, are considered important users to be retained.

[0264] Users who meet the eighth preset condition in the user portrait model in terms of user stickiness, user activity and user voice are considered important retained users.

[0265] Furthermore, the device further includes: an establishment unit for establishing an analysis and prediction model; and an establishment unit for:

[0266] Collect historical user profile reports and historical recommended decision plans;

[0267] Preprocessing the historical user portrait reports and the historical recommended decision plans to obtain processed historical user portrait reports and processed historical recommended decision plans;

[0268] Constructing a data set using the processed historical user profile reports and the processed historical recommended decision solutions;

[0269] The large model is trained using the data set to obtain a trained large model, which is an analysis and prediction model. Furthermore, the calculation formula for the user access ratio includes:

[0270] A1=uv / uvt

[0271] The calculation formula for user turnover rate includes:

[0272] A2=uar-ucr

[0273] The calculation formula for user stickiness includes:

[0274] A=W1*a1+W2*a2+W3*a3

[0275] In the above formula, A1 is the user access ratio, uv is the number of visiting users, uvt is the total number of unit users, A2 is the user change rate, uar is the user addition rate, ucr is the user churn rate, A is the user stickiness, W1 is the weight of the user access ratio score, a1 is the user access ratio score, W2 is the weight of the user change rate score, a2 is the user change rate score, W3 is the weight of the access frequency score, and a3 is the access frequency score.

[0276] Furthermore, the calculation formula for user activity includes:

[0277] B=W4*b1+W5*b2

[0278] In the above formula, B is the user activity, W4 is the weight of the proportion of the number of access functions, b1 is the proportion of the number of access functions, W5 is the weight of the proportion of access time, and b2 is the proportion of access time.

[0279] Furthermore, the calculation formula for the user's voice includes:

[0280] C=W6*c1+W7*c2

[0281] In the above formula, C is the user's voice, W6 is the weight of the proportion of demand number, c1 is the proportion of demand number, W7 is the weight of the proportion of feasibility study amount, and c2 is the proportion of feasibility study amount.

[0282] It can be understood that the system embodiment provided above corresponds to the method embodiment described above, and the corresponding specific contents can be referenced to each other and will not be repeated here.

[0283] It can be understood that the same or similar parts of the above embodiments can be referenced to each other, and the contents not described in detail in some embodiments can refer to the same or similar contents in other embodiments.

[0284] Embodiment 3

[0285] like Figure 4 As shown, the present invention also provides an electronic device, which may be a computer device, a single-chip device, an intelligent mobile device, etc. The electronic device in this embodiment may include a processor, a memory, a transceiver component, etc. The memory, the processor, and the transceiver component are connected via a bus; the memory may be used to store an execution program, and an exemplary execution program may include instructions; the processor is used to execute the instructions stored in the memory. The memory may also be used to store data, which may be called and / or modified when the instructions are executed.

[0286] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions in a storage medium to implement corresponding method flows or corresponding functions, so as to implement the steps of a recommendation decision-making method based on a user portrait model of a front-end application system in the above-mentioned embodiment.

[0287] Embodiment 4

[0288] Based on the same inventive concept, the present invention also provides a readable storage medium, specifically an electronic device readable storage medium (Memory), which is a memory device in an electronic device for storing programs and data. It can be understood that the storage medium here can include both built-in storage media in electronic devices and, of course, extended storage media supported by electronic devices. The storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by a processor are also stored in the storage space, and these instructions can be one or more execution programs (including program codes). It should be noted that the storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory. The processor loads and executes one or more instructions stored in the storage medium, which can implement the steps of a recommendation decision method based on a user portrait model of a front-end application system in the above embodiment.

[0289] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0290] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0291] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0292] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0293] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A suggestion decision method based on a user portrait model of a front-end application system, characterized in that: include: Collect user behavior data of each user based on the pre-built user access behavior monitoring indicator system; Building a user portrait model based on the user behavior data; Obtaining a user portrait report using the user portrait model; Based on the user portrait report, using a pre-established analysis and prediction model, obtain a predicted recommended decision plan; The analysis and prediction model is constructed by using historical user portrait reports and historical recommended decision plans.

2. The method according to claim 1, characterized in that: The user behavior data of each user is collected according to the pre-built user access behavior monitoring indicator system, including: According to the pre-built user access behavior monitoring indicator system, the user behavior data of each user is automatically captured by the front-end embedding point, and the user behavior data of each user is recorded in the pre-built user footprint log ledger.

3. The method according to claim 1, characterized in that The user behavior data includes at least the following nine types: The proportion of access functions, the proportion of access time, the proportion of demand, the proportion of feasibility study amount, the number of visiting users, the total number of unit users, the rate of new users, the rate of user churn and the frequency of access.

4. The method according to claim 3, characterized in that The constructing of a user portrait model based on the user behavior data includes: Utilizing the user behavior data, the user stickiness, user activity and user voice are calculated respectively; The user portrait model is constructed by utilizing the user stickiness, the user activity and the user voice.

5. The method according to claim 4, characterized in that The calculating user stickiness by using the user behavior data includes: The user access ratio is calculated using the number of access users and the total number of unit users; Based on a first preset rule, determining the user access ratio score according to the user access ratio; The user change rate is calculated using the user addition rate and user churn rate; Based on a second preset rule, determining the user change rate score according to the user change rate; Based on a third preset rule, determining the access frequency score according to the user's access frequency; The user stickiness is calculated using the user access ratio score, the user change rate score and the access frequency score.

6. The method according to claim 4, characterized in that The calculating the user activity by using the user behavior data includes: The user activity is calculated using the proportion of the number of access functions and the proportion of the access duration.

7. The method according to claim 4, characterized in that The calculating the user's speaking rights by using the user behavior data includes: The user's voice is calculated by using the proportion of the number of demands and the proportion of the feasibility study amount.

8. The method according to claim 4, characterized in that The obtaining of a user portrait report by using the user portrait model includes: Determine the user type of each user by using the user portrait model; Based on a preset report template, a user portrait report is generated according to the user type of each user.

9. The method according to claim 8, characterized in that Determining the user type of each user by using the user portrait model includes: The users corresponding to the user stickiness, user activity and user voice in the user portrait model that meet the first preset condition are important development users; The users corresponding to the user stickiness, user activity and user voice in the user portrait model that meet the second preset condition are core value users; The users corresponding to the user stickiness, user activity and user voice in the user portrait model that meet the third preset condition are general development users; The users corresponding to the user stickiness, user activity and user voice in the user portrait model that meet the fourth preset condition are the users to be mined; The users corresponding to the user stickiness, user activity and user voice in the user portrait model that meet the fifth preset condition are general retention users; The users corresponding to the user stickiness, user activity and user voice in the user portrait model that meet the sixth preset condition are generally retained users; The users who meet the seventh preset condition in the user portrait model, such as user stickiness, user activity and user voice, are considered important users to be retained; The users whose user stickiness, user activity and user voice meet the eighth preset condition in the user portrait model are important retained users.

10. The method according to claim 1, characterized in that The process of establishing the analysis and prediction model includes: Collect historical user profile reports and historical recommended decision plans; Preprocessing the historical user portrait reports and the historical recommended decision solutions to obtain processed historical user portrait reports and processed historical recommended decision solutions; constructing a data set using the processed historical user profile reports and the processed historical recommended decision solutions; The large model is trained using the data set to obtain a trained large model, and the trained large model is the analysis and prediction model.

11. The method according to claim 5, characterized in that The calculation formula of the user access ratio includes: A1=uv / uvt The calculation formula of the user change rate includes: A2=uar-ucr The calculation formula of the user stickiness includes: A=W1*a1+W2*a2+W3*a3 In the above formula, A1 is the user access ratio, uv is the number of visiting users, uvt is the total number of unit users, A2 is the user change rate, uar is the user addition rate, ucr is the user churn rate, A is the user stickiness, W1 is the weight of the user access ratio score, a1 is the user access ratio score, W2 is the weight of the user change rate score, a2 is the user change rate score, W3 is the weight of the access frequency score, and a3 is the access frequency score.

12. The method according to claim 6, characterized in that The calculation formula of the user activity includes: B=W4*b1+W5*b2 In the above formula, B is the user activity, W4 is the weight of the proportion of the number of access functions, b1 is the proportion of the number of access functions, W5 is the weight of the proportion of access time, and b2 is the proportion of access time.

13. The method according to claim 7, characterized in that The calculation formula of the user's voice rights includes: C=W6*c1+W7*c2 In the above formula, C is the user's voice, W6 is the weight of the proportion of demand number, c1 is the proportion of demand number, W7 is the weight of the proportion of feasibility study amount, and c2 is the proportion of feasibility study amount.

14. A front-end application system user portrait construction device, characterized in that: include: A collection unit, used to collect user behavior data of each user according to a pre-built user access behavior monitoring indicator system; A construction unit, used to construct a user portrait model based on the user behavior data; An acquisition unit, used for acquiring a user portrait report using the user portrait model; A prediction unit, configured to obtain a predicted recommended decision plan based on the user portrait report and using a pre-established analysis and prediction model; The analysis and prediction model is constructed by using historical user portrait reports and historical recommended decision plans.

15. An electronic device, characterized in that: include: at least one processor and memory; The memory and the processor are connected via a bus; The memory is used to store one or more programs; When the one or more programs are executed by the at least one processor, the front-end application system user portrait construction method as described in any one of claims 1 to 13 is implemented.

16. A readable storage medium, characterized in that: An execution program is stored thereon, and when the execution program is executed, the front-end application system user portrait construction method as described in any one of claims 1 to 13 is implemented.

Images