A user analysis method and a terminal

Abstract

The present invention provides a user analysis method and a terminal, which obtain first contract information of all contracts corresponding to a user identifier, where the first contract information includes a contract status; obtain second contract information whose contract status meets a preset contract status from the first contract information, and obtain user behaviors corresponding to the second contract information; and obtain a user behavior score based on the user behaviors and the second contract information. The present invention uses contract information as a consideration criterion for user behaviors, realizes an objective evaluation of user behaviors, and realizes a user behavior score according to the contract content and the obtained user behaviors, can automatically evaluate user behaviors without manual operation, and improves the efficiency of user analysis.

Description

Technical Field

[0001] The present invention relates to the field of user data processing, and particularly to a user analysis method and a terminal. Background Art

[0002] With the expansion of the company scale and the development of business, the amount of relevant business data is becoming increasingly huge, and it is becoming more and more difficult to carry out refined operation work based on user business behaviors. Therefore, we need a technical means to improve our work efficiency and enhance our refined operation work. Summary of the Invention

[0003] The technical problem to be solved by the present invention is: to provide a user analysis method and a terminal, so as to achieve an objective, accurate and automated evaluation of user behaviors.

[0004] To solve the above technical problem, a technical solution adopted by the present invention is:

[0005] A user analysis method, comprising the steps of:

[0006] Obtaining first contract information of all contracts corresponding to a user identifier, where the first contract information includes a contract status;

[0007] Obtaining second contract information whose contract status meets a preset contract status from the first contract information, and obtaining user behaviors corresponding to the second contract information;

[0008] Obtaining a user behavior score based on the user behaviors and the second contract information.

[0009] To solve the above technical problem, another technical solution adopted by the present invention is:

[0010] A user analysis terminal, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, where when the processor executes the computer program, the following steps are implemented:

[0011] Obtaining first contract information of all contracts corresponding to a user identifier, where the first contract information includes a contract status;

[0012] Obtaining second contract information whose contract status meets a preset contract status from the first contract information, and obtaining user behaviors corresponding to the second contract information;

[0013] Obtaining a user behavior score based on the user behaviors and the second contract information.

[0014] The beneficial effects of the present invention are as follows: obtaining the first contract information corresponding to all contracts of the user identifier, separating the second contract information therefrom according to the preset contract status, obtaining the user behavior corresponding to the second contract information, associating the user behavior with the contract corresponding to the user, using the contract information as a consideration criterion for the user behavior, realizing an objective evaluation of the user behavior, and realizing the scoring of the user behavior according to the contract content and the obtained user behavior, which can automatically evaluate the user behavior without manual operation and improve the efficiency of user analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a flowchart of the steps of a user analysis method according to an embodiment of the present invention;

[0016] Figure 2 It is a schematic structural diagram of a user analysis terminal according to an embodiment of the present invention;

[0017] Figure 3 It is a schematic flowchart of a user analysis method according to an embodiment of the present invention;

[0018] Reference numeral description:

[0019] 1. A user analysis terminal; 2. A processor; 3. A memory. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0020] To describe in detail the technical content, the achieved objectives and the effects of the present invention, the following is described in conjunction with the embodiments and the accompanying drawings.

[0021] Please refer to Figure 1 , a user analysis method, including the steps:

[0022] Obtaining the first contract information of all contracts corresponding to the user identifier, where the first contract information includes the contract status;

[0023] Obtaining the second contract information whose contract status meets the preset contract status from the first contract information, and obtaining the user behavior corresponding to the second contract information;

[0024] Obtaining the user behavior score based on the user behavior and the second contract information.

[0025] As can be seen from the above description, the beneficial effects of the present invention are as follows: obtaining the first contract information corresponding to all contracts corresponding to the user identifier, separating the second contract information therefrom according to the preset contract status, obtaining the user behavior corresponding to the second contract information, associating the user behavior with the contract corresponding to the user, using the contract information as a consideration criterion for the user behavior, realizing an objective evaluation of the user behavior, and realizing a user behavior score according to the contract content and the obtained user behavior, enabling an automated evaluation of the user behavior without manual operation, and improving the efficiency of user analysis.

[0026] Further, before obtaining the second contract information whose contract status meets the preset contract status from the first contract information, it includes:

[0027] Obtaining all possible contract statuses existing;

[0028] Judging whether the possible contract status indicates the end of the contract. If not, setting the possible contract status as the preset contract status.

[0029] As can be seen from the above description, first obtain all possible contract statuses, and judge whether the possible contract status indicates the end of the contract. If so, it means that the user has no subsequent contract on the platform and there is no need to analyze the user. If not, it means that the user's contract with the platform has not all ended, and the analysis of the user behavior can make corresponding predictions about the possible subsequent behavior of the user regarding the contract, and corresponding preparations can be made in advance. For example, if the user often defaults, the user can be reminded in advance to avoid default caused by the user's forgetting.

[0030] Further, the first contract information includes the first signing time;

[0031] The obtaining of the user behavior score based on the user behavior and the second contract information includes:

[0032] Obtaining the current scoring period and marking the start node of the current scoring period as the beginning of the period;

[0033] Judging whether there is first contract information whose first signing time is earlier than the beginning-of-period time. If so, marking the first user identifier corresponding to the first contract information as a stock user, obtaining all first user behaviors corresponding to the first user identifier before the beginning-of-period time and the second contract information corresponding to the first user identifier; obtaining the user behavior score according to the beginning-of-period scoring algorithm based on the first user behavior and the second contract information;

[0034] Otherwise, mark the first user identifier corresponding to the first contract information as an incremental user, and obtain all the second user behaviors corresponding to the first user identifier and the second contract information corresponding to the first user identifier; calculate the user behavior statistical data corresponding to the second user behaviors, and obtain the user behavior score based on the user behavior statistical data and the second contract information according to the behavior scoring algorithm.

[0035] As can be seen from the above description, a scoring period is set, and users are divided into incremental users and existing users according to the signing time of the contract. Since the user data closer to the current time is more valuable for judging the current user behavior, the user behaviors are scored in segments according to the scoring period, making the final judgment structure of the user behavior more objective.

[0036] Further, the determination of whether there is first contract information with a first signing time earlier than the start time of the period further includes:

[0037] If so, obtain the first score based on the first user behavior and the second contract information according to the start-of-period scoring algorithm;

[0038] Obtain all the user behavior statistical data up to the start time of the period and obtain the second score according to the behavior scoring algorithm;

[0039] Accumulate the first score and the second score to obtain the user behavior score.

[0040] As can be seen from the above description, if the user is an existing user, in addition to calculating the score of the user before the start of the period, the score of the user after the start of the period is also calculated, and different algorithms are used for the two segments of scoring, then the time attribute of the user behavior can be reflected, making the finally obtained user behavior score more objective.

[0041] Further, the preset contract status includes the normal status, the legal affairs intervention status, the vehicle repossession status, the accident status, and the transfer to vehicle service status.

[0042] As can be seen from the above description, the normal status, the legal affairs intervention status, the vehicle repossession status, the accident status, and the transfer to vehicle service status all indicate that there is still a continuing contract with the user. Therefore, taking them as the preset contract status can exclude the contract status that is no longer relevant to the user and improve the timeliness of the user behavior scoring.

[0043] Further, the obtaining of the user behavior score based on the first user behavior and the second contract information according to the start-of-period scoring algorithm includes:

[0044] Calculate the behavior index information based on the first user behavior and the second contract information;

[0045] Obtain the preset index scoring table;

[0046] Obtain the score corresponding to the behavior index information according to the said index score table;

[0047] Obtain the user behavior score according to the said initial score calculation algorithm and all the said scores.

[0048] As can be seen from the above description, taking the second contract information as the standard, the index information corresponding to the first user behavior is objectively calculated, and the index information is scored according to the index score table, quantifying the evaluation of the user behavior, and the quantification is based on an objective standard, excluding the influence of subjective factors, making the final evaluation of the user more objective and true.

[0049] Furthermore, the said first contract information includes the account period date;

[0050] The said behavior index information includes the proportion of overdue times;

[0051] The said first user behavior includes the payment time;

[0052] The calculation of the behavior index information based on the said first user behavior and the said second contract information includes:

[0053] Obtain the account period date and the corresponding payment time of the account period date, record the total number of overdue periods corresponding to the account period dates without corresponding payment times as the number of overdue times, then the number of performance periods = the total number of periods occupied by the first contract information - the number of overdue periods;

[0054] The said proportion of overdue times = the number of overdue times / the number of performance periods.

[0055] As can be seen from the above description, calculating the proportion of overdue times can objectively obtain the historical performance situation of the user, so as to make a certain degree of prediction on whether the user can perform in the future, and thus make corresponding responses; for example, if the proportion of overdue times exceeds the preset number of times, products with a price higher than a certain value cannot be rented.

[0056] Furthermore, the said user behavior statistical data includes the most recent repayment time;

[0057] The obtaining of the user behavior score according to the said user behavior statistical data and the said second contract information according to the behavior score calculation algorithm includes:

[0058] Obtain the second contract information corresponding to the said most recent repayment time, and mark it as the repayment contract;

[0059] Obtain the account period date of the said repayment contract;

[0060] Obtain the said user behavior score according to the said most recent repayment time and the said account period date according to the said behavior score calculation algorithm.

[0061] As described above, obtaining the most recent repayment time and the billing cycle date, and separately processing the most recent repayment behavior relative to the current time. Since the user's repayment behavior is usually related to time, for example, having the repayment ability for a period of time, considering the most recent repayment behavior separately can more accurately predict the user's recent behavior and help make decisions.

[0062] Further, before obtaining the user behavior score based on the user behavior and the second contract information, it further includes:

[0063] Classifying the user behavior;

[0064] The obtaining of the user behavior score based on the user behavior and the second contract information includes:

[0065] Classifying and statistically calculating the total behavior value corresponding to the user behavior of each category according to the second contract information;

[0066] Obtaining the user behavior score based on the total behavior value of each category.

[0067] As described above, classifying the user behavior, calculating the total behavior value corresponding to the user behavior of each category, and then obtaining the user score based on the total behavior value of each category. Since the influence of different user behaviors on their evaluations, such as credit evaluations, objectively exists with different strengths, after classification, first calculate the total behavior value in each category, and then calculate the user score, which can achieve a more objective scoring for the customer when different-dimensional evaluations of the user can be obtained based on the total behavior value.

[0068] Please refer to Figure 2 , a user analysis terminal, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements the steps in the above-mentioned user analysis method.

[0069] The above-mentioned user analysis method and terminal of the present invention can be applied to scenarios where users need to be graded, classified, evaluated, or profiled, especially when there is a contractual relationship with the user. The following is illustrated through specific embodiments.

[0070] Please refer to Figures 1 - 3 , Embodiment 1 of the present invention is:

[0071] A user analysis method, including the steps:

[0072] S1. Obtaining the first contract information of all contracts corresponding to the user identifier, where the first contract information includes the contract status and the first signing time;

[0073] In an optional implementation manner, the first contract information is the information corresponding to an auto financing lease contract;

[0074] S2. Obtain the second contract information whose contract status meets the preset contract status from the first contract information, and obtain the user behavior corresponding to the second contract information;

[0075] In an alternative implementation, the user behavior includes the payment time and the most recent repayment time; among them, the payment time takes different calculation methods according to different types of transactions. If it is a normal transaction, it is the time when the money arrives; if it is a carry-forward transaction, it is the carry-forward time of the money; if it is a transfer order transaction, it is the time when the associated money arrives;

[0076] For example, if the customer needs to repay a monthly installment of 1000 yuan, and at this time 1000 yuan is deducted from the customer, this 1000 yuan is a normal transaction, and the arrival time of this 1000 yuan is the payment time; if the customer needs to repay a monthly installment of 1000 yuan, and at this time 1500 yuan is deducted from the customer, this 500 yuan is a carry-forward transaction, and the arrival time of this 500 yuan is the payment time; if the customer needs to repay a monthly installment of 1000 yuan, and at this time 1000 yuan is deducted from the customer, and the customer service converts this 1000 yuan into a premium payment, which is not used to offset the monthly installment, this 1000 yuan is a transfer order transaction; then another 1000 yuan needs to arrive, and the arrival time of the other 1000 yuan except the transfer order transaction is the payment time;

[0077] S3. Obtain the user behavior score based on the user behavior and the second contract information, including:

[0078] S31. Obtain the current scoring period, and mark the start node of the current scoring period as the beginning of the period;

[0079] In an alternative implementation, the beginning of the period is the 1st and 15th of each month; obtaining the current scoring period means obtaining the current date. If 1 ≤ current date < 15, the 1st is the beginning of the current scoring period; if 15 ≤ current date < the 1st of the next month, the 15th is the beginning of the current scoring period;

[0080] S32. Determine whether there is first contract information whose first signing time is earlier than the time at the beginning of the period. If so, execute S33; otherwise, execute S34;

[0081] S33. Mark the first user identifier corresponding to the first contract information as a stock user, and obtain all the first user behaviors corresponding to the first user identifier before the beginning of the period and the second contract information corresponding to the first user identifier; obtain the user behavior score based on the first user behavior and the second contract information according to the initial scoring algorithm;

[0082] In an alternative embodiment, S33 includes: obtaining a first score according to the initial score algorithm based on the first user behavior and the second contract information; obtaining all user behavior statistical data up to the initial time point and getting a second score according to the behavior score algorithm; adding the first score and the second score to obtain the user behavior score; it can be seen that the calculation method of the first score here is the same as that of the user behavior score in the previous paragraph;

[0083] Among them, obtaining the user behavior score according to the initial score algorithm based on the first user behavior and the second contract information includes:

[0084] S331. Calculating behavior index information based on the first user behavior and the second contract information, including:

[0085] (11) Calculating the overdue frequency ratio: obtaining the billing date and the corresponding payment time of the billing date, and recording the total number of overdue periods corresponding to the billing dates without corresponding payment times as the overdue frequency, then the performance period = the total number of periods occupied by the first contract information - the overdue frequency; the overdue frequency ratio = overdue frequency / performance period;

[0086] (12) Calculating the average overdue duration: obtaining the billing date and the corresponding payment time of the billing date, and judging whether the payment time is later than the billing date. If so, calculating the difference between the billing date and the payment time in a preset unit, and the sum of all differences (i.e., the differences corresponding to each scoring period) is the total overdue duration; if the preset unit is days, a certain user is overdue for 2 days and 16 hours and 1 day and 8 hours in the 3rd and 4th periods respectively, then omitting the hours according to the preset unit, the total overdue duration is 3 days; the average overdue duration = total overdue duration / overdue frequency, and the calculation method of the overdue frequency is the same as that in (11);

[0087] In an alternative embodiment, the maximum difference in each scoring period is selected as the difference corresponding to that period; for example, if a user has two overdue payments in a scoring period, the difference between one billing date and the payment time is 2 days, and the difference between the other billing date and the payment time is 4 days, then 4 days is taken as the difference corresponding to that scoring period; another example is that the billing date is May 20th. Assuming that a payment was made on May 22nd but not fully paid off, and another payment was made on May 25th and then it was fully paid off, then the overdue duration is 5 days, not 7 days;

[0088] (13) Calculating the average claim amount: obtaining the cumulative value of the claim amounts in the scoring periods to get the total claim amount; obtaining the total number of scoring periods with claim behaviors to get the total number of claim times; the average claim amount = total claim amount / total number of claim times;

[0089] (14) Calculate the risk occurrence probability: Obtain the total number of risk occurrences, and the calculation method is the same as that in (13); obtain the number of performance periods, and the calculation method is the same as that in (1); then the risk occurrence probability = the total number of risk occurrences / the number of performance periods;

[0090] (15) Calculate the total price of violation handling: Obtain the total handling price of all unhandled violations. The handling price is obtained from the preset violation workbench and is based on the photo handling price. The photo handling price represents the price charged for handling the violations corresponding to the violation types in the violation photos on behalf of the user;

[0091] (16) Calculate the average cost of violation handling: Obtain the number of performance periods, and the calculation method is the same as that in (11). The average cost of violation handling = the total price of violation handling / the number of performance periods;

[0092] In an optional implementation manner, only calculate the overdue-related data of the contracts with a monthly payment relationship;

[0093] S332. Obtain the preset index scoring table;

[0094] In an optional implementation manner, the index scoring tables are shown in Table 1 - Table 3:

[0095] Table 1

[0096]

[0097] Table 2

[0098]

[0099] Table 3

[0100]

[0101] S333. Obtain the score corresponding to the behavior index information according to the index scoring table;

[0102] S334. Obtain the user behavior score according to the initial score calculation algorithm and all the scores;

[0103] Existing users are those who have signed contracts before the beginning. That is, there is behavior information, namely historical behavior information, before the beginning time. Integrating this part of historical behavior information to obtain the user behavior score can reflect the user's behavior before this scoring period; it is also possible to calculate the behavior of existing users in two time periods divided by the beginning node, making full use of the implicit information of the time point closer to the present;

[0104] S34. Mark the first user identifier corresponding to the first contract information as an incremental user, and obtain all second user behaviors corresponding to the first user identifier and the second contract information corresponding to the first user identifier; calculate the user behavior statistical data corresponding to the second user behavior, and obtain the user behavior score according to the behavior scoring algorithm based on the user behavior statistical data and the second contract information.

[0105] Among them, calculating the user behavior statistical data corresponding to the second user behavior includes:

[0106] (21) Obtain the most recent repayment status: Obtain the second contract information corresponding to the most recent repayment time, and mark it as the repayment contract; obtain the billing cycle date of the repayment contract; obtain the most recent repayment status based on the most recent repayment time and the billing cycle date. If the most recent repayment time is earlier than the billing cycle date, it is considered on-time repayment; otherwise, it is considered late repayment.

[0107] (22) Obtain the number of overdue times, and the calculation method is the same as that in (11).

[0108] (23) Obtain the overdue duration, and the calculation method is the same as the total overdue duration in (12). For incremental users, there is only one scoring cycle from the beginning to the current time.

[0109] (24) Obtain the number of claim occurrences, and the calculation method is the same as the total number of claim occurrences in (13).

[0110] (25) Obtain the claim amount, and the calculation method is the same as the total claim amount in (13).

[0111] (26) Obtain the most recent claim status: If the number of claim occurrences in the current scoring cycle is 0, the most recent claim status is no recent claim.

[0112] (27) Obtain the duration of violations: Obtain the existence duration of all unprocessed violations, and take the value in the preset unit as the duration of violations; if there are multiple unprocessed violations, take the existence duration of the unprocessed violation with the longest existence duration as the duration of violations. For example, if the preset unit is days, and the user has 1 violation that has existed for 3 days and 12 hours, it is only counted as 3 days.

[0113] (28) Obtain the violation handling price; obtain the handling prices corresponding to all violations.

[0114] (29) Obtain the most recent violation status: If there are no violations on the current date or all violations have been processed, the most recent violation status is no recent violation.

[0115] The behavior scoring algorithm includes a behavior scoring table, as shown in Tables 4 - 7:

[0116] Table 4

[0117]

[0118] Table 5

[0119]

[0120] Table 6

[0121]

[0122] Table 7

[0123]

[0124] Incremental users are users who register after the beginning of the period. Since the behaviors after the beginning of the period are closer to the current time, the scoring items will be changed accordingly. At this time, a scoring method different from the scoring method before the beginning of the period is used to make the scoring of users more objective.

[0125] In an optional implementation, after obtaining the user behavior score, it also includes: matching a collection template according to the user behavior score, the collection template is a speech template, which helps to recover the money in a more targeted manner; and subsequent operational means (points redemption, etc.).

[0126] Embodiment 2 of the present invention is:

[0127] A user analysis method, which is different from the first embodiment in that before step S2, it also includes:

[0128] S011. Obtain all possible existing contract states;

[0129] S012. Determine whether the possible contract status indicates the termination of the contract. If not, set the possible contract status as the preset contract status;

[0130] In an optional implementation, the preset contract is converted and stored in a scoring contract list library, and the scoring contract list library can be directly called to read the preset contract status to determine which of the first contract information needs to be included in the scoring;

[0131] Table 8 gives an example of a scoring contract list library:

[0132] Table 8

[0133]

[0134] In an optional implementation, the preset contract status includes a normal status, a legal intervention status, a vehicle collection status, an accident status, and a vehicle transfer status;

[0135] Embodiment 3 of the present invention is:

[0136] A user analysis method, which is different from the first embodiment or the second embodiment in that:

[0137] Before S3, it further includes:

[0138] S021. Classify the user behaviors;

[0139] In an optional implementation manner, the user behavior categories include violations, insurance claims, and overdue; it can be further subdivided into violation amount, number of violations, insurance claim amount, number of insurance claims, overdue duration, and number of overdue times;

[0140] Then S3 includes:

[0141] S301. Classify and statistically calculate the total behavior value corresponding to the user behaviors of each category according to the second contract information;

[0142] S302. Obtain the user behavior score based on the total behavior value of each category;

[0143] For example, if the user has 2 violations, the first time is 200 yuan and the second time is 300 yuan, then first summarize the total behavior value as the violation amount of 500 yuan, and then obtain the user behavior score according to 500 yuan. Specifically, according to the correspondence between the total behavior value and the index scoring table, the user behavior score is obtained.

[0144] Please refer to Figure 2 , the second embodiment of the present invention is:

[0145] A user analysis terminal 1 includes a processor 2, a memory 3, and a computer program stored on the memory 3 and executable on the processor 2. When the processor 2 executes the computer program, it implements each step in the first embodiment, the second embodiment, or the third embodiment.

[0146] In summary, the present invention provides a user analysis method and terminal. By automatically reading the information in the contract as a reference for scoring the obtained user behaviors, it realizes the automatic calculation of user behavior scores, forms a standard user grading system, improves work efficiency, and realizes refined operation management; the user grading scheme based on calculating scores from user behaviors is particularly suitable for providing refined operation means in the after-market of auto finance leasing services and improving the conversion of graded users.

[0147] The above are only the embodiments of the present invention, and do not limit the patent scope of the present invention accordingly. All equivalent transformations made using the content of the specification and drawings of the present invention, or directly or indirectly applied in related technical fields, are equally included in the patent protection scope of the present invention.

Claims

1. A user analysis method, characterized in that, Including the steps: Obtain the first contract information of all contracts corresponding to the user identifier, where the first contract information includes the contract status and the first signing time; Obtain the second contract information whose contract status meets the preset contract status from the first contract information, and obtain the user behavior corresponding to the second contract information; Obtain the user behavior score based on the user behavior and the second contract information, including: Obtain the current scoring period, and mark the start node of the current scoring period as the beginning of the period; Judge whether there is first contract information whose first signing time is earlier than the beginning-of-period time. If so, mark the first user identifier corresponding to the first contract information as a stock user, and obtain all the first user behaviors corresponding to the first user identifier before the beginning-of-period time and the second contract information corresponding to the first user identifier; obtain the first score according to the beginning-of-period scoring algorithm based on the first user behavior and the second contract information; obtain the second score according to the behavior scoring algorithm based on all the user behavior statistical data up to the beginning-of-period time point; accumulate the first score and the second score to obtain the user behavior score; Otherwise, mark the first user identifier corresponding to the first contract information as an incremental user, and obtain all the second user behaviors corresponding to the first user identifier and the second contract information corresponding to the first user identifier; calculate the user behavior statistical data corresponding to the second user behavior, and obtain the user behavior score according to the behavior scoring algorithm based on the user behavior statistical data and the second contract information; Before obtaining the second contract information whose contract status meets the preset contract status from the first contract information, it includes: Obtain all possible contract statuses that exist; Judge whether the possible contract status indicates the end of the contract. If not, set the possible contract status as the preset contract status.

2. The user analysis method according to claim 1, characterized in that, The preset contract status includes the normal status, the legal affairs intervention status, the vehicle repossession status, the accident status, and the transfer to vehicle service status.

3. The user analysis method according to claim 1, characterized in that, Obtaining the user behavior score according to the beginning-of-period scoring algorithm and the behavior scoring algorithm based on the first user behavior and the second contract information includes: Calculate the behavior index information based on the first user behavior and the second contract information; Obtain the preset index scoring table; Obtain the score corresponding to the behavior index information according to the index scoring table; Obtain the user behavior score according to the beginning-of-period scoring algorithm and all the scores.

4. The user analysis method according to claim 3, characterized in that, The first contract information includes the billing cycle date; The behavior index information includes the overdue frequency ratio; The first user behavior includes the payment time; Calculating the behavior index information based on the first user behavior and the second contract information includes: Obtain the billing cycle date and the payment time corresponding to the billing cycle date, and record the total number of overdue periods corresponding to the billing cycle dates without corresponding payment times as the number of overdue times. Then, the number of performance periods = the total number of periods occupied by the first contract information - the number of overdue periods; Overdue frequency ratio = number of overdue times / number of performance periods.

5. The user analysis method according to claim 1, characterized in that, The user behavior statistical data includes the most recent repayment time; Obtaining the user behavior score according to the behavior scoring algorithm based on the user behavior statistical data and the second contract information includes: Obtain the second contract information corresponding to the most recent repayment time and mark it as the repayment contract; Obtain the billing cycle date of the repayment contract; Obtain the user behavior score according to the most recent repayment time, the billing cycle date, and the behavior scoring algorithm.

6. The user analysis method according to claim 1, characterized in that, Before obtaining the user behavior score based on the user behavior and the second contract information, it further includes: Classify the user behavior; Obtaining the user behavior score based on the user behavior and the second contract information includes: Classify and statistically calculate the total behavior value corresponding to the user behavior of each category according to the second contract information; Obtain the user behavior score based on the total behavior value of each category.

7. A user analysis terminal, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps in a user analysis method according to any one of claims 1-6.

Images