Instant lease signing payment management method and system

Abstract

The invention discloses an instant lease signing payment management method and system. The method comprises the following steps: collecting signing payment data corresponding to user data; data cleaning and clustering are carried out on the signed payment data to obtain a plurality of step-by-step operation data sets, and each step-by-step operation data set corresponds to the operation condition of one signed payment step with a time period label; determining user feature information according to user data corresponding to the signing payment data in the multiple step-by-step operation data sets; and formulating a system optimization scheme based on the feature information of each user under the labels at different time periods and the corresponding signing payment step operation condition. According to the invention, through collection and analysis of the contract signing payment data of the user, refined process analysis and data clustering, identification and optimization of bottlenecks in each link, and combination of labels in different time periods and user characteristics, an optimization scheme is formulated, the operation efficiency of the platform is improved, more personalized services can be provided, and the user experience is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of online rental platform optimization, and particularly to an instant rental signing and payment management method and system. Background Art

[0002] An online instant rental platform refers to a business platform that provides instant rental services through an Internet platform. Users can use a mobile device to log in to the online rental platform when a need occurs and select the goods or services to be rented. Online rental effectively reduces the economic burden on users, provides great convenience and flexibility for users, improves the utilization rate of items, and solves the problem of resource waste. It is the product of the continuous development of modern society and the upgrading of consumption patterns. With the increase in rental platforms, users have put forward higher requirements for the service quality of instant online rental platforms.

[0003] Currently, many instant rental platforms ensure the smooth signing and payment through static optimization methods. Usually, the signing and payment are set as fixed processes. When receiving feedback with poor user experience, the method of replacing the overall template is often adopted to take effect on all users. This method has high input costs but poor effects, resulting in the platform resource allocation system being unable to make dynamic adjustments according to actual user needs and actual resource changes. There may be situations of overload or resource waste, lacking flexibility, unable to deeply explore the details of user behavior, unable to perform targeted fine-grained optimization on the rental platform, and unable to meet the usage needs of customers.

[0004] Therefore, it is necessary to propose an instant rental signing and payment management method and system that can comprehensively analyze the resource usage of each link in online rental signing and payment and user characteristic data, improve the signing and payment problems in the instant rental platform through fine-grained analysis, optimize the allocation of system resources, and enhance the operation efficiency and user experience of the instant rental platform. Summary of the Invention

[0005] In view of this, the present invention provides an instant rental signing and payment management method and system to solve the technical problem that the existing online rental platform cannot comprehensively analyze the resource usage of each link in online rental signing and payment and user characteristic data, resulting in the inability to effectively improve the operation efficiency and user experience of the instant rental platform.

[0006] To achieve the above technical objectives, the present invention adopts the following technical solutions:

[0007] On the one hand, the present invention provides an instant rental signing and payment management method, including:

[0008] Collecting signing and payment data corresponding to user data;

[0009] Clean and cluster the signed payment data to obtain multiple step-by-step operation data sets, where each step-by-step operation data set corresponds to the operation status of a signed payment step with a time period label;

[0010] Determine user characteristic information based on the user data corresponding to the signed payment data in multiple step-by-step operation data sets;

[0011] Formulate a system optimization plan based on the user characteristic information and the corresponding signed payment step operation status under different time period labels.

[0012] Further, the user data includes user basic information, user credit score, user labels, and user evaluation data;

[0013] The signed payment data includes terminal device data, signing process data, number of authentication times, payment process data, process timestamp, and system load data.

[0014] Further, cleaning and clustering the signed payment data to obtain multiple step-by-step process data sets, including:

[0015] Remove duplicates and complete the data for the signed payment data;

[0016] Extract the time period characteristics of the signed payment process according to the process timestamp in the signed payment data, and determine the time period label of the signed payment based on the time period characteristics;

[0017] Perform vector operations on the signing process data, number of authentication times, payment process data, and system load data under the time period label to obtain multiple signed payment feature vectors;

[0018] Cluster the multiple signed payment feature vectors under each time period label to obtain multiple clusters of vector sets;

[0019] Select the corresponding step-by-step operation data set for each cluster of vector sets according to the preset criteria to obtain the category of the signed payment step operation status.

[0020] Further, the user basic information includes user personal information and rental history records; the user credit score includes the user credit score obtained from a third-party platform; the user labels include member users, ordinary users, and risk users; the user evaluation data includes the number of connections with the customer service, connection time, and service satisfaction score;

[0021] The terminal device data includes the terminal type interface, system version, and APP version; the signing process data includes the signing step sequence, e - contract access data, signing step - by - step time consumption, and the number of signing failures; the payment process data includes the payment amount, payment method, payment step sequence, payment step - by - step time consumption, and the number of payment failures; the number of authentication times includes the statistical values of the verification times in the signing process and the payment process; the system load data includes the total number of users and network resource occupancy data.

[0022] Further, cluster multiple signing - payment feature vectors under each time - period label to obtain a multi - cluster vector set, including:

[0023] Use the reference feature vector corresponding to the preset signing - payment step operation situation category as the initial centroid;

[0024] Calculate the distance from each signing - payment feature vector to each initial centroid, and assign the data point to the cluster with the closest distance;

[0025] Recalculate the updated initial centroid of the signing - payment feature vectors in each cluster vector set, and re - assign multiple signing - payment feature vectors based on the updated initial centroid until the preset maximum number of iterations is reached to obtain a multi - cluster vector set.

[0026] Further, calculate the distance from each signing - payment feature vector to each initial centroid, which is expressed by the formula:

[0027] D(x,c)=β·M(x,c)+(1 - β)·(1 - cos sim (x,c))

[0028] where x is the current signing - payment feature vector, c is the reference feature vector corresponding to the initial centroid, D(x,c) represents the similarity distance between the current signing - payment feature vector and the initial centroid, M(x,c) represents the Euclidean distance between the feature vector and the initial centroid, cos sim (x,c) represents the cosine distance between the current signing - payment feature vector and the initial centroid, and β represents the weight coefficient.

[0029] Further, recalculate the updated initial centroid of the signing - payment feature vectors in each cluster vector set, including:

[0030] Update the initial centroid through the following formula:

[0031]

[0032] where c (k) represents the k - th clustering centroid, S k represents the set of all data points in the step - by - step operation dataset, |S k | represents the set S kThe number of data points, x represents the current signed payment feature vector, m t represents the weight vector, and ⊙ represents the element-wise multiplication of vectors.

[0033] Furthermore, according to the user data corresponding to the signed payment data in multiple step-by-step process data sets, user feature information is determined, including:

[0034] Taking each numerical data of the user data as a vector element, an initial feature vector corresponding to each user data is established;

[0035] Performing a normalization operation on each initial feature vector to obtain the user feature information corresponding to each user data.

[0036] Furthermore, according to the running conditions of the signed payment steps corresponding to the user feature information under different time period labels, a system optimization plan is formulated, including:

[0037] According to the running conditions of the signed payment steps corresponding to the user feature information under different time period labels, adjusting the resource allocation parameters for the payment step settings, which is expressed by the formula:

[0038]

[0039] where, L t is the system resource allocation amount at the current moment, L t-1 is the system resource allocation amount at the previous time step, Y i (t) is the resource consumption of the i-th user at the current moment, R t is the total system resources that can be allocated currently, η 1 、η 2 and η 3 respectively represent weight parameters used to balance the influence of different factors on resource allocation.

[0040] On the other hand, the present invention also provides an instant rental signing payment management system, including:

[0041] A data collection module for collecting signed payment data corresponding to user data;

[0042] A data analysis module for performing data cleaning and clustering on the signed payment data to obtain multiple step-by-step running data sets, and each step-by-step running data set corresponds to the running condition of a signed payment step with a time period label;

[0043] A feature determination module for determining user feature information according to the user data corresponding to the signed payment data in multiple step-by-step running data sets;

[0044] An optimization module, configured to formulate a system optimization plan based on the user characteristic information and the operation status of the corresponding signing and payment steps under different time period tags.

[0045] Compared with the prior art, the advantages provided by the present invention are as follows:

[0046] (1) By performing data cleaning and clustering on the signing and payment data, the operation status of the signing and payment steps with different time period tags is obtained. Through refined process analysis and data clustering, bottlenecks in each link are identified and optimized, so as to improve the probability of successful signing and payment of users.

[0047] (2) According to the signing and payment data and the corresponding user data in multiple step-by-step operation data sets, user characteristic information is determined, which can perform a more fine-grained analysis of user characteristics and the operation status of signing steps, and improve the user experience; based on the user characteristic information and the corresponding operation status categories of signing and payment steps under different time period tags, a targeted system optimization plan is formulated, which can intelligently allocate resources according to the system load and user behavior data in different time periods, optimize the waiting time of specific links in the signing and payment process for users with different characteristics, avoid resource waste, and improve system stability.

[0048] In summary, the present invention can, through precise process optimization, intelligent resource management and personalized user analysis, perform targeted fine-grained analysis and improvement on the signing and payment problems in the instant rental platform, optimize the allocation of system resources, significantly improve system efficiency and user satisfaction, and enhance the overall operation efficiency and user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a schematic flowchart of the instant rental signing and payment management method provided by the present invention;

[0050] Figure 2 It is a schematic flowchart of the vector cluster clustering processing provided by the present invention;

[0051] Figure 3 It is a schematic structural diagram of the instant rental signing and payment management system provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0052] The following will specifically describe the preferred embodiments of the present invention in conjunction with the drawings. The drawings form a part of this application and are used together with the embodiments of the present invention to explain the principles of the present invention, rather than to limit the scope of the present invention.

[0053] Please refer to Figure 1 , this embodiment provides an instant rental signing and payment management method, including:

[0054] Step S101: Collect the signing and payment data corresponding to the user data;

[0055] Step S102: Clean and cluster the signed payment data to obtain multiple step-by-step operation datasets, where each step-by-step operation dataset corresponds to the operation situation of a signed payment step with a time period label;

[0056] Step S103: Determine user characteristic information according to the user data corresponding to the signed payment data in multiple step-by-step operation datasets;

[0057] Step S104: Develop a system optimization plan based on the operation situations of the signed payment steps corresponding to each user characteristic information under different time period labels.

[0058] The method of this embodiment, by collecting and analyzing the signed payment data of users, performing refined process analysis and data clustering to identify and optimize the bottlenecks in each link, combining different time period labels and user characteristics to develop an optimization plan, not only improves the operation efficiency of the system, but also can provide more personalized services, enhances the user experience, helps the online rental system to gradually improve during use, better adapt to the changes in the market and user needs, maintain long-term competitiveness, and promote the development and progress of the instant rental business.

[0059] As a preferred embodiment, in step S101, the user data includes user basic information, user credit score, user tags, and user evaluation data;

[0060] The signed payment data includes terminal device data, signing process data, number of identity verification times, payment process data, process timestamp, and system load data.

[0061] Furthermore, the user basic information includes user personal information and rental history records; the user credit score includes the user credit score obtained from a third-party platform; the user tags include member users, ordinary users, and risk users; the user evaluation data includes the number of times of connecting with the customer service, connection time, and service satisfaction score;

[0062] The terminal device data includes terminal type interface, system version, and APP version; the signing process data includes signing step sequence, e-contract access data, signing step-by-step time consumption, and number of signing failures; the payment process data includes payment amount, payment method, payment step sequence, payment step-by-step time consumption, and number of payment failures; the number of identity verification times includes the statistical value of the verification times in the signing process and the payment process; the system load data includes the total number of users and network resource occupancy data.

[0063] As a specific embodiment, taking the user data of User A as a specific example, the user data is as follows: Basic information of User A: age 28, gender and address; Rental history: has rented a mobile phone and a laptop on this platform; Based on the user's rental history records and evaluation data, the user's preferences can be analyzed, so as to optimize the display page of the platform and provide more targeted product or service recommendations for customers.

[0064] Credit score: 750 points (obtained from a third-party platform); Tags of User A: ordinary user; Combining the user's credit score, rental history records and user tags, the user can be risk-assessed.

[0065] Evaluation data: has communicated with the customer service 3 times, with an average service score of 4.2 (out of 5); By analyzing the number of times the user has connected to the customer service, the time, and the service satisfaction score, the platform can identify and solve the user's pain points and improve the overall customer satisfaction.

[0066] As a preferred embodiment, in step S102, data cleaning and clustering are performed on the signing and payment data to obtain multiple step-by-step process data sets, including:

[0067] Perform data deduplication and data supplementation on the signing and payment data;

[0068] Extract the time period characteristics of the signing and payment process according to the process timestamp in the signing and payment data, and determine the time period label of the signing and payment based on the time period characteristics;

[0069] Perform vector operations on the signing process data, number of identity verifications, payment process data, and system load data under the time period label to obtain multiple signing and payment feature vectors;

[0070] Cluster the multiple signing and payment feature vectors under each time period label to obtain multiple clusters of vector sets;

[0071] Select the corresponding step-by-step operation data set for each cluster of vector sets according to the preset criteria to obtain the category of the signing and payment step operation situation.

[0072] In the above processing process, extracting the time period characteristics of the signing and payment process according to the process timestamp in the signing and payment data is to classify and analyze the user needs during holidays and non-holiday periods, as well as different time periods of each day, compare the signing and payment process data of different users after removing the influence of time periods horizontally, and compare the signing and payment experiences of the same user under the influence of different time periods vertically. Analyzing from different dimensions is more timely and targeted, and can identify potential problems or characteristics of the process under different time periods.

[0073] Converting the signing payment process, the number of identity verifications, payment process data, and system load data into feature vectors enables more refined calculations and modeling. After vectorization, the model can more easily identify similarities, correlations, and patterns among the data. After clustering, the system can identify typical signing payment processes, and then analyze optimization solutions for each type of operating situation separately, avoiding overly generalized improvement measures.

[0074] As a preferred embodiment, cluster multiple signing payment feature vectors under each time period label to obtain multiple cluster vector sets, including:

[0075] Use the reference feature vector corresponding to the preset signing payment step operating situation category as the initial centroid;

[0076] Calculate the distance from each signing payment feature vector to each initial centroid, and assign the data point to the closest cluster;

[0077] Recalculate the updated initial centroid of the signing payment feature vectors in each cluster vector set, and reassign multiple signing payment feature vectors based on the updated initial centroid until the preset maximum number of iterations is reached to obtain multiple cluster vector sets.

[0078] As Figure 2 shown, Figure 2 Figure 1 shows a schematic flow diagram of clustering signing payment feature vectors. It should be noted that in some embodiments, when some feature vectors cannot be clustered and matched to existing clusters, new vector clusters can be created according to the feature drift amount and the total drift, to achieve dynamic clustering with self-evolution ability. System operators' analysis of the new vector clusters helps to further analyze the actual situation or guide the generation of a more reasonable preset range during operation.

[0079] As a preferred embodiment, calculate the distance from each signing payment feature vector to each initial centroid, which is expressed by the formula:

[0080] D(x,c) = β·M(x,c) + (1 - β)·(1 - cos sim (x,c))

[0081] where x is the current signing payment feature vector, c is the reference feature vector corresponding to the initial centroid, D(x,c) represents the similarity distance between the current signing payment feature vector and the initial centroid, M(x,c) represents the Euclidean distance between the feature vector and the initial centroid, cos sim (x,c) represents the cosine distance between the current signing payment feature vector and the initial centroid, and β represents the weight coefficient.

[0082] By combining the Euclidean distance and cosine similarity and adjusting the influence of both through the weight parameter β, it is more suitable for the usage scenario of accurately classifying by evaluating the similarity between the current signed payment feature vector and the centroid. In conventional clustering operations, the Euclidean distance is usually used to measure the actual physical distance between data points, but the signed payment feature vector contains various different types of representation data.

[0083] Specifically, in practical applications, the operation steps are usually converted into a Markov transition probability matrix, the payment amount, the number of failure times, and the number of verification steps are converted into numerical data, and the payment methods, etc. are discretized and represented using one-hot encoding. Therefore, in order to adapt to the needs of different data types and clustering tasks, the cosine distance is also introduced when calculating the similarity distance. After combining the two methods, the clustering results can not only reflect the numerical differences (such as the payment amount, the time-consuming of signing step by step, etc.), but also identify the similarity of behavior patterns (such as the payment step sequence, the signing process). This multi-level similarity measurement makes the clustering results more comprehensive, detailed, and can more accurately identify potential user behavior patterns and system load rules.

[0084] As a preferred embodiment, recalculating the updated initial centroid of the signed payment feature vector in each cluster vector set includes:

[0085] Updating the initial centroid through the following formula:

[0086]

[0087] where c (k) represents the k-th clustering centroid, S k represents the set of all data points in the step-by-step operation dataset, |S k | represents the number of data points in the set S k , x represents the current signed payment feature vector, m t represents the weight vector, and ⊙ represents the element-wise multiplication of vectors.

[0088] As a preferred embodiment, in step S103, according to the user data corresponding to the signed payment data in multiple step-by-step process datasets, determining the user feature information includes:

[0089] Taking each numerical data of the user data as a vector element to establish an initial feature vector corresponding to each user data;

[0090] Performing a normalization operation on each initial feature vector to obtain the user feature information corresponding to each user data. By analyzing the signed payment data in multiple datasets, the feature information reflecting the behavior characteristics of the corresponding users is extracted.

[0091] As a preferred embodiment, in step S104, according to the operation conditions of the signing and payment steps corresponding to each user feature information under different time period tags, a system optimization plan is formulated, including:

[0092] According to the operation conditions of the signing and payment steps corresponding to each user feature information under different time period tags, adjust the resource allocation parameters of the payment step settings, which can be expressed by the formula:

[0093]

[0094] Where L t is the system resource allocation amount at the current moment, and L t-1 is the system resource allocation amount at the previous time step. Y i (t) is the resource consumption of the i-th user at the current moment, and R t is the total system resources that can be allocated currently. η 1 , η 2 and η 3 respectively represent weight parameters used to balance the influence of different factors on resource allocation.

[0095] By combining the system resource allocation amounts before and after different time steps and the resource consumption of users at the current moment, the resource growth requirements between steps can be accurately judged, so as to intelligently allocate resources, avoid over-allocation or under-allocation, and effectively improve the utilization efficiency of platform resources.

[0096] As a specific embodiment, during the weekend evening peak period, it is detected through the clustering result that the failure rate in the payment process is higher than the set threshold. Through user feature analysis, it is found that the main affected group is non-member users of iOS users + old version SDK. Then the problem can be traced, and it is located and analyzed that the compatibility between the old version SDK and the Apple Pay interface is poor, and the bandwidth resources allocated to non-member users are insufficient. At this time, an emergency treatment plan can be formulated: push the forced update of the SDK and automatically increase the bandwidth resources to switch to the backup payment channel adapted to the new version interface. Subsequently, during the overall optimization, a feature version management mechanism and an online A / B test can also be cooperated to achieve closed-loop optimization of the platform and improve the system's adaptive optimization ability.

[0097] As Figure 3 shown, the embodiment of the present invention also provides an instant rental signing and payment management system 300, including:

[0098] A data collection module 301 for collecting signing and payment data corresponding to user data;

[0099] The data analysis module 302 is used to clean and cluster the signed payment data to obtain multiple step-by-step operation data sets, and each step-by-step operation data set corresponds to the operation situation of a signed payment step with a time period label.

[0100] The feature determination module 303 is used to determine user feature information according to the user data corresponding to the signed payment data in multiple step-by-step operation data sets.

[0101] The optimization module 304 is used to formulate a system optimization plan based on the user feature information and the corresponding signed payment step operation situation under different time period labels.

[0102] The instant rental signing and payment management method and system provided by the present invention collect user data and obtain the corresponding signed payment data, providing a strong foundation for comprehensively analyzing user behaviors and needs; by cleaning and clustering the collected signed payment data and dividing the data into multiple step-by-step operation data sets, it can better understand each link in the signing and payment process and can conduct targeted analysis on the signing and payment behaviors in different time periods according to the time period label; by analyzing multiple step-by-step operation data sets, it can accurately identify user feature information, and based on the behavior characteristics shown by different users in the signing and payment process, it can provide more personalized services for different user groups, which helps to improve the user experience; according to the user feature information and the signed payment step operation situation under different time period labels, a targeted optimization plan is formulated, enabling the system to adjust the working process or service content of the system according to different time periods and user characteristics, so as to better meet user needs, improve the efficiency of rental signing and payment, and enhance the flexibility and response speed of the system.

[0103] The above is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present invention should be covered by the protection scope of the present invention.

Claims

1. A method for managing instant leasing contract payment, characterized in that: include; Collect contract payment data corresponding to user data; Clean and cluster the contract payment data to obtain multiple step-by-step operation data sets, each of which corresponds to the operation status of a contract payment step with a time period label; Determine user characteristic information according to user data corresponding to the contract payment data in the plurality of step-by-step operation data sets; Develop a system optimization plan based on the user characteristic information under different time period labels and the corresponding signing and payment step operation status.

2. The instant rental contract payment management method according to claim 1 is characterized in that: The user data includes user basic information, user credit score, user tags and user evaluation data; The contract payment data includes terminal device data, contract process data, identity authentication times, payment process data, process timestamp and system load data.

3. The instant rental contract payment management method according to claim 2 is characterized in that: The contract payment data is cleaned and clustered to obtain multiple step-by-step process data sets, including: De-duplicate and complete the contract payment data; Extract the time period characteristics of the contract payment process according to the process timestamp in the contract payment data, and determine the time period label of the contract payment based on the time period characteristics; Performing vector operations on the contract signing process data, the number of identity authentications, the payment process data, and the system load data under the time period label to obtain a plurality of contract signing and payment feature vectors; Cluster multiple contract payment feature vectors under each time period label to obtain multiple cluster vector sets; According to the preset standard, the step-by-step operation data set corresponding to each cluster vector set is selected to obtain the category of the operation status of the contract signing and payment steps.

4. The instant rental contract payment management method according to claim 3 is characterized in that: User basic information includes user personal information and rental history records; user credit score includes user credit score obtained from third-party platforms; user tags include member users, ordinary users and risk users; user evaluation data includes the number of connections with customer service, connection time, and service satisfaction scores; Terminal device data includes terminal type interface, system version and APP version; contract signing process data includes contract signing step sequence, electronic contract access data, contract signing step-by-step time and number of contract signing failures; Payment process data includes payment amount, payment method, payment step sequence, payment step-by-step duration and number of payment failures; identity authentication times include statistical values ​​of verification times for the signing process and payment process; system load data includes total number of users and network resource usage data.

5. The instant rental contract payment management method according to claim 3 is characterized in that: Cluster multiple contract payment feature vectors under each time period label to obtain multiple cluster vector sets, including: The reference feature vector corresponding to the preset category of the operation status of the signing and payment step is used as the initial centroid; Calculate the distance between each contract payment feature vector and each initial centroid, and assign the data point to the cluster with the closest distance; The updated initial centroid of the contract payment feature vector in each cluster vector set is recalculated, and multiple contract payment feature vectors are reallocated based on the updated initial centroid until a preset maximum number of iterations is reached to obtain multiple cluster vector sets.

6. The instant rental contract payment management method according to claim 5 is characterized in that: Calculate the distance from each contract payment feature vector to each initial centroid, expressed as: D(x,c)=β·M(x,c)+(1-β)·(1-cos sim (x,c)) Where x is the current contract payment feature vector, c is the reference feature vector corresponding to the initial centroid, D(x,c) represents the similarity distance between the current contract payment feature vector and the initial centroid, M(x,c) represents the Euclidean distance between the feature vector and the initial centroid, cos sim (x,c) represents the cosine distance between the current contracted payment feature vector and the initial centroid, and β represents the weight coefficient.

7. The instant rental contract payment management method according to claim 5 is characterized in that: Recalculate the updated initial centroid of the contract payment feature vector in each cluster vector set, including: The initial centroid is updated using the following formula: Among them, c (k) Indicates k Cluster centroids, S k represents the set of all data points in the step-by-step running dataset, |S k | represents the set S k The number of data points in , x represents the current contract payment feature vector, m t represents the weight vector, and ⊙ represents the element-by-element multiplication of the vector.

8. The instant rental contract payment management method according to claim 1, characterized in that: Determine user feature information based on user data corresponding to the contract payment data in multiple step-by-step process data sets, including: Take each numerical data of the user data as a vector element and establish an initial feature vector corresponding to each user data; Each initial feature vector is normalized to obtain user feature information corresponding to each user data.

9. The instant rental contract payment management method according to claim 1, characterized in that: According to the operation status of the signing and payment steps corresponding to the characteristic information of each user under different time period labels, formulate a system optimization plan, including: According to the operation status of the contract payment step corresponding to each user's characteristic information under different time period labels, adjust the payment step to set the resource allocation parameters, which can be expressed as: Among them, L t is the current system resource allocation, L t-1 is the system resource allocation at the previous time step, Y i (t) is the resource consumption of the i-th user at the current moment, R t is the total system resource currently available for allocation, and η1, η2, and η3 represent weight parameters, which are used to balance the impact of different factors on resource allocation.

10. An instant leasing contract payment management system, characterized in that: include: A data collection module is used to collect contract payment data corresponding to user data; The data analysis module is used to clean and cluster the contract payment data to obtain multiple step-by-step operation data sets, each of which corresponds to the operation status of a contract payment step with a time period label; A feature determination module, used to determine user feature information according to user data corresponding to the contract payment data in the multiple step-by-step operation data sets; The optimization module is used to formulate system optimization plans based on the characteristic information of each user under different time period labels and the corresponding signing and payment step operation status.

Images