Information rearrangement method, device, electronic device and medium based on financial system

Abstract

The embodiments of the present application provide a method, device, electronic device, and medium for rearranging information based on a financial system, belonging to the field of financial technology. The method includes: obtaining all financial products in the financial system; for each financial product, collecting data generated by the user's interaction with the financial product within a preset time period to obtain sample data; generating triple data based on the user's user information, sample data, and financial products; performing a reverse sort operation on the triple data based on the interaction frequency to obtain a sample sequence; setting multiple cutoff values ​​based on the sample sequence and the interaction frequency, and setting a suppression function based on the cutoff values; calculating the score of the triple data based on the suppression function to obtain a scoring result; rearranging the financial products based on the scoring result to obtain a target display sequence; and recommending to the user target financial products corresponding to the order of the target display sequence. The embodiments of the present application can improve the accuracy of product recommendations.

Description

Technical Field

[0001] The present application relates to the field of financial technology, and in particular to a method, device, electronic device and medium for rearranging information based on a financial system. Background Art

[0002] In the financial field, insurance recommendation systems play a key role in the financial field. They not only provide users with personalized insurance product recommendations, but also help insurance companies improve customer stickiness and market competitiveness.

[0003] However, traditional systems often rely on static user information and simple product matching logic, ignoring the dynamic nature of user behavior and the complexity of insurance needs. Furthermore, with the frequent updates and changes to insurance products, the insurance products recommended by traditional recommendation systems may not necessarily meet the user's current needs. For example, a user may have just purchased a specific type of insurance, but the system may still recommend similar or duplicate products without considering the user's existing insurance coverage. This overlap can not only waste users' time but also lead to frustration and dissatisfaction with the system. If the system consistently recommends products with high click-through rates, it can easily create a Matthew effect, where popular products receive more exposure while those that are not clicked on or have low click-through rates are gradually marginalized, reducing their exposure opportunities and resulting in a lack of personalized and accurate recommendations. Summary of the Invention

[0004] The main purpose of the embodiments of the present application is to propose an information rearrangement method, device, electronic device and medium based on a financial system, which can improve the accuracy of product recommendations and avoid the Matthew effect.

[0005] To achieve the above objectives, a first aspect of an embodiment of the present application provides an information rearrangement method based on a financial system, the method comprising:

[0006] Access all financial products in the financial system;

[0007] For each of the financial products, data generated by the user's interaction with the financial product within a preset time period is collected to obtain sample data, wherein the sample data includes the interaction frequency;

[0008] generating triplet data according to the user information of the user, the sample data, and the financial product;

[0009] Performing an inverted sort operation on the triple data based on the interaction frequency to obtain a sample sequence;

[0010] Setting a plurality of cutoff values ​​according to the sample sequence and the interaction frequency, and setting a suppression function according to the cutoff values;

[0011] Calculating the score of the triplet data based on the suppression function to obtain a scoring result;

[0012] Rearranging the financial products according to the scoring results to obtain a target display sequence;

[0013] Target financial products corresponding to the order of the target display sequence are recommended to the user.

[0014] In some embodiments, for each of the financial products, collecting data generated by the user's interaction with the financial product within a preset time period to obtain sample data includes:

[0015] For each financial product, record the user's click behavior, collection behavior, and purchase behavior of the financial product within a preset time period;

[0016] Collecting a first frequency of the user's click behavior, a second frequency of the collection behavior, and a third frequency of the purchase behavior;

[0017] Obtaining an interaction frequency according to the first frequency, the second frequency, and the third frequency;

[0018] Obtaining interactive behavior according to the click behavior, the collection behavior, and the purchase behavior;

[0019] Sample data is obtained according to the interaction frequency and the interaction behavior.

[0020] In some embodiments, setting a plurality of cutoff values ​​according to the sample sequence and the interaction frequency includes:

[0021] Dividing the sample sequence into intervals based on a preset Pareto analysis method to obtain a first interval, a second interval, and a third interval, wherein the intersection of any two of the first interval, the second interval, and the third interval is empty;

[0022] The minimum value of the second interval is set as the first delimiting value, and the maximum value of the second interval is set as the second delimiting value.

[0023] In some embodiments, setting a suppression function according to the demarcation value includes:

[0024] Setting a first piecewise function corresponding to the first interval, a second piecewise function corresponding to the second interval, and a third piecewise function corresponding to the third interval;

[0025] The segmentation value of the first segmentation function is set according to the first boundary value, the segmentation value of the second segmentation function is set according to the first boundary value and the second boundary value, and the segmentation value of the third segmentation function is set according to the second boundary value to obtain a suppression function.

[0026] In some embodiments, dividing the sample sequence into intervals based on a preset Pareto analysis method to obtain a first interval, a second interval, and a third interval includes:

[0027] Based on a preset Pareto analysis method, determining an interval in the sample sequence that is within a first preset ratio as a first interval, determining an interval in the sample sequence that is within a second preset ratio as a second interval, and determining an interval in the sample sequence that is within a third preset ratio as a third interval;

[0028] or,

[0029] Setting a first ratio threshold and a second ratio threshold according to a preset Pareto analysis method;

[0030] The sample sequence is segmented according to the first ratio threshold and the second ratio threshold, and a sequence portion smaller than the first ratio threshold is determined as a first interval, a sequence portion greater than or equal to the first ratio threshold and smaller than the second ratio threshold is determined as a second interval, and a sequence portion greater than or equal to the second ratio threshold is determined as a third interval.

[0031] In some embodiments, the triple data includes multiple groups of data to be scored; and the scoring of the triple data based on the scoring function to obtain a scoring result includes:

[0032] For each set of scoring data in the triplet data, inputting the scoring data into a preset deep learning model to perform refined ranking score calculation, and outputting the refined ranking score corresponding to the scoring data;

[0033] Dividing the triple data according to the first dividing value, the second dividing value, and the interaction frequency to obtain a first data set, a second data set, and a third data set;

[0034] Determining a first refined ranking score set corresponding to the first data set, a second refined ranking score set corresponding to the second data set, and a third refined ranking score set corresponding to the third data set;

[0035] Input the first data set into the first piecewise function to output a first rearranged score set, input the second data set into the second piecewise function to output a second rearranged score set, and input the third data set into the third piecewise function to output a third rearranged score set;

[0036] determining a first score based on the first refined-ranked score set and the first rearranged score set, determining a second score based on the second refined-ranked score set and the second rearranged score set, and determining a third score based on the third refined-ranked score set and the third rearranged score set;

[0037] A scoring result is obtained according to the first score, the second score, and the third score.

[0038] In some embodiments, dividing the triple data according to the first cutoff value, the second cutoff value, and the interaction frequency to obtain a first data set, a second data set, and a third data set includes:

[0039] comparing the interaction frequency with the first cutoff value and the second cutoff value;

[0040] taking the data in the triple data whose interaction frequency is less than the first cutoff value as the first data set;

[0041] taking the data in the triplet data, wherein the interaction frequency is greater than or equal to the first cutoff value and less than or equal to the second cutoff value, as a second data set;

[0042] The data in the triplet data whose interaction frequency is greater than the second cutoff value is used as the third data set.

[0043] To achieve the above-mentioned purpose, a second aspect of an embodiment of the present application provides an information rearrangement device based on a financial system, the device comprising:

[0044] Product acquisition module, used to obtain all financial products in the financial system;

[0045] a data collection module configured to collect, for each of the financial products, data generated by user interactions with the financial product within a preset time period to obtain sample data, wherein the sample data includes interaction frequency;

[0046] A data generation module, configured to generate triplet data based on the user information of the user, the sample data, and the financial product;

[0047] a sequence inversion module, configured to perform an inversion operation on the triple data based on the interaction frequency to obtain a sample sequence;

[0048] A cutoff value setting module, configured to set a plurality of cutoff values ​​according to the sample sequence and the interaction frequency, and to set a suppression function according to the cutoff values;

[0049] A score calculation module, configured to calculate the score of the triple data based on the suppression function to obtain a score result;

[0050] a target generation module, configured to rearrange the financial products according to the scoring results to obtain a target display sequence;

[0051] The product recommendation module is used to recommend target financial products corresponding to the order of the target display sequence to the user.

[0052] To achieve the above-mentioned purpose, the third aspect of an embodiment of the present application proposes an electronic device, an electronic device comprising a memory and a processor, the memory storing a computer program, and the processor implementing the information rearrangement method based on the financial system as described in the first aspect when executing the computer program.

[0053] To achieve the above-mentioned purpose, the fourth aspect of an embodiment of the present application proposes a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the information rearrangement method based on the financial system as described in the first aspect.

[0054] The information rearrangement method, device, electronic device and storage medium based on the financial system proposed in this application first obtain all financial products in the financial system. For each financial product, the data generated by the user's interaction with the financial product within a preset time period is collected to obtain sample data, so that the user's recent interests and demand changes can be accurately reflected through the sample data. Then, triple data is generated based on the user's user information, sample data and financial products to achieve summary statistics of the data, which is convenient for subsequent analysis of the user's interaction frequency with different financial products. Afterwards, the triple data is reversed based on the interaction frequency, and the triple data can be sorted in combination with the user's active behavior and interaction frequency to obtain a sample sequence. Multiple cutoff values ​​are set according to the sample sequence and interaction frequency, so as to identify multiple interaction frequency values ​​that affect the sorting result, which is convenient for subsequent improvement of decision quality, and a suppression function is set according to the cutoff value to target different positions of the sample sequence. Different degrees of suppression are given to the financial products with different interactions, thereby increasing the degree of suppression on high-frequency financial products and reducing their probability of appearing at the head, while reducing the degree of suppression on financial products with low interaction frequency and increasing the recommendation exposure opportunities of financial products with low interaction frequency. The score of the triple data is calculated based on the suppression function to suppress financial products with higher initial scores and increase the scores of financial products with lower scores to obtain the scoring results, thereby avoiding the Matthew effect and increasing the exposure opportunities of financial products at the tail of the sample sequence. The financial products are rearranged according to the scoring results to obtain the target display sequence to meet the user's needs for diversity and freshness. Finally, the target financial products corresponding to the order of the target display sequence are recommended to the user, which can guide users to discover new products while providing personalized financial product recommendations that are more in line with user needs, increase user participation in new products, and further improve the accuracy and effectiveness of recommendations. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 This is a flowchart of a financial system-based information rearrangement method provided in an embodiment of the present application;

[0056] Figure 2 yes Figure 1 Flowchart of step S102 in FIG.

[0057] Figure 3 A flow chart of a method for setting multiple cutoff values ​​according to sample sequence and interaction frequency provided in an embodiment of the present application;

[0058] Figure 4 A flow chart of a method for setting a suppression function according to a threshold value provided in an embodiment of the present application;

[0059] Figure 5 yes Figure 3 Flowchart of step S301 in FIG.

[0060] Figure 6 yes Figure 1 Flowchart of step S106 in FIG.

[0061] Figure 7 yes Figure 6 Flowchart of step S602 in FIG.

[0062] Figure 8 This is a schematic diagram of the structure of an information rearrangement device based on a financial system provided in an embodiment of the present application;

[0063] Figure 9 This is a schematic diagram of the hardware structure of the electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0064] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0065] It should be noted that although the device schematics illustrate functional module divisions and the flowcharts illustrate logical sequences, in certain circumstances, the steps shown or described may be performed in a sequence that differs from the module divisions in the device or the sequence in the flowcharts. The terms "first," "second," and so on, in the specification, claims, and drawings, are used to distinguish similar items and are not necessarily used to describe a specific sequence or precedence.

[0066] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application pertains. The terms used herein are for the purpose of describing the embodiments of this application only and are not intended to limit this application.

[0067] First, let’s analyze some of the terms used in this application:

[0068] Natural Language Processing (NLP): NLP uses computers to process, understand, and apply human languages ​​(such as Chinese and English). NLP is a branch of artificial intelligence and an interdisciplinary subject between computer science and linguistics. It is often referred to as computational linguistics. Natural language processing includes grammatical analysis, semantic analysis, and text understanding. Natural language processing is commonly used in technical fields such as machine translation, handwritten and printed character recognition, speech recognition and text-to-speech conversion, information intent recognition, information extraction and filtering, text classification and clustering, public opinion analysis, and opinion mining. It involves data mining related to language processing, machine learning, knowledge acquisition, knowledge engineering, artificial intelligence research, and linguistic research related to language computing.

[0069] Pareto Analysis: Often used to identify and focus on the most important issues or causes, based on the Pareto Principle (80 / 20 Principle), which states that 80% of the results typically come from 20% of the causes.

[0070] The information rearrangement method and device based on the financial system, electronic device and storage medium provided in the embodiments of the present application can improve the accuracy of product recommendations and avoid the Matthew effect.

[0071] The following examples are used to illustrate the method. First, the information rearrangement method based on the financial system in the embodiment of the present application is described.

[0072] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Artificial Intelligence (AI) is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use knowledge to achieve optimal results.

[0073] Fundamental AI technologies generally include sensors, specialized AI chips, cloud computing, distributed storage, big data processing, operating / interaction systems, and mechatronics. AI software technologies primarily encompass computer vision, robotics, biometrics, module management for online customer service systems, natural language processing, and machine learning / deep learning.

[0074] The information re-arrangement method based on the financial system provided in the embodiment of the present application relates to the field of financial technology. The information re-arrangement method based on the financial system provided in the embodiment of the present application can be applied to the terminal, can also be applied to the server side, and can also be software running in the terminal or the server side. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc.; the server side can be configured as an independent physical server, or can be configured as a server cluster or distributed system composed of multiple physical servers, and can also be configured as a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application that implements the information re-arrangement method based on the financial system, etc., but is not limited to the above forms.

[0075] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld devices or portable devices, tablet devices, multi-processor systems, microprocessor-based systems, set-top boxes, programmable consumer electronic devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data information types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0076] It should be noted that in each specific embodiment of the present application, when it comes to the need to perform relevant processing based on data related to the user's identity or characteristics, such as user information, user behavior data, user historical data, and user location information, the user's permission or consent will be obtained first, and the collection, use, and processing of such data will comply with the relevant laws, regulations, and standards of the relevant countries and regions. In addition, when the embodiment of the present application needs to obtain the user's sensitive personal information, the user's separate permission or consent will be obtained through a pop-up window or by jumping to a confirmation page. After clearly obtaining the user's separate permission or consent, the necessary user-related data for the normal operation of the embodiment of the present application will be obtained.

[0077] In the financial field, insurance recommendation systems play a key role in the financial field. They not only provide users with personalized insurance product recommendations, but also help insurance companies improve customer stickiness and market competitiveness.

[0078] However, traditional systems often rely on static user information and simple product matching logic, ignoring the dynamic nature of user behavior and the complexity of insurance needs. Furthermore, with the frequent updates and changes to insurance products, the insurance products recommended by traditional recommendation systems may not necessarily meet the user's current needs. For example, a user may have just purchased a specific type of insurance, but the system may still recommend similar or duplicate products without considering the user's existing insurance coverage. This overlap can not only waste users' time but also lead to frustration and dissatisfaction with the system. If the system consistently recommends products with high click-through rates, it can easily create a Matthew effect, where popular products receive more exposure while those that are not clicked on or have low click-through rates are gradually marginalized, reducing their exposure opportunities and resulting in a lack of personalized and accurate recommendations.

[0079] In order to solve the above problems, this embodiment provides an information rearrangement method, device, electronic device and storage medium based on the financial system. First, all financial products in the financial system are obtained. For each financial product, data generated by the user's interaction with the financial product within a preset time period is collected to obtain sample data, so that the user's recent interests and demand changes can be accurately reflected through the sample data. Then, triple data is generated based on the user's user information, sample data and financial products to achieve summary statistics of the data, which is convenient for subsequent analysis of the user's interaction frequency with different financial products. Thereafter, the triple data is reversed based on the interaction frequency, and the triple data can be sorted in combination with the user's active behavior and interaction frequency to obtain a sample sequence. Multiple cutoff values ​​are set according to the sample sequence and interaction frequency, so as to identify multiple interaction frequency values ​​that affect the sorting result, which is convenient for subsequent improvement of decision quality. A suppression function is set according to the cutoff value to target the sample. Financial products at different positions in the sequence are given different degrees of suppression, thereby increasing the degree of suppression on high-frequency financial products and reducing their probability of appearing at the head, while reducing the degree of suppression on financial products with low interaction frequency and increasing the recommendation exposure opportunities for financial products with low interaction frequency. The triple data is scored based on the suppression function to suppress financial products with higher initial scores and increase the scores of financial products with lower scores to obtain scoring results, thereby avoiding the Matthew effect and increasing the exposure opportunities of financial products at the tail of the sample sequence. The financial products are rearranged according to the scoring results to obtain the target display sequence to meet the user's needs for diversity and freshness. Finally, the target financial products corresponding to the order of the target display sequence are recommended to the user, which can guide users to discover new products while providing personalized financial product recommendations that are more in line with user needs, increase user participation in new products, and further improve the accuracy and effectiveness of recommendations.

[0080] The following is a detailed description with reference to the accompanying drawings.

[0081] Figure 1 This is an optional flowchart of the information rearrangement method based on the financial system provided in the embodiment of the present application. Figure 1 The method may include but is not limited to steps S101 to S108.

[0082] Step S101: Acquire all financial products in the financial system.

[0083] In step S101 of some embodiments, all financial products in the financial system are obtained, such as pension insurance products, automobile insurance products, financial management insurance products, etc., to facilitate subsequent analysis of their interaction frequencies.

[0084] Step S102 : For each financial product, data generated by the user's interaction with the financial product within a preset time period is collected to obtain sample data, wherein the sample data includes the interaction frequency.

[0085] In some embodiments, in step S102, since users may click on or purchase certain financial products while browsing, data generated by user interactions with the financial product over a preset period of time is collected for each financial product. By analyzing the user's interaction behavior and frequency table over a period of time, the relevance and accuracy of the financial system's recommendations can be improved. Sample data is obtained, enabling real-time updates of user interaction behavior and frequency. With the continuous updating of user behavior data, the recommendation system can quickly adapt to market changes and changes in user needs. This real-time or near-real-time data processing capability makes the recommendation system more flexible and adaptable.

[0086] It is understandable that the interaction frequency in the embodiment of the present application can be the number of user clicks, the number of purchases, etc. The interaction frequency will be specifically explained below and will not be repeated here in the embodiment of the present application.

[0087] Step S103: generating triplet data according to the user information, sample data and financial products.

[0088] In step S103 of some embodiments, the user information includes the user identifier, the sample data includes the interaction frequency, and the financial product includes the financial product identifier. The embodiment of the present application generates triplet data based on the user identifier, the financial product identifier, and the interaction frequency to facilitate subsequent analysis of financial products with different interaction frequencies.

[0089] It can be understood that the triplet data in the embodiment of the present application can be expressed as (uid, iid, count), where uid is the user identifier, iid is the financial product identifier, and count is the interaction frequency.

[0090] Step S104: performing an inverted sort operation on the triple data based on the interaction frequency to obtain a sample sequence.

[0091] In step S104 of some embodiments, the triple data is inverted based on the interaction frequency. Specifically, the interaction behavior is used as the keyword to establish an inverted list, and the list of users and financial products under each interaction behavior is recorded to obtain a sample sequence, thereby improving the accuracy of recommendations and user satisfaction.

[0092] Step S105 , setting multiple cutoff values ​​according to the sample sequence and the interaction frequency, and setting a suppression function according to the cutoff values.

[0093] In step S105 of some embodiments, multiple cutoff values ​​are set according to the sample sequence and the interaction frequency, so that the activity of user behavior can be distinguished more finely, the key factors that have the greatest impact on the results can be identified more accurately, and a suppression function is set according to the cutoff value to achieve differentiated suppression of financial products with different interaction frequencies, thereby meeting users' needs for diversity and freshness.

[0094] Step S106: Score the triplet data based on the suppression function to obtain a scoring result.

[0095] In step S106 of some embodiments, a score is calculated for the triple data based on a suppression function to obtain a scoring result, thereby achieving a re-ranking and scoring of financial products with different interaction frequencies. This can increase the degree of suppression for financial products with high interaction frequencies and reduce the degree of suppression for financial products with low interaction frequencies, thereby increasing the recommendation exposure opportunities for financial products with low interaction frequencies, thereby meeting user needs, facilitating users to discover new products, and improving user experience.

[0096] Step S107: Rearrange the financial products according to the scoring results to obtain a target display sequence.

[0097] In step S107 of some embodiments, the financial products are rearranged according to the scoring results to reduce the probability of appearance of the top financial products and increase the probability of appearance of the bottom financial products, thereby obtaining a target display sequence.

[0098] Step S108: recommending target financial products corresponding to the order of the target display sequence to the user.

[0099] In step S108 of some embodiments, target financial products corresponding to the order of the target display sequence are recommended to the user, satisfying the user's demand for diversity and novelty, guiding the user to discover new products, and improving user participation.

[0100] See also Figure 2 In some embodiments, step S102 may also include but is not limited to steps S201 to S205.

[0101] Step S201: For each financial product, record the user's click behavior, collection behavior, and purchase behavior of the financial product within a preset time period.

[0102] Step S202 , collecting the first frequency of the user's click behavior, the second frequency of the collection behavior, and the third frequency of the purchase behavior.

[0103] Step S203: obtaining an interaction frequency according to the first frequency, the second frequency, and the third frequency.

[0104] Step S204: Obtain interactive behaviors based on click behaviors, collection behaviors, and purchase behaviors.

[0105] Step S205: obtaining sample data according to the interaction frequency and interaction behavior.

[0106] In some embodiments, in steps S201 to S205, during the process of collecting data generated by a user's interaction with a financial product within a preset time period, the user may browse, add to favorites, or purchase the financial product. To capture the user's preferences, embodiments of the present application record, for each financial product in the financial system, the user's click behavior, add to favorites, and purchase behavior of the financial product within the preset time period. These behaviors reflect the user's preference for the financial product, enabling accurate understanding of the user's preferences and purchase intentions. The frequencies of the user's various interaction behaviors are then recorded. Specifically, a first frequency of the user's click behavior, a second frequency of the add to favorites, and a third frequency of the purchase behavior are collected. This allows the number of occurrences of various interaction behaviors to be determined based on the different frequencies, further accurately reflecting changes in the user's interests and needs and accurately capturing the user's real-time needs. Subsequently, the user's interaction frequency with the financial product is obtained based on the first, second, and third frequencies, and the interaction behavior is then obtained based on the click behavior, add to favorites, and purchase behavior. By analyzing the user's interaction behavior and frequency table over a period of time, the relevance and accuracy of the financial system's recommendations can be improved. Finally, sample data is obtained based on the interaction frequency and interaction behavior, enabling real-time updates of the user's interaction behavior and frequency. As user behavior data is continuously updated, recommendation systems can quickly adapt to market changes and changes in user needs. This real-time or near-real-time data processing capability makes recommendation systems more flexible and adaptable.

[0107] It should be noted that the preset duration in the embodiment of the present application can be set according to user needs, for example, recording the user's interactive behavior within one month, recording the user's interactive behavior within one week, recording the user's interactive behavior within two weeks, etc. The embodiment of the present application does not make specific restrictions.

[0108] In some embodiments, the embodiments of the present application can also record the user's browsing behavior of financial products within a preset time period, and record the browsing duration of the user's browsing behavior, and use click behavior, collection behavior, purchase behavior and browsing behavior as interactive behavior. At the same time, sample data is obtained based on the interaction frequency, interactive behavior and browsing duration to further improve the relevance and accuracy of financial system recommendations.

[0109] Specifically, the first frequency in the embodiment of the present application is the number of click behaviors, for example, the number of times a user clicks on a financial product or link; the second frequency is the number of collection behaviors, for example, the number of times a user adds a product or content to a collection list; the third frequency is the number of purchase behaviors, for example, the number of times a user actually purchases a product.

[0110] See also Figure 3 , Figure 3 A flowchart of a method for setting multiple cutoff values ​​according to a sample sequence and an interaction frequency provided in an embodiment of the present application, wherein the method includes but is not limited to steps S301 to S302.

[0111] Step S301 : dividing the sample sequence into intervals based on a preset Pareto analysis method to obtain a first interval, a second interval, and a third interval.

[0112] It should be noted that the intersection of any two of the first interval, the second interval, and the third interval is empty.

[0113] Step S302: Set the minimum value of the second interval as the first delimiting value, and set the maximum value of the second interval as the second delimiting value.

[0114] In steps S301 to S302 of some embodiments, in the process of setting multiple dividing values ​​according to the sample sequence and the interaction frequency, first, the sample sequence is divided into intervals based on the preset Pareto analysis method to obtain a first interval, a second interval and a third interval, wherein the first interval, the second interval and the third interval do not intersect with each other, thereby achieving effective division of the intervals, being able to more finely distinguish the activity of user behavior, and facilitating the subsequent formulation of different personalized suppression strategies for different intervals. Finally, the minimum value of the second interval is set as the first dividing value, and the maximum value of the second interval is set as the second dividing value. By setting different dividing values, the key factors that have the greatest impact on the results can be more accurately identified.

[0115] It can be understood that, taking the Pareto analysis method as an example to illustrate the division of the sample sequence into 20%, 40% and 40% ratios, according to the 28th law, the first 20% is regarded as the high-active interaction number interval (first interval), followed by 40% as the medium-active interaction number interval (second interval), and the last 40% as the low-active interaction number interval (third interval), and then determining the interval values ​​of the first interval, the second interval and the third interval. Specifically, in the embodiment of the present application, the minimum value of the first interval is 0 and the maximum value is 3, the minimum value of the second interval is 4 and the maximum value is 10, and the minimum value of the third interval is 11 and the maximum value is 15. At this time, the minimum value of the second interval is set to the first dividing value, that is, 4 is set to the first dividing value, and the maximum value of the second interval is set to the second dividing value, that is, 10 is set to the second dividing value, so as to facilitate the subsequent setting of different suppression strategies for different intervals.

[0116] See also Figure 4 , Figure 4 A flow chart of a method for setting a suppression function according to a threshold value provided in an embodiment of the present application, the method includes but is not limited to steps S401 to S402.

[0117] Step S401 : setting a first piecewise function corresponding to the first interval, a second piecewise function corresponding to the second interval, and a third piecewise function corresponding to the third interval.

[0118] Step S402, setting the segmentation value of the first segmentation function according to the first threshold value, setting the segmentation value of the second segmentation function according to the first threshold value and the second threshold value, and setting the segmentation value of the third segmentation function according to the second threshold value to obtain a suppression function.

[0119] In steps S401 to S402 of some embodiments, in the process of setting the suppression function according to the threshold value, since the financial system has a higher probability of recommending financial products with high interaction frequency and a lower probability of recommending financial products with low interaction frequency, the embodiment of the present application will set a first piecewise function corresponding to the first interval, a second piecewise function corresponding to the second interval, and a third piecewise function corresponding to the third interval, and adopt differentiated suppression strategies for different intervals, and adopt piecewise functions to increase the degree of suppression on financial products with high interaction frequency and reduce the degree of suppression on financial products with low interaction frequency. Afterwards, the segmentation value of the first piecewise function is set according to the first threshold value, the segmentation value of the second piecewise function is set according to the first and second threshold values, and the segmentation value of the third piecewise function is set according to the second threshold value to obtain the suppression function, thereby achieving differentiated suppression of financial products with different interaction frequencies and meeting users' needs for diversity and freshness.

[0120] Specifically, since the financial products that have been purchased more times are likely to be the most popular top financial products, if you only focus on the top few products, it is easy to cause the Matthew effect. In order to avoid the Matthew effect, the embodiment of the present application adopts differentiated financial product recommendation suppression strategies for different parts of the sample sequence, adopts a piecewise function, and designs a suppression function according to the interaction frequency of the financial product. Among them, the first piecewise function of the embodiment of the present application is expressed as f(c)=log(c+2), the second piecewise function is expressed as f(c)=c, and the third piecewise function is expressed as f(c)=c 0.75 +5, thus obtaining the suppression function, and setting the segmentation value of the segmentation function according to different threshold values. Taking the first threshold value of 4 and the second threshold value of 10 as an example, the suppression function is expressed as follows:

[0121]

[0122] See also Figure 5 In some embodiments, step S301 may also include but is not limited to steps S501 to S503.

[0123] Step S501 : Based on a preset Pareto analysis method, an interval in a sample sequence with a first preset ratio is determined as a first interval, an interval in a sample sequence with a second preset ratio is determined as a second interval, and an interval in a sample sequence with a third preset ratio is determined as a third interval.

[0124] In step S501 of some embodiments, in the process of dividing the sample sequence into intervals based on the preset Pareto analysis method, the embodiments of the present application can determine the interval in the sample sequence that is in a first preset proportion as the first interval, determine the interval in the sample sequence that is in a second preset proportion as the second interval, and determine the interval in the sample sequence that is in a third preset proportion as the third interval based on the preset Pareto analysis method, thereby achieving effective division of the intervals, thereby being able to identify the minority of factors that have the greatest impact on the results.

[0125] Specifically, the first preset ratio, the second preset ratio and the third preset ratio can be set according to the needs of the user. The embodiment of the present application is illustrated by taking the first preset ratio of 20%, the second preset ratio of 40%, and the third preset ratio of 40% as an example. The Pareto analysis method is used to divide the top 20% of the sample sequence into the first interval, that is, the high active interaction number interval, and then the 40% interval in the sample sequence is divided into the second interval, that is, the medium active interaction number interval, and finally the 40% interval in the sample sequence is divided into the third interval, that is, the high active interaction number interval.

[0126] Step S502: setting a first ratio threshold and a second ratio threshold according to a preset Pareto analysis method.

[0127] Step S503 : The sample sequence is segmented according to the first ratio threshold and the second ratio threshold. The portion of the sequence smaller than the first ratio threshold is determined as a first interval, the portion of the sequence greater than or equal to the first ratio threshold and smaller than the second ratio threshold is determined as a second interval, and the portion of the sequence greater than or equal to the second ratio threshold is determined as a third interval.

[0128] In steps S502 to S503 of some embodiments, in the process of dividing the sample sequence into intervals based on the preset Pareto analysis method, the embodiments of the present application can also set a first ratio threshold and a second ratio threshold according to the preset Pareto analysis method, and then divide the sample sequence into intervals according to the first ratio threshold and the second ratio threshold, and determine the sequence portion that is less than the first ratio preset as the first interval, the sequence portion that is greater than or equal to the first ratio threshold and less than the second ratio threshold as the second interval, and the sequence portion that is greater than or equal to the second ratio threshold as the third interval, so as to more finely distinguish the activity of user behavior, and facilitate the subsequent formulation of different personalized suppression strategies for different intervals.

[0129] Specifically, in an embodiment of the present application, two thresholds can be set according to the Pareto analysis method. Taking the first ratio threshold of 20% and the second ratio threshold of 60% as an example, the sample sequence is cut according to the first ratio threshold and the second ratio threshold, and the sequence portion of the sample sequence that is less than 20% is determined as the first interval, the sequence portion of the sample sequence that is greater than or equal to 20% and less than 60% is determined as the second interval, and the sequence portion of the sample sequence that is greater than or equal to 60% is determined as the third interval.

[0130] See also Figure 6 In some embodiments, step S106 may also include but is not limited to steps S601 to S606.

[0131] It should be noted that the triplet data includes multiple groups of data to be scored.

[0132] Step S601: For each set of scoring data in the triplet data, the scoring data is input into a preset deep learning model to perform refined ranking score calculation, and the refined ranking score corresponding to the scoring data is output.

[0133] Step S602 : dividing the triplet data according to the first cutoff value, the second cutoff value, and the interaction frequency to obtain a first data set, a second data set, and a third data set.

[0134] Step S603 : Determine a first refined ranking score set corresponding to the first data set, a second refined ranking score set corresponding to the second data set, and a third refined ranking score set corresponding to the third data set.

[0135] Step S604: input the first data set into the first piecewise function to output a first rearranged score set, input the second data set into the second piecewise function to output a second rearranged score set, and input the third data set into the third piecewise function to output a third rearranged score set.

[0136] Step S605 , determining a first score based on the first refined score set and the first rearranged score set, determining a second score based on the second refined score set and the second rearranged score set, and determining a third score based on the third refined score set and the third rearranged score set.

[0137] Step S606: Obtain a scoring result according to the first score, the second score, and the third score.

[0138] In some embodiments, in steps S601 to S606, during the process of calculating scores for triple data based on a suppression function, the triple data includes multiple sets of data to be scored. For each set of scoring data in the triple data, the scoring data is first input into a preset deep learning model for fine-ranking score calculation. Specifically, the deep learning model extracts product features of the financial product corresponding to the scoring data and user features corresponding to the user, and then constructs a feature matrix based on the product features and user features. The trained deep learning model then scores the financial products corresponding to the feature matrix and outputs a fine-ranking score corresponding to the scoring data. By integrating multi-dimensional information about users and products, the accuracy of recommendations and user satisfaction can be improved. Subsequently, the triple data is divided according to a first cutoff value, a second cutoff value, and interaction frequency, and data with different interaction frequencies are divided into a first data set, a second data set, and a third data set. By classifying the data with different interaction frequencies, accurate analysis of the interaction frequencies of different financial products is achieved, thereby enabling suppression processing for data sets with different interaction frequencies, thereby achieving differentiated suppression processing for different interaction frequencies. Afterwards, a deep learning model is used to determine the first refined ranking score set corresponding to the first data set, the second refined ranking score set corresponding to the second data set, and the third refined ranking score set corresponding to the third data set, thereby accurately determining the products that the user can accept and the products that are most likely to be recommended to the user. Different data sets are then segmented using a suppression function. Specifically, the first data set is input into the first segmentation function, which outputs the first rearranged score set, the second data set is input into the second segmentation function, which outputs the second rearranged score set, and the third data set is input into the third segmentation function, which outputs the third rearranged score set. This increases the degree of suppression of financial products with high interaction frequencies, reduces their probability of appearing at the top, and reduces the degree of suppression of financial products with low interaction frequencies, thereby increasing the recommendation exposure opportunities for financial products with low interaction frequencies. Afterwards, the scores of the suppressed financial products are recalculated. Specifically, based on the first refined ranking score set and the first rearranged score set, the scores of the financial products with low interaction frequencies are recalculated. The score set determines a first score. In an embodiment of the present application, the first score is obtained by dividing the first precise ranking score by the first re-ranked score. Similarly, the second score is obtained by dividing the score in the second precise ranking score set by the score in the second re-ranked score set, and the third score is obtained by dividing the score in the third precise ranking score set by the score in the third re-ranked score set. Finally, a scoring result is obtained based on the first score, the second score, and the third score, thereby realizing re-ranking and scoring of financial products with different interaction frequencies, reducing the probability of appearance of top financial products, and increasing the probability of appearance of tail financial products, thereby meeting user needs, facilitating users to discover new products, and improving user experience.

[0139] It should be noted that the deep learning models in the embodiments of the present application include but are not limited to logistic regression models, random forest models, gradient boosting function models, etc., and the embodiments of the present application do not impose specific restrictions.

[0140] See also Figure 7 In some embodiments, step S602 may also include but is not limited to steps S701 to S704.

[0141] Step S701: Compare the interaction frequency with a first threshold value and a second threshold value.

[0142] Step S702: taking the data in the triple data whose interaction frequency is less than the first cutoff value as the first data set.

[0143] Step S703: taking the data in the triplet data whose interaction frequency is greater than or equal to the first threshold value and less than or equal to the second threshold value as the second data set.

[0144] Step S704: taking the data in the triplet data whose interaction frequency is greater than the second cutoff value as the third data set.

[0145] In steps S701 to S704 of some embodiments, in the process of dividing the triplet data according to the first threshold value, the second threshold value and the interaction frequency, the embodiment of the present application compares the interaction frequency with the first threshold value and the second threshold value, and then takes the data in the triplet data with an interaction frequency less than the first threshold value as the first data set, takes the data in the triplet data with an interaction frequency greater than or equal to the first threshold value and less than or equal to the second threshold value as the second data set, and takes the data in the triplet data with an interaction frequency greater than the second threshold value as the third data set, thereby realizing the division of data sets with different interaction frequencies, so that suppression processing can be performed on data sets with different interaction frequencies, and differentiated suppression processing of different interaction frequencies can be realized.

[0146] Specifically, taking the first dividing value as 4, the second dividing value as 10, and the interaction frequency as the historical number of interactions as an example, the data with a historical number of interactions less than 4 in the triplet data is taken as the data in the first data set, the data with a historical number of interactions greater than or equal to 4 and less than or equal to 10 in the triplet data is taken as the data in the second data set, and the data with a historical number of interactions greater than 10 in the triplet data is taken as the data in the third data set, thereby completing the division of data sets with different interaction frequencies, which is convenient for subsequent personalized processing of different data sets.

[0147] See also Figure 8 The embodiment of the present application further provides an information rearrangement device based on a financial system, the device comprising:

[0148] Product acquisition module 801, used to acquire all financial products in the financial system;

[0149] The data collection module 802 is configured to collect data generated by user interactions with each financial product within a preset time period to obtain sample data, wherein the sample data includes interaction frequency;

[0150] The data generation module 803 is used to generate triplet data based on the user information, sample data and financial products;

[0151] A sequence inversion module 804 is used to perform an inversion operation on the triple data based on the interaction frequency to obtain a sample sequence;

[0152] A cutoff value setting module 805 is used to set multiple cutoff values ​​according to the sample sequence and the interaction frequency, and to set a suppression function according to the cutoff values;

[0153] The score calculation module 806 is used to calculate the score of the triple data based on the suppression function to obtain a score result;

[0154] A target generation module 807 is used to rearrange the financial products according to the scoring results to obtain a target display sequence;

[0155] The product recommendation module 808 is configured to recommend target financial products corresponding to the order of the target display sequence to the user.

[0156] The specific implementation of the financial system-based information rearrangement device is substantially the same as the specific embodiment of the financial system-based information rearrangement method described above, and will not be described in detail here.

[0157] An embodiment of the present application further provides an electronic device comprising: a memory, a processor, a program stored in the memory and executable on the processor, and a data bus for enabling communication between the processor and the memory. When the program is executed by the processor, the aforementioned method for rearranging information based on a financial system is implemented. The electronic device may be any intelligent terminal, including a tablet computer and an in-vehicle computer.

[0158] See also Figure 9 , Figure 9 The hardware structure of an electronic device according to another embodiment is shown. The electronic device includes:

[0159] The processor 901 may be implemented as a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is configured to execute relevant programs to implement the technical solutions provided in the embodiments of the present application.

[0160] The memory 902 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 902 can store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 902 and is called by the processor 901 to execute the information reordering method based on the financial system in the embodiments of this application.

[0161] Input / output interface 903, used to implement information input and output;

[0162] Communication interface 904, used to implement communication interaction between this device and other devices, which can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WiFi, Bluetooth, etc.);

[0163] Bus 905 , which transmits information between various components of the device (e.g., processor 901 , memory 902 , input / output interface 903 , and communication interface 904 );

[0164] The processor 901 , the memory 902 , the input / output interface 903 and the communication interface 904 are connected to each other in communication within the device via a bus 905 .

[0165] An embodiment of the present application further provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-mentioned information rearrangement method based on the financial system.

[0166] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0167] The information rearrangement method, device, electronic device and storage medium based on the financial system provided by the embodiments of the present application first obtain all financial products in the financial system, and for each financial product, collect data generated by the user's interaction with the financial product within a preset time period to obtain sample data, so that the user's recent interests and demand changes can be accurately reflected through the sample data, and then generate triple data based on the user's user information, sample data and financial products to achieve summary statistics of the data, which is convenient for subsequent analysis of the user's interaction frequency with different financial products. Thereafter, the triple data is reversed based on the interaction frequency, and the triple data can be sorted in combination with the user's active behavior and interaction frequency to obtain a sample sequence, and multiple cutoff values ​​are set according to the sample sequence and interaction frequency to identify multiple interaction frequency values ​​that affect the sorting result, which is convenient for subsequent improvement of decision quality, and a suppression function is set according to the cutoff value to target the sample sequence that is not Financial products in the same position are given different degrees of suppression, thereby increasing the degree of suppression for high-frequency financial products and reducing their probability of appearing at the head, while reducing the degree of suppression for financial products with low interaction frequency and increasing the recommendation exposure opportunities for financial products with low interaction frequency. The triple data is scored based on the suppression function to suppress financial products with higher initial scores and increase the scores of financial products with lower scores to obtain scoring results, thereby avoiding the Matthew effect and increasing the exposure opportunities of financial products at the tail of the sample sequence. The financial products are rearranged according to the scoring results to obtain the target display sequence to meet the user's needs for diversity and freshness. Finally, the target financial products corresponding to the order of the target display sequence are recommended to the user, which can guide users to discover new products while providing personalized financial product recommendations that are more in line with user needs, increase user participation in new products, and further improve the accuracy and effectiveness of recommendations.

[0168] The embodiments described in the embodiments of this application are intended to more clearly illustrate the technical solutions of the embodiments of this application and do not constitute a limitation on the technical solutions provided by the embodiments of this application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0169] It will be understood by those skilled in the art that Figure 1-9 The technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than those shown in the figures, or a combination of certain steps, or different steps.

[0170] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, i.e., they may be located in one place or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of this embodiment.

[0171] Those skilled in the art will appreciate that all or some of the steps in the methods, systems, and functional modules / units in the devices disclosed above may be implemented as software, firmware, hardware, or appropriate combinations thereof.

[0172] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0173] It should be understood that in this application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following items" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0174] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the above-mentioned units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0175] The units described above as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0176] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0177] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes multiple instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application. The aforementioned storage medium includes: various media that can store programs, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0178] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but are not intended to limit the scope of the present invention. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and essence of the present invention should be within the scope of the present invention.

Claims

1. A method for rearranging information based on a financial system, characterized in that: The method comprises: Obtain all financial products in the financial system; for each financial product, collect data generated by user interactions with the financial product within a preset time period to obtain sample data, wherein the sample data includes interaction frequency; generating triplet data according to the user information of the user, the sample data, and the financial product; performing a reverse sort operation on the triplet data based on the interaction frequency to obtain a sample sequence; Setting a plurality of cutoff values ​​according to the sample sequence and the interaction frequency, and setting a suppression function according to the cutoff values; performing score calculation on the triple data based on the suppression function to obtain a scoring result; Rearranging the financial products according to the scoring results to obtain a target display sequence; and recommending target financial products corresponding to the order of the target display sequence to the user; The setting of multiple cutoff values ​​according to the sample sequence and the interaction frequency includes: Dividing the sample sequence into intervals based on a preset Pareto analysis method to obtain a first interval, a second interval, and a third interval, wherein any two intersections of the first interval, the second interval, and the third interval are empty; setting the minimum value of the second interval as the first delimiting value, and setting the maximum value of the second interval as the second delimiting value; The step of setting a suppression function according to the demarcation value includes: A first piecewise function corresponding to the first interval, a second piecewise function corresponding to the second interval, and a third piecewise function corresponding to the third interval are set, wherein the first piecewise function is expressed as f(c)=log(c+2), the second piecewise function is expressed as f(c)=c, and the third piecewise function is expressed as f(c)=c 0.75 +5; Setting the segmentation value of the first segmentation function according to the first threshold value, setting the segmentation value of the second segmentation function according to the first threshold value and the second threshold value, and setting the segmentation value of the third segmentation function according to the second threshold value to obtain a suppression function; The triple data includes multiple groups of data to be scored, and the scoring result is obtained by performing score calculation on the triple data based on the suppression function, including: For each set of scoring data in the triple data, the scoring data is input into a preset deep learning model to perform refined ranking score calculation, and the refined ranking score corresponding to the scoring data is output; the triple data is divided according to the first cutoff value, the second cutoff value and the interaction frequency to obtain a first data set, a second data set and a third data set; a first refined ranking score set corresponding to the first data set, a second refined ranking score set corresponding to the second data set, and a third refined ranking score set corresponding to the third data set are determined; the first data set is input into the first A piecewise function is provided to output a first rearranged score set, the second data set is input into the second piecewise function to output a second rearranged score set, and the third data set is input into the third piecewise function to output a third rearranged score set; a first score is obtained by dividing the first refined sorting score by the first rearranged score, a second score is obtained by dividing the score in the second refined sorting score set by the score in the second rearranged score set, and a third score is obtained by dividing the score in the third refined sorting score set by the score in the third rearranged score set; and a scoring result is obtained according to the first score, the second score and the third score.

2. The information rearrangement method based on the financial system according to claim 1, characterized in that: For each of the financial products, data generated by the user's interaction with the financial product within a preset time period is collected to obtain sample data, including: For each financial product, record the user's click behavior, collection behavior, and purchase behavior of the financial product within a preset time period; Collecting a first frequency of the user's click behavior, a second frequency of the collection behavior, and a third frequency of the purchase behavior; Obtaining an interaction frequency according to the first frequency, the second frequency, and the third frequency; Obtaining interactive behavior according to the click behavior, the collection behavior, and the purchase behavior; Sample data is obtained according to the interaction frequency and the interaction behavior.

3. The information rearrangement method based on the financial system according to claim 1, characterized in that: The sample sequence is divided into intervals based on a preset Pareto analysis method to obtain a first interval, a second interval, and a third interval, including: Based on a preset Pareto analysis method, determining an interval in the sample sequence that is within a first preset ratio as a first interval, determining an interval in the sample sequence that is within a second preset ratio as a second interval, and determining an interval in the sample sequence that is within a third preset ratio as a third interval; or, Setting a first ratio threshold and a second ratio threshold according to a preset Pareto analysis method; The sample sequence is segmented according to the first ratio threshold and the second ratio threshold, and a sequence portion smaller than the first ratio threshold is determined as a first interval, a sequence portion greater than or equal to the first ratio threshold and smaller than the second ratio threshold is determined as a second interval, and a sequence portion greater than or equal to the second ratio threshold is determined as a third interval.

4. The information rearrangement method based on the financial system according to claim 1, characterized in that: The dividing the triple data according to the first dividing value, the second dividing value and the interaction frequency to obtain a first data set, a second data set and a third data set includes: comparing the interaction frequency with the first cutoff value and the second cutoff value; taking the data in the triple data whose interaction frequency is less than the first cutoff value as the first data set; taking the data in the triplet data whose interaction frequency is greater than or equal to the first cutoff value and less than or equal to the second cutoff value as the second data set; The data in the triplet data whose interaction frequency is greater than the second cutoff value is used as the third data set.

5. An information rearrangement device based on a financial system, characterized in that: The device comprises: Product acquisition module, used to obtain all financial products in the financial system; a data collection module configured to collect, for each of the financial products, data generated by user interactions with the financial product within a preset time period to obtain sample data, wherein the sample data includes interaction frequency; A data generation module, configured to generate triplet data based on the user information of the user, the sample data, and the financial product; a sequence inversion module, configured to perform an inversion operation on the triple data based on the interaction frequency to obtain a sample sequence; A cutoff value setting module, configured to set a plurality of cutoff values ​​according to the sample sequence and the interaction frequency, and to set a suppression function according to the cutoff values; A score calculation module, configured to calculate the score of the triple data based on the suppression function to obtain a score result; a target generation module, configured to rearrange the financial products according to the scoring results to obtain a target display sequence; A product recommendation module, configured to recommend target financial products corresponding to the order of the target display sequence to the user; The setting of multiple cutoff values ​​according to the sample sequence and the interaction frequency includes: Dividing the sample sequence into intervals based on a preset Pareto analysis method to obtain a first interval, a second interval, and a third interval, wherein any two intersections of the first interval, the second interval, and the third interval are empty; setting the minimum value of the second interval as the first delimiting value, and setting the maximum value of the second interval as the second delimiting value; The step of setting a suppression function according to the demarcation value includes: A first piecewise function corresponding to the first interval, a second piecewise function corresponding to the second interval, and a third piecewise function corresponding to the third interval are set, wherein the first piecewise function is expressed as f(c)=log(c+2), the second piecewise function is expressed as f(c)=c, and the third piecewise function is expressed as f(c)=c 0.75 +5; Setting the segmentation value of the first segmentation function according to the first threshold value, setting the segmentation value of the second segmentation function according to the first threshold value and the second threshold value, and setting the segmentation value of the third segmentation function according to the second threshold value to obtain a suppression function; The triple data includes multiple groups of data to be scored, and the scoring result is obtained by performing score calculation on the triple data based on the suppression function, including: For each set of scoring data in the triple data, the scoring data is input into a preset deep learning model to perform refined ranking score calculation, and the refined ranking score corresponding to the scoring data is output; the triple data is divided according to the first cutoff value, the second cutoff value and the interaction frequency to obtain a first data set, a second data set and a third data set; a first refined ranking score set corresponding to the first data set, a second refined ranking score set corresponding to the second data set, and a third refined ranking score set corresponding to the third data set are determined; the first data set is input into the first A piecewise function is provided to output a first rearranged score set, the second data set is input into the second piecewise function to output a second rearranged score set, and the third data set is input into the third piecewise function to output a third rearranged score set; a first score is obtained by dividing the first refined sorting score by the first rearranged score, a second score is obtained by dividing the score in the second refined sorting score set by the score in the second rearranged score set, and a third score is obtained by dividing the score in the third refined sorting score set by the score in the third rearranged score set; and a scoring result is obtained according to the first score, the second score and the third score.

6. An electronic device, characterized in that: The electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the information rearrangement method based on the financial system according to any one of claims 1 to 4 when executing the computer program.

7. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the information rearrangement method based on a financial system according to any one of claims 1 to 4 is implemented.

Images