CN116823493A

CN116823493A - Feature processing method and device, storage medium and electronic equipment

Info

Publication number: CN116823493A
Application number: CN202310511452.5A
Authority: CN
Inventors: 靳乾乾; 唐觊隽; 张文博; 吴斌
Original assignee: Alipay Hangzhou Information Technology Co Ltd
Current assignee: Alipay Hangzhou Information Technology Co Ltd
Priority date: 2023-05-06
Filing date: 2023-05-06
Publication date: 2023-09-29

Abstract

The specification discloses a feature processing method, a device, a storage medium and an electronic apparatus, wherein the method comprises the following steps: determining a user characteristic insurance occurrence wide table based on at least one initial insurance user characteristic and insurance user occurrence information, determining at least one characteristic box body of the user characteristic insurance occurrence wide table, then acquiring target insurance user characteristics respectively corresponding to the characteristic box bodies, and determining insurance characteristic evaluation indexes corresponding to the target insurance user characteristics based on box body risk characteristic data of the characteristic box bodies so as to perform characteristic screening recommendation processing on each target insurance user characteristic.

Description

Feature processing method and device, storage medium and electronic equipment

Technical Field

The present disclosure relates to the field of computer technologies, and in particular, to a feature processing method, a device, a storage medium, and an electronic apparatus.

Background

In the development of the insurance wind control strategy, effective one or more insurance user characteristics are often selected by combining mass insurance user data of a platform, characteristic combination is carried out according to the insurance user characteristics, and a combination strategy rule is developed according to the characteristic combination in an insurance wind control strategy development environment, wherein the combination strategy rule refers to a rule strategy consisting of a plurality of insurance user characteristics serving as characteristic variables, for example, model participation of a relevant insurance wind control strategy model is carried out according to the characteristic combination obtained by the insurance user characteristics, and after the sensitivity of the insurance wind control strategy model is determined based on corresponding evaluation indexes, strategy online deployment is carried out on the relevant insurance wind control strategy model.

Disclosure of Invention

The specification provides a feature processing method, a device, a storage medium and electronic equipment, and the technical scheme is as follows:

in a first aspect, the present specification provides a feature processing method, the method comprising:

acquiring at least one initial insurance user characteristic and insurance user insurance information aiming at insurance transactions;

determining a user characteristic insurance policy table based on each of the initial insurance user characteristics and the insurance user insurance policy information, and determining at least one characteristic box of the user characteristic insurance policy table;

acquiring target insurance user characteristics corresponding to the characteristic boxes respectively, and determining insurance characteristic evaluation indexes corresponding to the target insurance user characteristics based on box risk characteristic data of the characteristic boxes;

and carrying out feature screening recommendation processing on the features of each target insurance user based on the insurance feature evaluation indexes.

In a second aspect, the present specification provides a feature processing apparatus, the apparatus comprising:

the information acquisition module is used for acquiring at least one initial insurance user characteristic and insurance user insurance information aiming at the insurance transaction;

the data determining module is used for determining a user characteristic insurance-issuing wide table based on the initial insurance user characteristics and the insurance user insurance-issuing information and determining at least one characteristic box body of the user characteristic insurance-issuing wide table;

And the feature screening module is used for acquiring the target insurance user features corresponding to the feature boxes respectively, determining the insurance feature evaluation indexes corresponding to the target insurance user features based on the box risk feature data of the feature boxes, and carrying out feature screening recommendation processing on the target insurance user features based on the insurance feature evaluation indexes.

In a third aspect, the present description provides a computer storage medium storing at least one instruction adapted to be loaded by a processor and to perform the method steps of one or more embodiments of the present description.

In a fourth aspect, the present description provides a computer program product storing at least one instruction adapted to be loaded by a processor and to perform the method steps of one or more embodiments of the present description.

In a fifth aspect, the present description provides an electronic device, which may include: a processor and a memory; wherein the memory stores a computer program adapted to be loaded by the processor and to perform the method steps of one or more embodiments of the present description.

The technical scheme provided by some embodiments of the present specification has the following beneficial effects:

In one or more embodiments of the present disclosure, a service platform may automatically combine at least one initial insurance user feature and insurance user risk information to form a user feature risk-occurrence wide table, and may obtain a corresponding target insurance user feature by determining at least one feature box of the user feature risk-occurrence wide table, and automatically determine an insurance feature evaluation index corresponding to the target insurance user feature based on box risk feature data of the feature box, so that feature screening recommendation processing may be automatically performed on each target insurance user feature according to the insurance feature evaluation index, and the whole feature processing screening and recommendation process may not need to complete manually and automatically screening effective target insurance user features from massive insurance transaction data, thereby optimizing a feature processing screening process, improving the degree of automation of feature processing, and improving feature processing efficiency.

Drawings

In order to more clearly illustrate the technical solutions of the present specification or the prior art, the following description will briefly introduce the drawings that are required to be used in the embodiments or the prior art descriptions, it is obvious that the drawings in the following description are only some embodiments of the present specification, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1a is a schematic illustration of a scenario of a feature processing system provided herein;

FIG. 1b is a schematic diagram of a service platform feature process architecture provided in the present specification;

FIG. 2 is a schematic flow chart of a feature processing method provided in the present specification;

FIG. 3 is a schematic diagram of a flow chart of a broad table and box determination provided in the present specification;

FIG. 4 is a schematic illustration of a determination of an initial user feature risk delivery width table provided herein;

FIG. 5 is a schematic flow chart of a feature binning provided in the present specification;

FIG. 6 is a schematic view of a feature processing apparatus provided in the present specification;

FIG. 7 is a schematic diagram of an electronic device provided in the present specification;

FIG. 8 is a schematic diagram of the architecture of the operating system and user space provided herein;

FIG. 9 is an architecture diagram of the android operating system of FIG. 8;

FIG. 10 is an architecture diagram of the IOS operating system of FIG. 8.

Detailed Description

The following description of the embodiments of the present invention will be made apparent from, and elucidated with reference to, the drawings of the present specification, in which embodiments described are only some, but not all, embodiments of the present specification. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are intended to be within the scope of the present disclosure.

In the description of the present specification, it should be understood that the terms "first," "second," and the like are used for descriptive purposes only and are not to be construed as indicating or implying relative importance. In the description of the present specification, it should be noted that, unless expressly specified and limited otherwise, "comprise" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those listed steps or elements but may include other steps or elements not listed or inherent to such process, method, article, or apparatus. The specific meaning of the terms in this specification will be understood by those of ordinary skill in the art in the light of the specific circumstances. In addition, in the description of the present specification, unless otherwise indicated, "a plurality" means two or more. "and/or", describes an association relationship of an association object, and indicates that there may be three relationships, for example, a and/or B, and may indicate: a exists alone, A and B exist together, and B exists alone. The character "/" generally indicates that the context-dependent object is an "or" relationship.

In the related art, a manual feature screening mode is often adopted to manually find the feature data of the insurance users in various dimensions in combination with the massive insurance user data of the platform, and one or more insurance user features are manually screened on the feature data of the insurance users by adopting a corresponding feature screening mode in feature engineering. However, manual feature screening is time-consuming and labor-consuming, which results in problems of low degree of automation and low efficiency of feature processing.

The present specification is described in detail below with reference to specific examples.

Referring to fig. 1a, a schematic view of a feature processing system is provided in the present specification. As shown in fig. 1a, the feature processing system may include at least a client cluster and a service platform 100.

The client cluster may include at least one client, as shown in fig. 1, specifically including a client 1 corresponding to a user 1, a client 2 corresponding to a user 2, …, and a client n corresponding to a user n, where n is an integer greater than 0.

Each client in the client cluster may be a communication-enabled electronic device including, but not limited to: wearable devices, handheld devices, personal computers, tablet computers, vehicle-mounted devices, smart phones, computing devices, or other processing devices connected to a wireless modem, etc. Electronic devices in different networks may be called different names, for example: a user equipment, an access terminal, a subscriber unit, a subscriber station, a mobile station, a remote terminal, a mobile device, a user terminal, a wireless communication device, a user agent or user equipment, a cellular telephone, a cordless telephone, a personal digital assistant (personal digital assistant, PDA), an electronic device in a 5G network or future evolution network, and the like.

The service platform 100 may be understood as an insurance transaction service platform for providing related insurance services to the outside, and the service platform 100 may be a separate server device, for example: rack-mounted, blade, tower-type or cabinet-type server equipment or hardware equipment with stronger computing capacity such as workstations, mainframe computers and the like is adopted; the server cluster may also be a server cluster formed by a plurality of servers, and each server in the server cluster may be formed in a symmetrical manner, wherein each server is functionally equivalent and functionally equivalent in a transaction link, and each server may independently provide services to the outside, and the independent provision of services may be understood as no assistance of another server is needed.

In one or more embodiments of the present disclosure, the service platform 100 may establish a communication connection with at least one client in the client cluster, based on the interaction of insurance transaction data during the feature processing of the communication connection, the service platform 100 may be a platform for a plurality of dangerous services including, but not limited to, health insurance, medical insurance, heavy illness insurance, etc., the insurance transaction request received by the service platform 100 by the client may include, but not limited to, insurance claim settlement request, insurance purchase request, etc., the service platform 100 may specifically be a transaction service platform of any insurance company capable of handling (including purchasing, claim settlement, etc.) health insurance, and the related health insurance may include, but not limited to, medical fee insurance, serious illness insurance, disability harvest loss insurance, nursing insurance, etc.;

The service platform 100 may provide insurance transactions to the outside, and the users of one or more clients perform related insurance operations, such as selecting a certain insurance policy, initiating insurance claims, etc., based on the insurance transactions provided by the service platform 100, where the service platform 100 may record insurance transaction data related to the insurance transactions, such as history claims records, application time, time of emergence, insurance period, user information, etc., under the condition of being authorized by the users or being fully authorized by the parties.

Referring to fig. 1b, fig. 1b is a schematic architecture diagram of a service platform feature process according to the present specification, as shown in fig. 1b, the service platform 100 may be configured with metadata management (system), through which metadata information of an insurance user feature x and insurance user risk information Y may be configured, and various initial insurance user features-feature x and Y corresponding to insurance user risk information may be obtained based on a data repository corresponding to an insurance transaction, and the features x and Y may be uniformly managed, where information of the (initial) insurance user features-feature x and insurance user risk information-Y may be stored in the form of one or more database tables (the (initial) insurance user features-feature x may be referred to as an x table).

The service platform 100 may execute the feature processing method of one or more embodiments of the present disclosure, thereby implementing automatic production of a user feature risk-out broad table through an insurance feature evaluation index in the development of an insurance policy, automatically determining at least one feature box for the user feature risk-out broad table, performing feature screening recommendation processing on target insurance user features corresponding to each feature box based on the insurance feature evaluation index, selecting one or more effective insurance user features according to massive insurance user data of the platform, performing feature combination according to the insurance user features, performing feature combination analysis according to the feature combinations, and further, may also be combined with automatic machine learning AutoML to better develop a combination policy rule in an insurance policy development environment to assist in generating an insurance risk policy.

It should be noted that, the service platform 100 establishes a communication connection with at least one client in the client cluster through a network for interactive communication, where the network may be a wireless network, or may be a wired network, where the wireless network includes, but is not limited to, a cellular network, a wireless local area network, an infrared network, or a bluetooth network, and the wired network includes, but is not limited to, an ethernet network, a universal serial bus (universal serial bus, USB), or a controller area network. In one or more embodiments of the specification, techniques and/or formats including HyperText Mark-up Language (HTML), extensible markup Language (Extensible Markup Language, XML), and the like are used to represent data exchanged over a network (e.g., target compression packages). All or some of the links may also be encrypted using conventional encryption techniques such as secure socket layer (Secure Socket Layer, SSL), transport layer security (Transport Layer Security, TLS), virtual private network (Virtual Private Network, VPN), internet protocol security (Internet Protocol Security, IPsec), and the like. In other embodiments, custom and/or dedicated data communication techniques may also be used in place of or in addition to the data communication techniques described above.

The embodiment of the feature processing system provided in the present disclosure and the feature processing method in one or more embodiments belong to the same concept, and an execution subject corresponding to the feature processing method related to one or more embodiments in the present disclosure may be the service platform 100, and an implementation process of the embodiment of the feature processing system may be the following method embodiment, which is not described herein again.

The feature processing method provided in one or more embodiments of the present disclosure is described in detail below based on the schematic diagrams shown in fig. 1 a-1 b.

Referring to fig. 2, a flow diagram of a feature processing method, which may be implemented in dependence on a computer program and may be run on a feature processing device based on von neumann system, is provided for one or more embodiments of the present description. The computer program may be integrated in the application or may run as a stand-alone tool class application. The feature processing device may be a service platform.

Specifically, the feature processing method comprises the following steps:

s102: acquiring at least one initial insurance user characteristic and insurance user insurance information aiming at insurance transactions;

specifically, the service platform may provide an insurance transaction to the outside, and the user of one or more clients performs related insurance operations, such as selecting a certain insurance policy, initiating an insurance claim, etc., based on the insurance transaction provided by the service platform, where the service platform may record insurance transaction data related to the insurance transaction, such as history claim records, application time, time of emergence, insurance period, user information, etc., under the condition that the service platform is authorized by the user or the parties are fully authorized.

Alternatively, the insurance transaction data may be stored in a data repository associated with the service platform, from which at least one of the initial insurance user characteristics and insurance user risk information may be obtained.

The initial insurance user characteristic can be understood as acquired user characteristic data related to insurance transactions, such as user payment characteristics in the uid dimension, user lbs characteristics in the identity card dimension, user time characteristics in the identity information dimension, male and female characteristics, professional characteristics and the like;

furthermore, the initial insurance user characteristics can be pre-defined and configured in combination with the characteristic engineering application, so that the corresponding initial insurance user characteristics are extracted from the insurance transaction data;

in some embodiments, the (initial/target) insurance user feature may be represented by a character x, i.e., feature x represents the (initial/target) insurance user feature;

the insurance user risk information can be represented by a character y, travel information of the insurance user is specific to one or more users, the insurance user risk information records whether the one or more users belong to dangerous type risk groups or non-risk groups, and at least the risk users and the non-risk users are distinguished in the insurance user risk information according to different user types.

In some embodiments, the insurance user risk information is typically in the form of an insurance user risk table, i.e., a Y table: a group of users to be protected for a dangerous identity card dimension, wherein users who are not at risk are distinguished by marks;

in one or more embodiments of the present disclosure, a service platform may be configured with a metadata management system, through which metadata information of an insurance user feature x and insurance user risk information Y may be configured, and various initial insurance user features-feature x and Y tables corresponding to the insurance user risk information may be obtained based on a data warehouse corresponding to an insurance transaction, and the feature x and Y tables may be unified and managed, and classified, duplicate features or data may be removed, and information of feature data sources, feature granularity, feature storage mode, feature description, feature classification, etc. corresponding to the feature x may be recorded, so as to maintain and facilitate management of the feature x and Y tables.

It can be appreciated that the metadata management system can avoid the scattering of user features and risk-out data in the data warehouse, and can realize the regularity of the scattered user features and the risk-out data.

S104: determining a user characteristic insurance policy table based on each of the initial insurance user characteristics and the insurance user insurance policy information, and determining at least one characteristic box of the user characteristic insurance policy table;

The user characteristic insurance policy table is a database table model with more fields, which is fused with insurance user characteristics and insurance user insurance policy information, and can be understood as a policy table model, and is a database table which correlates insurance policy indexes, user characteristics, insurance attributes and the like related to an insurance main body;

the user characteristic risk-out wide table can be used for data preparation before training of the insurance data mining model and even recommending of the insurance data mining strategy, and the related fields are placed in a wide table, so that characteristic screening and characteristic recommending work in data mining can be greatly improved, and the computing efficiency can be improved.

In one or more embodiments of the present disclosure, a service platform automatically merges, based on each initial insurance user feature and insurance user risk information, a user feature risk-out wide table, and performs a box-division processing on the user feature risk-out wide table, where the box-division processed user feature risk-out wide table corresponds to at least one feature box.

S106: acquiring target insurance user characteristics corresponding to the characteristic boxes respectively, and determining insurance characteristic evaluation indexes corresponding to the target insurance user characteristics based on box risk characteristic data of the characteristic boxes;

The target insurance user features are usually associated with box names which correspond to all box feature data in the feature box together, for example, the box names are usually used as target insurance user features;

the risk feature data of the box body can be understood as risk data associated with all insurance user features in the box body, namely risk information about whether the user associated with the user features is at risk.

In a possible embodiment, the acquiring the target insurance user features corresponding to the feature boxes respectively may be:

and updating the characteristic parameters of at least one initial insurance user characteristic in the characteristic box body to obtain the characteristic box body names of the characteristic box bodies, and determining the characteristic box body names respectively corresponding to the characteristic box bodies as target insurance user characteristics.

Illustratively, the box division processing can divide the wide table numerical data one by one in the user characteristic risk-out wide table based on the partition configuration to obtain a plurality of partitions, namely, rough classification is carried out on table data corresponding to a large number of characteristics x to obtain a plurality of partitions, then each partition is divided into box bodies in parallel to obtain at least one characteristic box body, the classified characteristic names corresponding to one characteristic box body, namely, the characteristic box body names, which are the abstract representation of the safety user characteristics, are obtained, then a large number of original safety user characteristics in the box bodies are updated, namely, the safety user characteristics after abstract representation is carried out on all box body data in the characteristic box bodies, namely, target safety user characteristics are updated, i box body data are assumed to correspond to one certain box body, i box body data are respectively corresponding to i large number of initial safety user characteristics before box division, i box body data are partitioned together through box division operation at this time to generate one characteristic box body, all i box body data in the characteristic box bodies are regarded as abstract representation of one safety user characteristic, namely, i box body data corresponding to one abstract characteristic is represented by one target safety user characteristic, namely, i box body data are represented by the abstract characteristic, namely, i box body data are represented by the target user characteristics are obtained by the abstract characteristic, namely, i box body data are represented by the target user characteristics, i box body data are represented by the corresponding to the target user characteristics, namely, i box body data are represented by the initial user characteristics, i box body data are obtained by the abstract user characteristics, namely, i box body data.

It will be appreciated that the target insurance user features are a combination of the original initial insurance user features of any one of the original i box data, the range of target insurance user features covering all of the original initial insurance user features.

The insurance characteristic evaluation index is an index for quantitatively evaluating the characteristic quality of a certain insurance user by calculating a corresponding characteristic evaluation quantity for the characteristics of the insurance user and taking the characteristic evaluation quantity as a characteristic;

optionally, the insurance feature evaluation index may be a fit of one or more of evaluation indexes such as a feature lifting index (feature lift index), a feature information value index (IV index), a feature weight index (woe index), and the like;

in a possible implementation manner, the determining, based on the box risk feature data of the feature box, an insurance feature evaluation index corresponding to the target insurance user feature may be:

the service platform calculates a characteristic lifting index based on the box risk characteristic data of the characteristic box, and takes the characteristic lifting index as an insurance characteristic evaluation index corresponding to the target insurance user characteristic;

the characteristic lifting index is determined based on the target risk user proportion and the total risk proportion of the target insurance risk in the characteristic box body risk characteristic data.

The box risk feature data is the combination of box user feature data corresponding to the feature box and insurance risk data corresponding to the user (the user feature data corresponds to the mapped user, the box user feature data is the user feature data corresponding to the user, the box user feature data may be different from the original initial user feature data), and a list coding data format is usually adopted in the box with a wide list.

The target risk user proportion is the proportion of the number of risk users of the target insurance risk (such as serious risk and medical risk) corresponding to the risk feature data of all the cases in the feature case and the total number of users corresponding to the feature case.

The total risk-out user proportion is the proportion of all the safety users and all the appearance users of the target safety risk.

Illustratively, the calculation of the feature improvement index may be: the service platform determines the proportion of the target risk users based on the box risk feature data of the feature box, and the service platform acquires the proportion of the total risk users of the target insurance risk;

and then the service platform takes the ratio of the target risk user proportion to the total risk user proportion as a characteristic lifting index lift.

In one or more embodiments of the present disclosure, based on the insurance-feature-evaluation index of the target insurance user feature, the effectiveness of the user feature in referencing the model may be quantified to determine whether to make a recommendation of the target insurance user feature.

S108: and carrying out feature screening recommendation processing on the features of each target insurance user based on the insurance feature evaluation indexes.

Illustratively, different target insurance user features correspond to different insurance feature evaluation indexes, the insurance feature evaluation indexes can be one or more, the effectiveness degree of one or more target insurance user features can be measured by combining the insurance feature evaluation indexes, and relatively effective insurance user features are automatically screened out according to preset screening criteria to be recommended, so that effective insurance user features and high-performance automatic feature screening computing capacity are provided for model insurance service, insurance transaction systems, offline training and the like.

Illustratively, the numerical values of the insurance feature evaluation indexes corresponding to the plurality of target insurance user features can be used for obtaining the priority of each target insurance user feature, selecting the target insurance user features with the target number according to the priority order for recommendation,

Illustratively, the service platform executing the feature screening recommendation processing for each target insurance user feature based on the insurance feature evaluation index may be:

specifically, an evaluation index threshold for the insurance feature evaluation index may be obtained;

the evaluation index threshold is a threshold value or a critical value set for the insurance feature evaluation index, and is used for feature screening based on the insurance feature evaluation index and the set evaluation index threshold.

Specifically, an evaluation index threshold for the insurance feature evaluation index may be obtained; matching each insurance characteristic evaluation index with an evaluation index threshold;

if a first evaluation index is matched with the evaluation index threshold value in each insurance feature evaluation index, acquiring a first insurance user feature corresponding to the first evaluation index, and performing feature recommendation processing on the first insurance user feature

Illustratively, the matching of each insurance feature evaluation index with the evaluation index threshold may be performed based on a matching rule, for example, if the matching rule indicates that the insurance feature evaluation index greater than the evaluation index threshold is the first evaluation index, each insurance feature evaluation index is traversed, and whether the insurance feature evaluation index is greater than the evaluation index threshold is detected.

Further, the evaluation index threshold may be preset, and the batch automatic calculation of the evaluation index is performed on the target insurance user features corresponding to each feature box in batches by calling the index calculation service of the insurance feature evaluation index, and feature validity evaluation screening is performed in combination with the evaluation index threshold, so that feature recommendation is performed on the first insurance user features expected to meet the set evaluation index threshold.

Optionally, fig. 3 is a schematic flow chart of determining a broad table and a box body according to the present specification, as shown in fig. 3, the service platform executes the determining a broad table of user characteristics based on each initial insurance user characteristic and the insurance user risk information, and determines at least one characteristic box body of the broad table of user characteristics, which may be:

s202: carrying out data fusion processing on the initial insurance user characteristics and the insurance user insurance information to obtain an initial user characteristic insurance width table;

illustratively, the following steps may be performed:

a2: acquiring a plurality of user characteristic information tables corresponding to the initial insurance user characteristics, and acquiring a user characteristic insurance base table corresponding to the insurance user insurance information;

in some embodiments, the service platform may record and store in a data repository, such as historical claim records, application time, time of risk, insurance period, user information, etc., insurance transaction data related to the insurance transaction with user authorization or with sufficient authorization from parties. Meanwhile, the service platform configures information of (initial) insurance user characteristics-characteristics x and insurance user insurance output information-Y based on the cooperation of the metadata management system and the data warehouse;

Illustratively, the value correspondence of insurance user risk information-Y may be configured by 0 and 1, for example, with respect to the user marked for risk, different insurance risk types: heavy-disease insurance, Y-configured users who are in danger and develop heavy diseases, medical insurance, Y-configured users who are in medical reimbursement, the value of Y is defined according to different dangerous seeds,

illustratively, x is an insurance user feature, for example, a user health feature, a user basic information feature, and the like of an insurance user, and a set of universal capabilities of a bottom layer is adopted, which is equivalent to that the user feature x is universal, but insurance user insurance information-Y of different dangerous types needs to define Y value attributes, namely a specific Y value indicates whether the insurance dangerous types are in danger;

in one or more embodiments herein, after the information of (initial) insurance user feature-feature x and insurance user risk information-Y is configured, in the data repository, the information of (initial) insurance user feature-feature x and insurance user risk information-Y may be stored in the form of one or more database tables (initial) insurance user feature-feature x corresponds to a user feature information table (which may be referred to as an x-table).

The user characteristic risk-free base table is a base table (base table) containing insurance user risk-free information as a Y tag, and can also be called a Y table.

In one or more embodiments of the present disclosure, the service platform may obtain a user feature information table corresponding to a plurality of initial insurance user features from the data repository, and obtain a user feature risk-out base table corresponding to the insurance user risk-out information.

A4: and automatically carrying out data association fusion processing on the user characteristic risk-out base table and the plurality of user characteristic information tables to obtain an initial user characteristic risk-out wide table.

In one or more embodiments of the present disclosure, the user feature information x may then be a feature of a large data size, and is characterized by a plurality of underlying user feature information tables, i.e., feature risk-occurrence tables, using a base table containing y-tags, i.e., a user feature risk-occurrence table, to associate the plurality of user feature information tables, and then performing table fusion on the user feature risk-occurrence table and the associated user feature information table to obtain a user feature risk-occurrence width table, also referred to as an initial user feature risk-occurrence width table, that is at least feedback of the user feature risk-occurrence dimension and the user feature.

S204: and carrying out characteristic box division processing on the initial user characteristic risk-free wide list to obtain the user characteristic risk-free wide list comprising at least one characteristic box body.

According to some embodiments, the box division processing may obtain corresponding partition configuration (information), partition the wide table numerical data one by one in the user feature risk-out wide table based on the partition configuration, that is, rough classification is performed on table data corresponding to a large number of features x to obtain each partition, then each partition is divided into boxes in parallel to form a box body, so as to obtain at least one feature box body, the classified feature name corresponding to one feature box body, that is, the feature box name is an abstract representation of a safe user feature, then update a large number of original initial safe user features in the box body, update the abstract representation of all box body data in the feature box body, that is, target safe user features, and assume that i box body data corresponding to a certain box body correspond to i box body features, the i box body data are respectively corresponding to i large number of initial safe user features before box division, partition the i box body data together through box division operation to generate a feature box body, use of all i box body data corresponding to the feature box body as abstract representation of the feature user features, that is also implement the abstract representation of the corresponding to i box body data corresponding to the target user features (i is also the abstract representation of the user features of the box body, that is based on the abstract representation of the target user features).

Alternatively, the partition configuration (information) may be one or more of a data period, a data length/width, a data equidistance, a data equiprequency, a data cluster, and the like, which are set based on the data timestamp.

In one or more embodiments of the present disclosure, converting a large number of variables, that is, insurance user features x, into categories presents feature meanings that are more capable of feeding back a category of insurance users, and at the same time, enabling a subsequent model to participate in an application phase algorithm such as a parameter to reduce noise interference, and dividing the insurance user features x in a certain range into determined sub-blocks to form a feature box. For example, a user who predicts which insurance user features will have a tendency to be more prone to certain insurance business dimensions (e.g., choose to purchase a dangerous seed, insurance claim) than others, and the insurance user's years are a continuous feature variable x, typically the foregoing manner may bin the insurance user features as years, thus yielding a plurality of feature bins, e.g., 18 or less, 18-24, 25-36, 36-56, 56 and above. Moreover, the multiple feature boxes exhibit a similar set of risk attributes due to similar years.

In one or more embodiments of the present disclosure, the feature box is more meaningful when the risk-out attribute can be divided into a simple range based on a priori, that is, similar insurance user features can present common attribute features when they fall into a partition to form the feature box, and in subsequent practical applications, such as situations of model overfitting, large errors and the like can be avoided.

Fig. 4 is a schematic diagram illustrating a determination of an initial user feature risk-out broad table according to the present specification, as shown in fig. 4, and the performing the data association fusion processing on the user feature risk-out base table and the multiple user feature information tables to obtain the initial user feature risk-out broad table may be:

s3002: acquiring a feature field processing threshold value and a base table association threshold value corresponding to a data processing service, and determining feature field quantity and feature table quantity corresponding to all the user feature information tables;

the data processing service, i.e. the service that combines the user profile risk basis table and the associated plurality of user profile information tables, may be an ODPS service in some embodiments.

In practical applications, there is a certain system processing limitation on the processing capacity of the data processing service objectively, a feature field processing threshold is generally set according to the upper limit of the single table data field of the data processing service, and a base table association threshold is generally set according to the upper limit of the table association fusion of the number of processing services.

The feature field processing threshold is a threshold or a critical value set for the upper limit of feature field processing of the user feature information table, which is set for a single table, for example, the upper limit of feature field processing of the data processing service limiting single table is 1200 fields, the data fields processed in the processing stages of subsequent merging, binning and the like are usually smaller than the upper limit, and a threshold smaller than the upper limit of feature field processing of the single table can be set, for example, the feature field processing threshold can be set as 1000 fields.

Schematically, if the feature field quantity is greater than the feature field processing threshold, the original user feature risk-out base table may be segmented to obtain a plurality of temporary user feature risk-out base tables.

The base table association threshold is a threshold or a critical value set for the number of user feature information tables associated with the user feature risk base table, for example, the data processing service ODPS service limits one base table to be associated with 15 user feature information tables at most.

Schematically, if the number of the feature tables is greater than the base table association threshold, the original user feature risk-out base table may be segmented to obtain a plurality of temporary user feature risk-out base tables.

S3004: based on the feature field quantity, the feature table quantity, the feature field processing threshold value and the base table association threshold value, carrying out data association fusion processing on the user feature risk-out base table and a plurality of user feature information tables, and carrying out distribution fusion association processing to obtain a plurality of reference user feature risk-out wide tables;

b2: if the characteristic field quantity is larger than the characteristic field processing threshold value, determining a first segmentation quantity of the user characteristic risk-out base table;

illustratively, the first number of cuts may be a first target value obtained by rounding a ratio of the feature field amount to the feature field processing threshold, and the first number of cuts may be determined based on the first target value.

Illustratively, a reference processing field value smaller than the processing threshold of the feature field may be set, a ratio of the feature field value to the reference processing field value is calculated, and a value obtained by rounding the ratio is used as the first score amount.

It can be understood that the determination mode of the first score amount can be automatically set based on an actual scene, and after a plurality of user feature risk-out base tables are obtained by cutting according to the first score amount, each base table of the plurality of user feature risk-out base tables can meet a feature field processing threshold set by a feature field processing upper limit of a data processing service limiting list.

B4: if the number of the feature tables is larger than the base table association threshold, determining a second segmentation number of the user feature risk-out base table;

illustratively, the second number of cuts may be a second target value obtained by rounding a ratio of the number of feature tables to a base table association threshold, and the second number of cuts may be determined based on the second target value.

Illustratively, a reference processing table value smaller than the base table association threshold may be set, a ratio of the number of feature tables to the reference processing table value is calculated, and a value obtained by rounding the ratio is used as the second segmentation number.

It can be understood that the determination manner of the second segmentation number can be automatically set based on the actual scene, and after the multiple user feature risk-out base tables are obtained by segmentation according to the second segmentation number, each base table of the multiple user feature risk-out base tables can meet a base table association threshold set by the association processing upper limit of the data processing service limiting list table.

B6: performing segmentation processing on the user characteristic risk-out base table based on the first segmentation number and/or the second segmentation number to obtain a third number of user characteristic risk-out base tables, and determining a plurality of user characteristic information tables associated with each user characteristic risk-out base table;

it will be appreciated that after a third number of user feature risk basis tables are obtained, a plurality of user feature information tables associated with each of the user feature risk basis tables are determined. For example, if the associated user feature information table x has 40 tables equivalent to the base table association threshold being worse, because one user feature risk-out base table can only be associated with 15 tables according to the base table association threshold, the user feature risk-out base table is equivalent to being divided into a third number-3 of indicated user feature risk-out base tables.

B8: and carrying out information fusion processing on each user characteristic risk-out base table and the plurality of user characteristic information tables associated with the user characteristic risk-out base tables to obtain the third number of reference user characteristic risk-out wide tables.

The information fusion processing can be understood as data merging and fusion processing is performed on the user characteristic risk-out base table and the user characteristic information table, so that risk-out dimension information of any user and user characteristic information can be associated.

The reference user characteristic risk-out wide table is a temporary reference user characteristic risk-out wide table obtained by information fusion of a plurality of associated user characteristic information tables on the basis of the user characteristic risk-out basic table.

S3006: and carrying out table data fusion on the multiple reference user characteristic risk-out broad tables to obtain an initial user characteristic risk-out broad table.

It can be understood that the table data fusion of the temporary broad table is performed after the multiple reference user characteristic risk-out broad tables are obtained, so as to obtain the initial user characteristic risk-out broad table. Illustratively, the initial user feature risk profile table corresponds to an insurance user feature-x feature having n dimensions after each y-tag.

Then, carrying out characteristic box division processing on each initial user characteristic risk-free wide table to obtain a user characteristic risk-free wide table containing at least one characteristic box body

In one or more embodiments of the present disclosure, the foregoing manner realizes the automatic combination of the insurance user feature x and the user risk information Y table, breaks through the technical limitation of the word segment limitation and the table association number limitation of the data processing service list, and the feature expansion combination breaks through the upper processing limit, so that the method is applicable to the batch automatic combination of the massive insurance user feature x into the user feature risk wide table, and improves the automation capability and the data processing efficiency of the data processing.

Schematically, fig. 5 is a schematic flow chart of feature binning, as shown in fig. 5, where feature binning processing is performed on the initial user feature risk-out broad table to obtain a user feature risk-out broad table including at least one feature box, where the feature risk-out broad table may be:

s4002: obtaining partition configuration information of the initial user characteristic risk-out wide table;

s4004: determining at least one feature partition corresponding to the initial user feature risk-out wide table based on the partition configuration information;

according to some embodiments, the box division processing may obtain partition configuration (information) of the corresponding initial user feature risk-giving wide table, partition the wide table numerical data one by one in the user feature risk-giving wide table based on the partition configuration (information), that is, rough classification is performed on table data corresponding to a large number of features x to obtain a plurality of partitions, then each partition limits parallel box division to form a box body according to a certain field, so as to obtain at least one feature box body, the classified feature name corresponding to one feature box body, that is, the feature box body name is an abstract representation of a safe user feature, then update a large number of original initial safe user features in the box body, that is, the target safe user feature, in the feature box body, after abstract representation is performed on all the data in the box body, assuming that i box body data correspond to i box body data, before the box division, the i box body data correspond to i large numbers of initial safe user features, the i box body data are partitioned together through box division operation, so as to generate a feature box body, that is, the abstract feature box body data corresponding to i is, that is, the target safe user feature is, i is, and the target safe user feature is obtained by reference to the abstract feature of all the box body data, that corresponds to the target feature i.

In one possible embodiment, it may be:

c2 determining at least one data period based on the partition configuration information;

illustratively, after the initial user feature risk broad table is formed, each piece of broad table data corresponds to an attribute corresponding to a timestamp, for example: assuming 2000 insurance user features, feature values of the insurance user features x are stored based on different times, for example, the time feature x of the user in 2002 corresponds to the current time status of the user, for example, the time is 23, the time feature x of the user in 2023 is 33, the insurance user features x are related to time, and the insurance user features x are required to be stored in a wide table according to different partitions;

and C4, dividing at least one feature partition corresponding to the data time period for the initial user feature risk-out broad based on the application time stamp corresponding to each piece of initial broad data in the initial user feature risk-out broad.

Illustratively, the obtaining the partition configuration information of the initial user feature risk-out broad table may be understood as a process of capturing partition configuration, and the partition configuration information may be a data time period and may be understood as follows: assuming that a certain insurance user characteristic is taken as a user insuring characteristic, insurance policies of all insurance users correspond to insuring time points when insuring, and the insuring time points are insuring time stamps corresponding to each piece of initial broad table data, wherein the data of a time point partition of the insuring time stamp in the user insuring characteristic need to be associated, at the moment, a platform can automatically allocate, so that the user insuring time and a large broad table stored in history, such as 700 partitions for two years, can grasp which partition is in decoupling with the characteristic of a data time period of the insuring time; the data time period can be configured as a partition, and based on the data time period it de-associates user's insuring features within the time period corresponding to your insuring time point;

For example, a user is 2020.1.1, the application time stamp is 2020.1.1, then data of 2018.1.1-2020.2.30 is stored in the historical data wide table, the application time 2023.1.1 is taken as data of a reference time period during analysis, for example, data of every 30 units before application is usually referential, a system is configured by taking an application time point as a reference, the data time period is 30 days, the data time period is configured to be 30, the data time period is 60, the data time period 90 is partitioned, the data in the time periods corresponding to 30, 60 and 90 are captured, then each partition can be formed into a box according to the partition, and one partition can be used for constructing at least one box.

Illustratively, each dangerous seed such as medical treatment, serious disease, accidental injury and the like is configured on the front and back of an initial user characteristic dangerous-out wide table-xy, the corresponding observation time period of each dangerous seed is different, partitioning is needed based on the corresponding observation time period of different dangerous seeds, for example, the observation time period of a certain dangerous seed A is 2 years, the observation time period of a certain dangerous seed B is 6 months, the xy table is partitioned according to the window period to be observed of each dangerous seed, and parallel box separation is carried out after the partitioning;

s4006: and carrying out feature data box-division processing on each feature partition to obtain a feature risk-out wide table containing at least one feature box body.

In one possible embodiment, it may be:

the service platform obtains a box-dividing characteristic field value;

and D4, carrying out feature data box-division conversion processing on each feature partition based on the box-division feature field values by the service platform to obtain at least one feature box body corresponding to each feature partition so as to generate a feature risk-out wide table containing at least one feature box body.

The binning feature field value may be understood as a field threshold value that bins feature data within a feature region, and if the binning feature field value is 100, each feature region is converted into one or more feature bins according to 100 fields.

Illustratively, after determining the value of the feature field of the sub-box, the service platform may perform feature data sub-box conversion processing on each feature partition based on the value of the feature field of the sub-box, to obtain at least one feature box corresponding to each feature partition, and after completing the foregoing steps for the initial feature risk-out wide table, generate a feature risk-out wide table including at least one feature box.

It will be appreciated that the previous initial feature risk-out broad table is a large base table, with each x being a value, say the time feature is 55 years old, and that after the binning has been completed it will show that 55 years old is displayed in a bin of 50-59, and will put all of the insurance user feature data in 50-59 in a bin of 50-59. And then detecting whether the risk of the label y is detected based on the corresponding index, namely, the situation that the label y is at risk is equivalent to the situation that the label y is detail data of each user, the user information of the same insurance user characteristic x is placed in a characteristic box, the concentration or risk degree of insurance of the group of users can be reflected through the characteristic box, after the characteristic box is divided into boxes, classification characteristic names, namely, characteristic box names, corresponding to the characteristic box, after the characteristic box is divided into boxes, the characteristic box names, namely, 50-59, are abstract characteristics of the insurance user, a large number of original insurance user characteristics in the box are updated, namely, target insurance user characteristics are updated, i-number of box data are corresponding to one box in the characteristic box, i-number of initial insurance user characteristics are respectively corresponding to one box, i-number of box data are partitioned together through box dividing operation, i-number of box data are generated, at this moment, i-number of initial insurance user characteristics are corresponding to one box names, i-number of the abstract characteristics are corresponding to one box, i-number of the characteristics of the user characteristics are also represented by the fact that i-number of box data are corresponding to one abstract characteristics are corresponding to one user characteristic, i-number of the initial user characteristics are represented by the box, i-number of the initial user characteristics are also represented by the abstract characteristics, namely, i-number of the user characteristics are represented by the box, i-number of the initial user characteristics are represented by the data, i-number of the box, namely, i-number of the user characteristics are represented by the abstract characteristics, and i-number of the user characteristics are represented by the corresponding to the initial user characteristics.

The feature processing apparatus provided in this specification will be described in detail with reference to fig. 6. The feature processing apparatus shown in fig. 6 is used to execute the method of the embodiment shown in fig. 1 to 5 of the present specification, and for convenience of explanation, only the portion relevant to the present specification is shown, and specific technical details are not disclosed, please refer to the embodiment shown in fig. 1 to 5 of the present specification.

Referring to fig. 6, a schematic structural diagram of the feature processing apparatus of the present specification is shown. The feature processing apparatus 1 may be implemented as all or part of a user terminal by software, hardware or a combination of both. According to some embodiments, the feature processing apparatus 1 includes an information acquisition module 11, a data determination module 12, and a feature screening module 13, specifically configured to:

an information acquisition module 11, configured to acquire at least one initial insurance user feature and insurance user risk information for an insurance transaction;

a data determining module 12, configured to determine a user feature risk-out broad table based on each of the initial insurance user features and the insurance user risk-out information, and determine at least one feature box of the user feature risk-out broad table;

and the feature screening module 13 is configured to obtain target insurance user features corresponding to the feature boxes respectively, determine an insurance feature evaluation index corresponding to the target insurance user features based on the box risk feature data of the feature boxes, and perform feature screening recommendation processing on the target insurance user features based on the insurance feature evaluation index.

Optionally, the data determining module 12 is configured to:

Carrying out data fusion processing on the initial insurance user characteristics and the insurance user insurance information to obtain an initial user characteristic insurance width table;

and carrying out characteristic box division processing on the initial user characteristic risk-free wide list to obtain the user characteristic risk-free wide list comprising at least one characteristic box body.

Optionally, the data determining module 12 is configured to:

acquiring a plurality of user characteristic information tables corresponding to the initial insurance user characteristics, and acquiring a user characteristic insurance base table corresponding to the insurance user insurance information;

and carrying out data association fusion processing on the user characteristic risk-out base table and the plurality of user characteristic information tables to obtain an initial user characteristic risk-out wide table.

Optionally, the data determining module 12 is configured to:

acquiring a feature field processing threshold value and a base table association threshold value corresponding to a data processing service, and determining feature field quantity and feature table quantity corresponding to all the user feature information tables;

based on the feature field quantity, the feature table quantity, the feature field processing threshold value and the base table association threshold value, carrying out data association fusion processing on the user feature risk-out base table and a plurality of user feature information tables, and carrying out distribution fusion association processing to obtain a plurality of reference user feature risk-out wide tables;

And carrying out table data fusion on the multiple reference user characteristic risk-out broad tables to obtain an initial user characteristic risk-out broad table.

Optionally, the data determining module 12 is configured to:

if the characteristic field quantity is larger than the characteristic field processing threshold value, determining a first segmentation quantity of the user characteristic risk-out base table;

if the number of the feature tables is larger than the base table association threshold, determining a second segmentation number of the user feature risk-out base table;

performing segmentation processing on the user characteristic risk-out base table based on the first segmentation number and/or the second segmentation number to obtain a third number of user characteristic risk-out base tables, and determining a plurality of user characteristic information tables associated with each user characteristic risk-out base table;

and carrying out information fusion processing on each user characteristic risk-out base table and the plurality of user characteristic information tables associated with the user characteristic risk-out base tables to obtain the third number of reference user characteristic risk-out wide tables.

Optionally, the data determining module 12 is configured to:

obtaining partition configuration information of the initial user characteristic risk-out wide table;

determining at least one feature partition corresponding to the initial user feature risk-out wide table based on the partition configuration information;

And carrying out feature data box-division processing on each feature partition to obtain a feature risk-out wide table containing at least one feature box body.

Optionally, the data determining module 12 is configured to:

acquiring a box-dividing characteristic field value;

and carrying out feature data box-division conversion processing on each feature partition based on the box-division feature field values to obtain at least one feature box body corresponding to each feature partition so as to generate a feature risk-out wide table containing at least one feature box body.

Optionally, the feature screening module 13 is configured to:

calculating a characteristic lifting index based on the box risk characteristic data of the characteristic box, and taking the characteristic lifting index as an insurance characteristic evaluation index corresponding to the target insurance user characteristic;

Optionally, the feature screening module 13 is configured to:

determining a target risk user proportion based on the box risk feature data of the feature box, and acquiring a total risk user proportion of a target insurance risk;

and taking the ratio of the target risk-giving user proportion to the total risk-giving user proportion as a characteristic lifting index.

Optionally, the feature screening module 13 is configured to:

acquiring an evaluation index threshold for the insurance feature evaluation index;

if the first evaluation index is matched with the evaluation index threshold value in the insurance feature evaluation indexes, acquiring a first insurance user feature corresponding to the first evaluation index, and performing feature recommendation processing on the first insurance user feature.

It should be noted that, in the feature processing apparatus provided in the foregoing embodiment, when executing the feature processing method, only the division of the foregoing functional modules is used as an example, in practical application, the foregoing functional allocation may be performed by different functional modules, that is, the internal structure of the device is divided into different functional modules, so as to complete all or part of the functions described above. In addition, the feature processing device and the feature processing method provided in the foregoing embodiments belong to the same concept, which embody detailed implementation procedures in the method embodiments, and are not described herein again.

The foregoing description is provided for the purpose of illustration only and does not represent the advantages or disadvantages of the embodiments.

The present disclosure further provides a computer storage medium, where the computer storage medium may store a plurality of instructions, where the instructions are adapted to be loaded by a processor and execute the feature processing method according to the embodiment shown in fig. 1 to 5, and the specific execution process may refer to the specific description of the embodiment shown in fig. 1 to 5, which is not repeated herein.

The present disclosure further provides a computer program product, where at least one instruction is stored, where the at least one instruction is loaded by the processor and executed by the processor, where the specific execution process may refer to the specific description of the embodiment shown in fig. 1 to 5, and details are not repeated herein.

Referring to fig. 7, a block diagram of an electronic device according to an exemplary embodiment of the present disclosure is shown. The electronic device in this specification may include one or more of the following: processor 110, memory 120, input device 130, output device 140, and bus 150. The processor 110, the memory 120, the input device 130, and the output device 140 may be connected by a bus 150.

Processor 110 may include one or more processing cores. The processor 110 utilizes various interfaces and lines to connect various portions of the overall electronic device, perform various functions of the electronic device 100, and process data by executing or executing instructions, programs, code sets, or instruction sets stored in the memory 120, and invoking data stored in the memory 120. Alternatively, the processor 110 may be implemented in at least one hardware form of digital signal processing (digital signal processing, DSP), field-programmable gate array (field-programmable gate array, FPGA), programmable logic array (programmable logic Array, PLA). The processor 110 may integrate one or a combination of several of a central processor (central processing unit, CPU), an image processor (graphics processing unit, GPU), and a modem, etc. The CPU mainly processes an operating system, a user interface, an application program and the like; the GPU is used for being responsible for rendering and drawing of display content; the modem is used to handle wireless communications. It will be appreciated that the modem may not be integrated into the processor 110 and may be implemented solely by a single communication chip.

The memory 120 may include a random access memory (random Access Memory, RAM) or a read-only memory (ROM). Optionally, the memory 120 includes a non-transitory computer readable medium (non-transitory computer-readable storage medium). Memory 120 may be used to store instructions, programs, code, sets of codes, or sets of instructions. The memory 120 may include a stored program area and a stored data area, wherein the stored program area may store instructions for implementing an operating system, which may be an Android (Android) system, including an Android system-based deep development system, an IOS system developed by apple corporation, including an IOS system-based deep development system, or other systems, instructions for implementing at least one function (such as a touch function, a sound playing function, an image playing function, etc.), instructions for implementing various method embodiments described below, and the like. The storage data area may also store data created by the electronic device in use, such as phonebooks, audiovisual data, chat log data, and the like.

Referring to FIG. 8, the memory 120 may be divided into an operating system space in which the operating system is running and a user space in which native and third party applications are running. In order to ensure that different third party application programs can achieve better operation effects, the operating system allocates corresponding system resources for the different third party application programs. However, the requirements of different application scenarios in the same third party application program on system resources are different, for example, under the local resource loading scenario, the third party application program has higher requirement on the disk reading speed; in the animation rendering scene, the third party application program has higher requirements on the GPU performance. The operating system and the third party application program are mutually independent, and the operating system often cannot timely sense the current application scene of the third party application program, so that the operating system cannot perform targeted system resource adaptation according to the specific application scene of the third party application program.

In order to enable the operating system to distinguish specific application scenes of the third-party application program, data communication between the third-party application program and the operating system needs to be communicated, so that the operating system can acquire current scene information of the third-party application program at any time, and targeted system resource adaptation is performed based on the current scene.

Taking an operating system as an Android system as an example, as shown in fig. 9, a program and data stored in the memory 120 may be stored in the memory 120 with a Linux kernel layer 320, a system runtime library layer 340, an application framework layer 360 and an application layer 380, where the Linux kernel layer 320, the system runtime library layer 340 and the application framework layer 360 belong to an operating system space, and the application layer 380 belongs to a user space. The Linux kernel layer 320 provides the underlying drivers for various hardware of the electronic device, such as display drivers, audio drivers, camera drivers, bluetooth drivers, wi-Fi drivers, power management, and the like. The system runtime layer 340 provides the main feature support for the Android system through some C/c++ libraries. For example, the SQLite library provides support for databases, the OpenGL/ES library provides support for 3D graphics, the Webkit library provides support for browser kernels, and the like. Also provided in the system runtime library layer 340 is a An Zhuoyun runtime library (Android run) which provides mainly some core libraries that can allow developers to write Android applications using the Java language. The application framework layer 360 provides various APIs that may be used in building applications, which developers can also build their own applications by using, for example, campaign management, window management, view management, notification management, content provider, package management, call management, resource management, location management. At least one application program is running in the application layer 380, and these application programs may be native application programs of the operating system, such as a contact program, a short message program, a clock program, a camera application, etc.; and may also be a third party application developed by a third party developer, such as a game-like application, instant messaging program, photo beautification program, etc.

Taking an operating system as an IOS system as an example, the programs and data stored in the memory 120 are shown in fig. 10, the IOS system includes: core operating system layer 420 (Core OS layer), core service layer 440 (Core Services layer), media layer 460 (Media layer), and touchable layer 480 (Cocoa Touch Layer). The core operating system layer 420 includes an operating system kernel, drivers, and underlying program frameworks that provide more hardware-like functionality for use by the program frameworks at the core services layer 440. The core services layer 440 provides system services and/or program frameworks required by the application, such as a Foundation (Foundation) framework, an account framework, an advertisement framework, a data storage framework, a network connection framework, a geographic location framework, a sports framework, and the like. The media layer 460 provides an interface for applications related to audiovisual aspects, such as a graphics-image related interface, an audio technology related interface, a video technology related interface, an audio video transmission technology wireless play (AirPlay) interface, and so forth. The touchable layer 480 provides various commonly used interface-related frameworks for application development, with the touchable layer 480 being responsible for user touch interactions on the electronic device. Such as a local notification service, a remote push service, an advertisement framework, a game tool framework, a message User Interface (UI) framework, a User Interface UIKit framework, a map framework, and so forth.

Among the frameworks illustrated in fig. 10, frameworks related to most applications include, but are not limited to: the infrastructure in core services layer 440 and the UIKit framework in touchable layer 480. The infrastructure provides many basic object classes and data types, providing the most basic system services for all applications, independent of the UI. While the class provided by the UIKit framework is a basic UI class library for creating touch-based user interfaces, iOS applications can provide UIs based on the UIKit framework, so it provides the infrastructure for applications to build user interfaces, draw, process and user interaction events, respond to gestures, and so on.

The manner and principle of implementing data communication between the third party application program and the operating system in the IOS system may refer to the Android system, and this description is not repeated here.

The input device 130 is configured to receive input instructions or data, and the input device 130 includes, but is not limited to, a keyboard, a mouse, a camera, a microphone, or a touch device. The output device 140 is used to output instructions or data, and the output device 140 includes, but is not limited to, a display device, a speaker, and the like. In one example, the input device 130 and the output device 140 may be combined, and the input device 130 and the output device 140 are a touch display screen for receiving a touch operation thereon or thereabout by a user using a finger, a touch pen, or any other suitable object, and displaying a user interface of each application program. Touch display screens are typically provided on the front panel of an electronic device. The touch display screen may be designed as a full screen, a curved screen, or a contoured screen. The touch display screen can also be designed to be a combination of a full screen and a curved screen, and a combination of a special-shaped screen and a curved screen is not limited in this specification.

In addition, those skilled in the art will appreciate that the configuration of the electronic device shown in the above-described figures does not constitute a limitation of the electronic device, and the electronic device may include more or less components than illustrated, or may combine certain components, or may have a different arrangement of components. For example, the electronic device further includes components such as a radio frequency circuit, an input unit, a sensor, an audio circuit, a wireless fidelity (wireless fidelity, wiFi) module, a power supply, and a bluetooth module, which are not described herein.

In this specification, the execution subject of each step may be the electronic device described above. Optionally, the execution subject of each step is an operating system of the electronic device. The operating system may be an android system, an IOS system, or other operating systems, which is not limited in this specification.

The electronic device of the present specification may further have a display device mounted thereon, and the display device may be various devices capable of realizing a display function, for example: cathode ray tube displays (cathode ray tubedisplay, CR), light-emitting diode displays (light-emitting diode display, LED), electronic ink screens, liquid crystal displays (liquid crystal display, LCD), plasma display panels (plasma display panel, PDP), and the like. A user may utilize a display device on electronic device 101 to view displayed text, images, video, etc. The electronic device may be a smart phone, a tablet computer, a gaming device, an AR (Augmented Reality ) device, an automobile, a data storage device, an audio playing device, a video playing device, a notebook, a desktop computing device, a wearable device such as an electronic watch, electronic glasses, an electronic helmet, an electronic bracelet, an electronic necklace, an electronic article of clothing, etc.

In the electronic device shown in fig. 7, where the electronic device may be a terminal, the processor 110 may be configured to invoke an application program stored in the memory 120 and specifically perform the following operations:

In one embodiment, the processor 110, when executing the determining a user characteristic risk profile based on each of the initial insurance user characteristics and the insurance user risk information, determines at least one characteristic box of the user characteristic risk profile, performs the following steps:

In one embodiment, the processor 110 performs the data fusion process on each of the initial insurance user features and the insurance user risk information to obtain an initial user feature risk width table, and performs the following steps:

In one embodiment, the processor 110 performs the data association fusion processing on the user feature risk base table and the plurality of user feature information tables to obtain an initial user feature risk width table, and performs the following steps:

In one embodiment, the processor 110 performs the data association fusion processing on the user feature risk base table and the plurality of user feature information tables to obtain a plurality of reference user feature risk broad tables after performing the data association fusion processing on the base table based on the feature field quantity, the feature table quantity, the feature field processing threshold and the base table association threshold, and performs the following steps:

In one embodiment, the processor 110 performs the feature binning process on the initial user feature risk width table to obtain a feature risk width table including at least one feature bin, and performs the following steps:

In one embodiment, the processor 110 performs the feature data binning process on each of the feature partitions to obtain a feature risk score table including at least one feature bin, and performs the following steps:

acquiring a box-dividing characteristic field value;

In one embodiment, the processor 110 performs the following steps when executing the acquiring the target insurance user features corresponding to the feature boxes respectively:

In one embodiment, the processor 110, when executing the determining the insurance feature evaluation index corresponding to the target insurance user feature based on the box risk feature data of the feature box, executes the following steps:

In one embodiment, the processor 110, when executing the computing of the feature boost index based on the tank risk feature data of the feature tank, includes:

In one embodiment, the processor 110 performs the following steps when performing the feature screening recommendation process on each of the target insurance user features based on the insurance feature evaluation index:

Those skilled in the art will appreciate that implementing all or part of the above-described methods in accordance with the embodiments may be accomplished by way of a computer program stored on a computer readable storage medium, which when executed may comprise the steps of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a read-only memory, a random access memory, or the like.

It should be noted that, information (including but not limited to user equipment information, user personal information, etc.), data (including but not limited to data for analysis, stored data, presented data, etc.), and signals according to the embodiments of the present disclosure are all authorized by the user or are fully authorized by the parties, and the collection, use, and processing of relevant data is required to comply with relevant laws and regulations and standards of relevant countries and regions. For example, the insurance user features, insurance user risk information, etc. referred to in this specification are all acquired with sufficient authorization.

The foregoing disclosure is only illustrative of the preferred embodiments of the present invention and is not to be construed as limiting the scope of the claims, which follow the meaning of the claims of the present invention.

Claims

1. A method of feature processing, the method comprising:

2. The method of claim 1, the determining a user characteristic risk profile based on each of the initial insurance user characteristics and the insurance user risk information, and determining at least one characteristic bin of the user characteristic risk profile, comprising:

3. The method of claim 2, wherein the data fusion processing is performed on each of the initial insurance user features and the insurance user risk information to obtain an initial user feature risk width table, and the method comprises:

4. The method according to claim 3, wherein the data association fusion processing is performed on the user feature risk base table and the plurality of user feature information tables to obtain an initial user feature risk width table, and the method comprises the following steps:

5. The method of claim 4, wherein the performing data association fusion processing on the user feature risk base table and the plurality of user feature information tables to perform allocation fusion association processing on the basis of the feature field quantity, the feature table quantity, the feature field processing threshold and the base table association threshold to obtain a plurality of reference user feature risk broad tables comprises:

6. The method of claim 2, wherein the feature binning the initial user feature risk width table to obtain a feature risk width table comprising at least one feature bin, comprises:

7. The method of claim 6, wherein the performing feature data binning on each of the feature partitions to obtain a feature risk profile including at least one feature box comprises:

acquiring a box-dividing characteristic field value;

8. The method of claim 1, wherein the obtaining the target insurance user features respectively corresponding to the feature boxes comprises:

9. The method of claim 1, wherein determining an insurance feature evaluation index corresponding to the target insurance user feature based on the tank risk feature data of the feature tank comprises:

10. The method of claim 9, the computing a feature boost indicator based on tank risk feature data for the feature tank, comprising:

11. The method of claim 1, wherein the performing feature screening recommendation processing on each of the target insurance user features based on the insurance feature evaluation index comprises:

12. A feature processing apparatus, the apparatus comprising:

13. A computer storage medium storing a plurality of instructions adapted to be loaded by a processor and to perform the method steps of any one of claims 1 to 11.

14. A computer program product storing at least one instruction for loading by a processor and performing the method steps of any one of claims 1 to 11.

15. An electronic device, comprising: a processor and a memory; wherein the memory stores a computer program adapted to be loaded by the processor and to perform the method steps of any of claims 1-11.