TWM678596U - Data sharing and analysis system - Google Patents

Abstract

A data sharing analysis system includes a processing unit and a storage unit. The processing unit is configured to: receive a communication data to be verified; determine whether the communication data to be verified matches one of multiple customer communication data; when a match is determined, obtain related party characteristic data and shared characteristic data based on the basic data of the target customer and the basic data of a reference customer corresponding to the matching reference customer communication data; generate a data sharing assessment result using an intelligent decision-making model based on the related party characteristic data, the shared characteristic data, and the reference customer risk data corresponding to the basic data of the reference customer corresponding to the matching reference customer communication data; and transmit a message corresponding to the data sharing assessment result to the client's electronic device.

Description

Data sharing and analysis system

This invention relates to an analysis system, and more particularly to a data sharing analysis system.

With the rapid development of digital transactions, fraud and identity theft incidents are on the rise, posing challenges to Know Your Customer (KYC) processes. While current systems can identify whether emails or phone numbers submitted by customers overlap with or are shared with other records in the database, they struggle to further analyze the actual relationships between these sharers (e.g., whether they are relatives, businesses, or associated with blacklists), thus failing to accurately determine the legitimacy of such sharing. This deficiency may increase fraud risks or reduce the user experience for legitimate customers due to excessive scrutiny.

Therefore, how to develop a system that automatically analyzes common causes and detects anomalies has become the subject of this invention.

Therefore, the purpose of this invention is to provide a data sharing and analysis system that can overcome the shortcomings of prior art.

This novel data sharing and analysis system is applicable to communication with a client electronic device operated by a target customer. It includes a processing unit and a storage unit electrically connected to the processing unit. The processing unit is electrically connected to the client electronic device. The storage unit stores basic target customer data corresponding to the target customer, multiple sets of basic reference customer data corresponding to multiple reference customers, and multiple sets of reference customer profiles corresponding to the reference customers. The data includes risk data and a smart decision-making model implemented using machine learning algorithms. Each reference customer's basic data includes reference customer communication data. The processing unit receives a data change request from the customer's electronic device. The data change request includes verification communication data corresponding to the target customer. Based on the data change request, the processing unit determines whether the verification communication data matches one of the customer's communication data. When the verification request is received, the processing unit determines whether the verification communication data matches one of the customer's communication data. When the processing unit determines that the communication data to be verified matches one of the reference customer communication data, it obtains related party characteristic data and common characteristic data based on the target customer's basic data and the basic data of the corresponding reference customer for the matching reference customer communication data. The related party characteristic data indicates the relationship information between the target customer and the corresponding reference customer, and the common characteristic data indicates that the communication data to be verified is shared. The processing unit uses the intelligent decision-making model to generate a data sharing assessment result based on the related party's characteristic data, the shared characteristic data, and the reference customer's basic information corresponding to the corresponding reference customer's communication data and risk data. The processing unit then transmits a message corresponding to the data sharing assessment result to the client's electronic device, indicating whether the data change request is allowed.

In some embodiments of this novel data sharing analysis system, the intelligent decision-making model includes multiple decision trees, each of which outputs a trust score. For each trust score, the processing unit generates a decision result corresponding to that trust score based on the trust score and a first preset rule. Based on these decision results and a second preset rule, the processing unit selects a target decision result from these decision results as the data sharing evaluation result.

In some implementations of this novel data sharing and analysis system, each reference customer risk data includes a high-risk eigenvalue, a historical risk eigenvalue, and a time factor eigenvalue.

In some embodiments of this novel data sharing and analysis system, the related party characteristic data includes a table of personal and legal person relationships, a parent-child account table, a legal representative relationship table, and a trusted contact table.

In some embodiments of this novel data sharing analysis system, the shared feature data includes a blacklist shared feature value, a family shared feature value, and a device shared feature value.

The advantages of this new system are as follows: When a user changes their email or phone number, the system automatically determines whether the entered email or phone number is shared by other users. It then uses a machine learning algorithm to output multiple trust scores to determine the relationship between the user and the sharer (e.g., whether they are relatives, businesses, or associated with a blacklist), thus reducing the risk of misjudgment and fraud. Furthermore, the system integrates these trust scores to generate a data sharing assessment result and sends the corresponding message to the client device to remind the user whether the change of email or phone number requires further enhanced KYC verification.

Before this invention is described in detail, it should be noted that, unless otherwise defined, "electrically connected" as used in this patent specification refers to the "coupled" relationship between computer hardware (such as electronic systems, devices, apparatuses, units, components), and generally refers to "wired electrical connections" that are physically connected to each other through conductive/semiconductor materials, as well as "radio connections" that utilize wireless communication technologies (such as, but not limited to, wireless networks, Bluetooth, and electromagnetic induction) to achieve wireless data transmission. On the other hand, unless otherwise defined, the term "electrical connection" as used in this patent specification also refers to "direct electrical connection" which is achieved by directly coupling multiple computer hardware components to each other, and "indirect electrical connection" which is achieved by indirectly coupling multiple computer hardware components to each other through other computer hardware components.

Before this invention is described in detail, it should be noted that the term "unit" in this patent specification refers to computer hardware, not software. For example, "processing unit" is used to represent computer hardware with data processing capabilities. On the other hand, the term "unit" in this patent specification can refer to a single piece of computer hardware with a specific function, or it can refer to a group of computer hardware with similar functions. For example, "processing unit" can refer to a single processor with data processing capabilities, but it can also refer to a collection of processors.

Referring to Figure 1, one embodiment of the novel data sharing and analysis system 100 is managed by a financial institution (e.g., a bank) and is adapted to communicate with a client electronic device 200 operated by a target customer. The client electronic device 200 may be, for example, a mobile phone, a tablet computer, a laptop computer, or a desktop computer.

In this embodiment, the data sharing analysis system 100 is a server device, and the data sharing analysis system 100 includes a processing unit 1 and a storage unit 2 electrically connected to the processing unit 1. The processing unit 1 is adapted to be electrically connected to the client electronic device 200 via a network, thereby communicating with the client electronic device 200.

More specifically, in this embodiment, the processing unit 1 is a processor implemented with integrated circuits and having data processing and instruction transceiver functions, and the storage unit 2 is a data storage device (such as a hard disk or other types of computer-readable recording media) for storing digital data. However, in similar embodiments, the processing unit 1 may also be a circuit component including a processor and a circuit board, and the storage unit 2 may be a collection of multiple storage devices of the same or different types. Furthermore, in other embodiments, the data sharing analysis system 100 can also be implemented as multiple server devices electrically connected to each other. In this case, the processing unit 1 can be implemented as a collection of multiple processors/circuit components respectively possessed by the server devices, and the storage unit 2 can be implemented as a collection of multiple storage devices respectively possessed by the server devices. Based on the above, the actual implementation of the data sharing analysis system 100 in terms of computer hardware is not limited to this embodiment.

In this embodiment, the storage unit 2 stores a target customer basic data 21 corresponding to the target customer, multiple reference customer basic data 22 corresponding to multiple reference customers, multiple reference customer risk data 23 corresponding to the reference customers, and a smart decision-making model M implemented using a machine learning algorithm.

The target customer's basic information 21 includes a customer identification code related to a natural person or legal entity.

Each reference customer basic information 22 includes another customer identification number related to a natural or legal person, and a reference customer contact information. This reference customer contact information may include, for example, an email address and a telephone number. It should be noted that the customer identification number and the other customer identification number are used by the bank to distinguish different customers, and each customer is assigned a unique customer identification number by the bank.

Each reference client's risk data 23 in this embodiment includes a high-risk eigenvalue, a historical risk eigenvalue, and a time factor eigenvalue. The high-risk eigenvalue indicates whether the corresponding reference client is on a blacklist (e.g., 1 represents that the reference client is on a blacklist, and 0 represents that the reference client is not on a blacklist). The historical risk eigenvalue indicates a comprehensive quantitative indicator calculated based on the number of times the reference client has entered a high-risk state (such as a blacklist), the duration, and the recidivism frequency, used to reflect the historical risk trajectory of the reference client. The time factor eigenvalue indicates the level and duration of the reference client's high-risk status. The calculated comprehensive quantitative indicator is used to reflect the changes in the reference client's risk level.

In this embodiment, the intelligent decision-making model M is implemented using a machine learning algorithm, and it includes multiple decision trees. These decision trees, in this embodiment, may be, for example, a blacklist sharing decision tree for comparing whether a data sharer is associated with a blacklist, a related person sharing decision tree for comparing whether a data sharer is a relative, and a device sharing decision tree for determining whether a changed login device is shared with others to confirm whether the data sharer is associated with a blacklist or a relative, but are not limited to these. Each decision tree is trained to output a trust score.

More specifically, each decision tree is trained using a set of training input data and a set of training target data. In this embodiment, the training input data of the blacklist-sharing decision tree includes multiple high-risk eigenvalues, multiple historical risk eigenvalues, and multiple time factor eigenvalues corresponding to multiple data sharers. The training target data of the blacklist-sharing decision tree corresponds to multiple trust scores of the multiple data sharers of the training input data, but is not limited to this. The training input data for the shared decision tree for related parties includes multiple kinship scores, multiple kinship shared eigenvalues, multiple personal and legal entity relationship tables, multiple parent-child account tables, multiple legal representative relationship tables, multiple trusted contact tables, and multiple historical risk eigenvalues, respectively. The training objective data for the shared decision tree for related parties corresponds to multiple trust scores of the multiple data sharers in the training input data, but is not limited to this. The training input data of the device-shared decision tree includes multiple device-shared feature values, multiple high-risk feature values, multiple kinship scores, multiple time factor feature values, and multiple historical risk feature values corresponding to multiple data sharers. The training target data of the device-shared decision tree corresponds to multiple trust scores of multiple data sharers of the training input data, but is not limited to these.

It should be further noted that in other embodiments, the intelligent decision-making model M may, for example, include only one or more of the blacklist shared decision tree, the related party shared decision tree, and the device shared decision tree. Furthermore, the decision tree described in this embodiment may be constructed, for example, using algorithms in the prior art. Since the construction details of the decision tree are not the focus of this patent specification, their details are not described here.

Referring to Figures 1 and 2, the following exemplarily illustrates how the data sharing analysis system 100 of this embodiment implements a data sharing analysis method.

First, in step S11, the processing unit 1 receives a data change request from the client electronic device 200. The data change request includes a verification communication data corresponding to the target customer, and the verification communication data is, for example, another email address or another phone number.

After the processing unit 1 receives the data change request, the process proceeds to step S12.

In step S12, the processing unit 1 determines whether the communication data to be verified matches one of the customer communication data based on the data change request. If the determination result is yes, the process proceeds to step S13; if the determination result is no, the process ends.

In step S13, the processing unit 1 obtains related party characteristic data and shared characteristic data based on the target customer's basic data 21 and the corresponding reference customer's basic data 22. The related party characteristic data indicates the relationship information between the target customer and the corresponding reference customer, and the shared characteristic data indicates the characteristic information shared by the communication data to be verified.

Specifically, in this embodiment, the related party characteristic data includes, but is not limited to, a personal-legal entity relationship table, a parent-child account table, a legal representative relationship table, and a trusted contact table. The personal-legal entity relationship table indicates the association between the natural person's identification number and the legal entity's account identification number (i.e., whether the target customer and the reference customer are in a natural person-legal entity account relationship). The parent-child account table indicates the parent-child relationship between the natural person's identification numbers (i.e., whether the target customer and the reference customer are in a parent-child relationship). The legal representative relationship table indicates the association between the legal representative's identification number and the legal entity's account identification number (i.e., whether the target customer and the reference customer are in a legal representative-legal entity account relationship). The trusted contact information form indicates the natural person's identifier and the relationship between the natural person and the trusted contact. On the other hand, the shared feature data in this embodiment includes a blacklist shared feature value, a family shared feature value, and a device shared feature value. The blacklist shared feature value indicates how many blacklisted customers share the email or phone number. The family shared feature value indicates how many family members share the email or phone number. The device shared feature value indicates how many electronic devices linked to the email or phone number have shared the email or phone number.

After processing unit 1 obtains the related party characteristic data and the shared characteristic data, the process proceeds to step S14.

In step S14, the processing unit 1 uses the intelligent decision-making model M to generate a data sharing assessment result based on the related party characteristic data, the shared characteristic data, and the reference customer basic data 22 corresponding to the matching reference customer communication data and the reference customer risk data 23.

In this embodiment, the processing unit 1 generates a decision result corresponding to the trust score based on the trust score and a first preset rule; then, the processing unit 1 selects a target decision result from the decision results as the data sharing evaluation result based on the decision results and a second preset rule.

In this embodiment, the first default rule is, for example, that when the trust score is greater than or equal to 0.8, the decision result is "pass"; when the trust score is greater than or equal to 0.5 and less than 0.8, the decision result is "fall into KYC"; and when the trust score is less than 0.5, the decision result is "reject". In this embodiment, the second default rule is, for example, "selecting the decision result with higher priority based on priority order: 'rejection' > 'under KYC' > 'pass'". Therefore, if the decision result corresponding to the blacklist shared decision tree is "rejection", the decision result corresponding to the related party shared decision tree is "under KYC", and the decision result corresponding to the device shared decision tree is "under KYC", then the target decision result can be, for example, "rejection".

After the data sharing evaluation result is generated by the processing unit 1, the process proceeds to step S15.

In step S15, the processing unit 1 transmits a message corresponding to the data sharing evaluation result to the client electronic device 200. The message indicates whether the data change request is allowed, and includes the trust score corresponding to the target decision result of the data sharing evaluation result.

In this embodiment, the message content may be, for example, "The system will automatically update your email/phone number, with a trust score of 0.9", "For security reasons, if you are unable to change your email/phone number, please contact customer service directly, with a trust score of 0.3", or "Please perform a biometric verification procedure, with a trust score of 0.5", and is not limited to this.

The above is an example of how the data sharing analysis system 100 in this embodiment implements the data sharing analysis method.

It should be specifically noted that steps S11 to S15 of this embodiment and the flowchart in Figure 2 are merely illustrative of one possible implementation of the data sharing analysis system 100 of this invention. It should be understood that even if steps S11 to S15 are merged, split, or rearranged, if the resulting process performs similar functions and yields similar results compared to this embodiment, it still falls within the implementable form of the data sharing analysis system 100 of this invention. Therefore, steps S11 to S15 of this embodiment and the flowchart in Figure 2 are not intended to limit the scope of implementation of this invention.

In summary, the data sharing analysis system 100 can automatically determine whether the email address or phone number entered by the user is shared by other users when the user changes their email address or phone number. It then uses a machine learning algorithm to output multiple trust scores to determine the relationship between the user and the sharer (e.g., whether they are relatives, businesses, or associated with a blacklist), thereby reducing the risk of misjudgment and fraud. Furthermore, the data sharing analysis system 100 integrates these trust scores to generate a data sharing assessment result and sends the corresponding information to the client electronic device 200 to remind the user whether the change of email address or phone number requires further enhanced KYC verification. Therefore, it effectively achieves the purpose of this novel system.

However, the above description is merely an embodiment of this utility model and should not be construed as limiting the scope of its implementation. Any simple equivalent changes and modifications made in accordance with the scope of the patent application and the contents of the patent specification shall still fall within the scope of this utility model.

100: Data Sharing and Analysis System 1: Processing Unit 2: Storage Unit 21: Target Customer Basic Data 22: Reference Customer Basic Data 23: Reference Customer Risk Data M: Intelligent Decision-Making Model 200: Client-Side Electronic Devices S11~S15: Steps

Other features and effects of this invention will be clearly shown in the embodiments with reference to the figures. Figure 1 is a schematic diagram of the hardware connections of an embodiment of the data sharing analysis system of this invention; and Figure 2 is a flowchart illustrating how the embodiment implements a data sharing analysis method.

100: Data Sharing and Analysis System

1: Processing Unit

2: Storage Unit

21: Target Customer Basic Information

22: Refer to customer basic information

23: Refer to customer risk data

M: Intelligent Decision-Making Model

200: Client Electronic Device

Claims (5)

A data sharing and analysis system is adapted to communicate with a client electronic device operated by a target customer, and includes: a processing unit adapted to be electrically connected to the client electronic device; a storage unit electrically connected to the processing unit, storing target customer basic data corresponding to the target customer, multiple sets of reference customer basic data corresponding to multiple reference customers, multiple sets of reference customer risk data corresponding to the reference customers, and a smart decision-making model implemented using a machine learning algorithm, each set of reference customer basic data including reference customer communication data; the processing unit receives a data change request from the client electronic device, the data change request including unverified communication data corresponding to the target customer; The processing unit determines whether the communication data to be verified matches one of the customer communication data based on the data change request. When the processing unit determines that the communication data to be verified matches one of the reference customer communication data, it obtains related party feature data and common feature data based on the target customer's basic data and the basic data of the reference customer corresponding to the matching reference customer communication data. The related party feature data indicates the relationship information between the target customer and the corresponding reference customer, and the common feature data indicates the feature information shared by the communication data to be verified. The processing unit uses the intelligent decision-making model to generate a data sharing assessment result based on the related party characteristic data, the shared characteristic data, and the reference customer risk data corresponding to the reference customer's basic information and corresponding to the reference customer's communication data. The processing unit then sends a message corresponding to the data sharing assessment result to the client's electronic device, indicating whether the data change request is allowed. The data sharing analysis system as described in claim 1, wherein the intelligent decision-making model includes multiple decision trees, each decision tree being used to output a trust score; for each trust score, the processing unit generates a decision result corresponding to the trust score based on the trust score and a first preset rule; the processing unit selects a target decision result from the decision results as the data sharing evaluation result based on the decision results and a second preset rule. The data sharing analysis system as described in claim 1, wherein each reference customer risk data includes a high-risk eigenvalue, a historical risk eigenvalue, and a time factor eigenvalue. The data sharing analysis system as described in claim 1, wherein the related party characteristic data includes a table of personal and legal person relationships, a parent-child account table, a legal representative relationship table, and a trusted contact table. The data sharing analysis system as described in claim 1, wherein the shared feature data includes a blacklist shared feature value, a family shared feature value, and a device shared feature value.