AU2018100566A4 - Decentralized financial intelligence based on decentralized consensus and anonymized transaction history - Google Patents

Abstract

DECENTRALIZED FINANCIAL INTELLIGENCE BASED ON DECENTRALIZED CONSENSUS AND ANONYMIZED TRANSACTION HISTORY A computerized transaction scoring system may include a decentralized network of multiple transaction scoring analytic nodes (computers of any kind) operatively coupled over a computerized network. The computerized transaction scoring system may include a distributed file system for storage of anonymized transaction history this file system will be capable of analysing transaction data to store unalterable data, remove duplicated files across the network, and obtain hashedversion information for accessing storage nodes to search for files in the network. The computerized transaction scoring system may also include automated consensus algorithm of any kind to identify faulty nodes or clusters and flag them as faulty to ensure the trust in between the nodes. The nodes in computerized transaction scoring system may include predictive analytics algorithms to score each new transaction based on the history of existing transactions in the file system. Transaction Origin Bank / Insttton Score Score Aggregation and --- --- --- Aggregation and Consensus Consensus Anltc Aayd ntcAalytic Analytc .-lti Noe Nd oe Nd ode Nod Distributed Storage Layer Figure 2

Description

FIELD OF INVENTION

[0001] The present invention relates generally to credit scoring systems and transaction scoring systems and methods of assessing and rating a consumer credibility or a consumer engagement in the transaction and, in particular, to a computer implemented decentralized network and system for rating a consumer or a transaction dynamically based on up to date data. Typically, any transaction between any two parties.

BACKGROUND OF THE INVENTION

[0002] Growth in the electronic payments sector has surpassed general economic growth and growth in other financial sectors. Electronic payments include credit, debit, and other electronic instruments used to transfer payments from consumers to merchants.

[0003] The growth in the electronic payments sector is accompanied by numerous economic and transactional benefits. Electronic payments improve economic inefficiencies, make payments more secure and convenient, and, as a corollary to the lessons learned from micro finance, provide the force for further economic and social development.

[0004] For developing countries, those gains could be significant, but they would depend on the concurrent development of the appropriate network and payments infrastructure, government regulation, consumer education, and competition within the sector.

[0005] As governments in developed economies have learned, adequate regulatory oversight in the electronic payments sector is essential to maintaining financial stability, consumer confidence, and data privacy and security of the sector.

[0006] Although electronic payments growth could represent an opportunity for developing countries to rebalance their economies by encouraging domestic consumption and it is essential to educate consumers to use electronic payments responsibly and securely.

[0007] Financial fraud continues to grow faster than consumer spending. Data breaches have resulted in more card details being compromised, and the growth in online shopping has led to more opportunities for ecommerce fraud.

[0008] Fraudulent transactions are the ones that are not initiated by the account holder. Hackers may obtain sensitive information about the card or the account and could make a transaction on behalf of the card user.

[0009] In order to decrease these losses, the financial industry has developed many different strategies that -predict fraud using data analytics, predictive analytics, artificial intelligence and machine learning.

SUMMARY OF INVENTION

[0010] The present invention is to provide decentralised transaction scores, which provide a dynamic marker for the customer engagement in a transaction. It is another objective of the present invention is to provide a method and system to generate the transaction scores in a decentralised way.

[0011] The present invention is to provide a system to identify the customer using hashes generated on the unique personal details which there by providing security. It is still another objective of the present invention to provide a system to assess the consumer rating transaction by transaction, in real-time, based on the previous transaction history and alternative data.

[0012] The present invention is directed to a computer-implemented method of rating a transaction initiated by the consumer. This method includes the steps of: (a) providing the unique customer hash as a part of the transaction, (b) providing the transaction data set including the details of the transaction, (c) providing the details of the origin, type and account hash of the transaction (d) calculating the transaction score based upon a scoring formula utilizing at least one data field value from at least one of the provided transaction data set and (e) presenting the transaction score to the card/account issuer, (f) The scoring formula is executed by many computers which are interconnected with a computer network, and these computers come to an agreement (consensus) about the transaction score. In addition, the computer implemented method described above is initiated substantially at the same time the credit transaction is initiated. Therefore, the presently invented method is performed in a real-time or dynamic process.

[0013] The present invention is also able to include a distributed file system for storage of anonymized transaction history this file system will be capable of analysing transaction data to store unalterable data, remove duplicated files across the network, and obtain address information for accessing storage nodes to search for files in the network.

[0014] The present invention is also able to include an automated consensus algorithm of any kind to identify faulty nodes or clusters and flag them as faulty to ensure the trust in between the nodes. The nodes in computerized transaction scoring system may include predictive analytics algorithms to score each new transaction based on the history of existing transactions in the file system.

[0015] The present invention, both as to its construction and its method of operation, together with the additional objects and advantages thereof, will best be understood from the following description of exemplary. Embodiments when read in connection with the accompanying drawings.

TECHNICAL PROBLEM

[0016] The global financial intelligence is centralised, incomplete, and not portable as they are specific to a country or a region. Along with the incompleteness it also has security risks. There have been multiple data breaches and hacks which has resulted in significant loss of money and sensitive information like identity and personal details of the customers.

[0017] As information becomes more centralized it becomes monopolised and incomplete. This leads to decisions being made without all the available information at hand, significantly increasing the associated risk. Moreover, the credit scores are not updated in real time, with the delay prejudicing millions of consumers and businesses as their current credit history is not factored into the decision-making process.

[0018] Lack of live transaction scores is the major problem the financial industry, by analysing and scoring the transactions in real time helps the institutions to minimize losses and allows them to take well informed decisions.

SOLUTION TO THE PROBLEM

[0019] To combat fraud at a global scale we need a decentralized system with high availability and providing live risk scores for transactions. Storing the anonymous transaction data securely on a decentralized data network and building a shared layer to access this network on a block chain will create an entirely new set of possibilities for Al capabilities and insights.

[0020] The benefits of decentralized/shared control, particularly as a foundation for Al results in more data, thus improved modelling capabilities and Qualitatively new data leading to entirely new models.

[0021] Such an open network will yield many interesting by-products such as Credit scores providing financial inclusion for Credit invisibles. Transaction scores to predict transaction fraud, Shared Global Intelligence which would make the risk models portable especially online transaction patterns and eventually becoming a playground for data scientists, which will encourage them to mine the up to date data and even come up with revolutionary risk models to safeguard the global community from fraud.

ADVANTAGEOUS EFFECTS OF THE INVENTION

[0022] The present invention will have the following positive effects to the existing industry problems, 1. Secures the identity information from any possible data breaches, because the identity is established only with the help of unique customer hash or hashes, which cannot be reverse engineered to obtain any of the customer's sensitive information. 2. By making the identity secure from (point 1), all the transaction data which will be received by the present invention (a decentralised system) will be anonymized, this provides another level of security and protection from any possible data breaches, as the data is useless as it cannot be linked to a particular individual or entity. 3. By providing a decentralised computerized consensus mechanism in providing the transaction scores, the present invention will assure that the decisions which were made are reliable and accurate. Consensus mechanism eradicates the necessity of human interaction in validating the results, there by providing a layer of trust which is completely independent and un-influenced. 4. The present invention brings a lot of value to the financial system as it will be able to predict and detect fraud in real time and is reported, unlike the existing methodologies where the fraud is realised only after it has happened. 5. The present invention also provides modularity and extensibility for the data models, where each data set can be independently scored by various models and they consent without any human interaction. This also means that the models can be dynamically adjusted and updated automatically based on the previous transaction.

BRIEF DESCRIPTION OF DRAWINGS

[0023] Fig. 1 is a flow diagram of computerised anonymization of the customer identity information according to the present invention.

[0024] Fig. 2 is a flow diagram of the computerised decentralised transaction scoring, with a consensus mechanism according to the present invention.

DESCRIPTION OF EMBODIMENTS

[0025] The present invention is a method of decentralised scoring of a transaction in real-time with a consensus mechanism using anonymised customer information. The method of present invention about ananomysing the customer information is illustrated with the help of Figure 1. The present invention also involves a computerized generation of real-time transaction scores with consensus mechanism is illustrated in Figure 2.

[0026] As per the present invention, Step 10 in Figure 1 illustrates the collection of unique details of a customer, including but not limited to details like, Name, date of birth, Social security number / Tax file number, Account Number.

[0027] The Step 11 in Figure 1, represents a hashing function, The input to this function are the unique details of the customer as represented in Step 10. this hash function is any function that can be used to map data of arbitrary size to data of fixed size, in this case the fixed size customer identity details. This function generates the hashes of the pairs of the unique customer identity information details, for example Hash(Name, date of birth), Hash(Name, Social security number) and so on.

[0028] The Step 12 in Figure 1 is the output of the hash function, all the hashes of the pairs of unique customer identity information as explained in the Step 11.

[0029] As depicted in Figure 2, the present invention is deployed to a computerised system consisting of multiple interconnected computers sharing a distributed file system for storage of anonymized transaction history this file system will be capable of analysing transaction data to store unalterable data, remove duplicated files across the network, and obtain address information for accessing storage nodes to search for files in the network.

[0030] The system 20, is the originator of the financial transaction, this origin may be a point of sale, a computer for online transaction including but not limited to a funds transfer request, loan request, online payment request, crypto currency transfer request. This transaction is then received by the system 21.

[0031] The system 21 represents the authority to process the transaction, which issued the account or instrument of the transaction originator, typically this could be a financial institution like bank, credit agency, crypto bank, lending institutions. This system requests the method 22 (the entry point to the present invention) to get a score of the transaction. Due to the nature of large customer base of the system 21, the system expects multiple simultaneous transaction scoring requests coming in to the present invention, typically several thousand requests per second. The present invention should and will scale to process those levels of transactions per second.

[0032] The method 22 is responsible for the distribution of the transaction across the decentralised processing nodes, there may be multiple nodes which are interconnected, we can call these master nodes which aggregatesand come to aconsensus with other master nodes and also keep an account of the performance statistics of the nodes represented as a group in system 23. The system 23 shall also record the processed transaction in the storage layer which is the method 24.

[0033] Further the system 23 is a group of computers which are interconnected and independently generate the score for a transaction, the scoring algorithm includes predictive analytic models which rely on the data models built on the transaction history of a unique customer hash.

[0034] The method 24 is implemented as a distributed file system storage of anonymized transaction history, this file system will be capable of analysing transaction data to store unalterable data, remove duplicated files across the network, and obtain an immutable version information and acts as a swarm database.

[0035] This invention has been described with reference to the preferred embodiments. Obvious modifications and alterations will occur to others upon reading and understanding the preceding detailed description. It is intended that the invention be construed as including all such modifications and alteration.

INDUSTRIAL APPLICABILITY

[0036] This present invention has been described with reference to the preferred embodiments can be applied to provide a wide range of analytic services like Up to date Credit Scores, Real time Transaction Scores to predict fraud, Business credit scores. The beneficiaries of this invention are Banks, Financial Institutions, Lenders, Crypto Banks, Credit unions.

[0037] Apart from the above services, the present invention shall also provide real time insights about transaction patterns and fraud patterns to thedata scientists. This helps them to build realistic and up to date portable data risk models based on the shared global financial intelligence.

Claims (5)

1. A computerized method of scoring a transaction initiated by the consumer including the following steps: a. providing the unique customer hash as a part of the transaction. b. providing the transaction data set including the details of the transaction. c. providing the details of the origin, type and account hash of the transaction d. calculating the transaction score based upon a scoring formula by various independent computers utilizing at least one data field value from at least one of the provided transaction data set, and the computers come to an agreement (consensus) about the transaction score and presenting them in real-time. e. presenting the transaction score to the card/account issuer.

2. The process of claim 1 is performed by a computing device and includes one or more steps in claim 1.

3. The process of claim 1 uses a hash generated by a hashing algorithm, which uniquely identifies a customer.

4. The process of claim 1, where the transaction scores are generated using statistic alanalytic models on the transaction history independently analysed and scored by multiple computing devices interconnected over a network and share a distributed file system which is capable of store unalterable data, remove duplicated files and provides versioning ability to the transaction history.

5. The process of claim 1, where a transaction is scored by multiple computing devices will also facilitate a decentralised consensus mechanism in which all the devices will come to a consensus with the help of a consensus algorithm without any external influence and human interaction.