At Metamap, we've developed a cutting-edge system for identifying the purpose and context of bank transactions. By using a combination of machine learning algorithms and expert knowledge, we've created a system that surpasses both traditional manual analysis by credit specialists and machine learning-only methods used by personal finance management apps. Our system is more efficient and accurate, thanks to the high quality of our training data.
The machine learning algorithms developed by MetaMap transforms raw transaction data into applicable insights.
Three-level hierarchical system
We understand the importance of having a clear understanding of your customer's data, which is why we offer both a broad overview and a detailed, granular examination, with each transaction being categorized into three distinct levels of granularity.
Automate Decision Process
With categorisation, you can accelerate the entire decision process as there is no need to do manually go through each and every transaction description
The proprietary categorization engine serves as the foundation for our system. It utilizes a three-level categorization, with level 0 i.e. Transaction Type serving as the highest level and having the ability to have level 1 i.e. Parent category and level 1 having the ability to have level 2 Child Category. Each transaction is assigned three levels, and the final category is always assigned to the lowest possible level.
With over 125 unique categories across all levels for both Inflow and outflow, it's crucial to utilize the full category tree hierarchy when analyzing a bank account to extract the most valuable information.
Here's a breakdown of the different levels of information we provide:
This categorization method identify the nature of a financial transaction, classifying it as income, expense, transfer, refund, loan deposit, penalty, investment, or bank charges. This information gives you a big picture view of your customer's financial cash flow.
This level delves deeper into the reasoning behind the money coming in and going out. For example, was that deposit a salary, pension, or investment income? And on the flip side, was that expense for a vacation, a night out, or paying off a loan? With this information, you'll have a clear understanding of where your money is coming from and where it's going.
This level provides even more specifics, like the exact details of an expense, like if the user's vacation expense was on a train, airways, or a resort. Sometimes, the parent and child category can be the same, but this level provides a more in-depth look.
|Transaction Type||Parent Category||Child Category|
|LOAN DEPOSIT||LOAN DEPOSIT||LOAN DEPOSIT|
|INCOME||AUTO CREDIT||AUTO CREDIT|
|INCOME||INVESTMENT INCOME||INTEREST INCOME|
|INCOME||INVESTMENT INCOME||INVESTMENT INCOME|
|INCOME||OTHER INCOME||CASH DEPOSIT|
|INCOME||OTHER INCOME||POINT OF SALE|
|INCOME||OTHER INCOME||OTHER INCOME|
Customer gives consent for connection to bank account or upload their bank statement via Bank Account Merit.
MetaMap retrieves and analyses the bank transactions data and generates insights based on sophisticated algorithms
Merchant receives the detailed insights via the web hook in real-time and takes the decision
Updated 7 months ago