Disruptive technologies like Machine Learning (ML) and Artificial Intelligence (AI) have opened many doors for the finance sector when it comes to understanding data. They are helping institutions to streamline processes, optimize portfolios, decrease risk and underwrite loans amongst a plethora of other things.
As a forerunner in the financial technology space, Montran believes in consistently innovating and keeping ahead of the curve, so that our clients have the edge they need to service their international and domestic customer base. As data becomes king, Machine Learning is just one of the many areas we have delved into in recent years. In this article, we take a quick look into the developments in the Artificial Intelligence at Montran.

IMAGE PROCESSING
Image processing involves using large amounts of unstructured data. Our team uses AI successfully to process this unstructured data with the help of deep-learning algorithms, powerful GPUs and a large amount of computing resources.

A) Offline Signature Verification
We launched the AI-based ‘offline signature verification system’. Conventional systems require two signatures per customer. Our improved system requires just one signature per customer. Test results show an effective accuracy rate of over 90%.

B) Image Processing Optical Character Recognition Feature
We have also developed an AI-based application for Optical Character Recognition (OCR) of handwritten fields on the mandate form. The application also tests the validity of the scanned mandate images and the accuracy levels are more than adequate. This is not only convenient but reduces human effort and saves time of several hours vis-à-vis manual data entry.

FORECASTING
Forecasting helps predict future values based on the past and present data and by the analysis of trends. Our development team has created AI-based forecasting applications. We have already introduced them along with our liquidity management products. These provide the trends, probabilities, insightful analysis of historical data and accurate forecasts based on the data captured by the system. The findings are presented through well-designed dashboards, statistics and graphs to provide you with insight to empower your managers to take well-informed decisions. The application has received good reviews from the industry.

DEVELOPMENTS IN THE PIPELINE
We are developing several applications for sentiment analysis, classification of SWIFT messages, fraud detection, and blockchain. The deployment of the latest machine-learning tools such as TensorFlow, Keras, scikit-learn and Python will help deliver better and enhanced solutions.

THE WAY AHEAD
Artificial Intelligence in banking is about digesting large amounts of data, extracting meaningful insights and using these to drive security-based and business decisions. In line with this core vision, our team is developing several applications using AI to simplify and automate business processes, improve security checks, make accurate forecasts, detect frauds and enhance customer experiences.