This article is more than 6 month old.

How Indian banks can use power of network intelligence to combat fraud


Criminal networks and banks’ fraud detection machinery are pitted against each other across a highly interconnected digital ecosystem.

How Indian banks can use power of network intelligence to combat fraud

Authored by: Damon Madden

Payments on the UPI platform in India surged to more than 1.6 billion transactions in August; a rise in digital transactions that signals that India’s dream of transforming into a cashless (or more realistically a less cash) economy is being realised. However, banks are nervously monitoring the rising volume of low-value digital payments. The reason being that the incremental cost of processing these "low-value, high-frequency" payments is high, as banks must constantly upgrade their systems to monitor fraud risks in line with the rising transaction load.

Criminal networks and banks’ fraud detection machinery are pitted against each other across a highly interconnected digital ecosystem. While individual banks are increasingly employing technologies such as machine learning and implementing multi-layer fraud detection strategies, staying a step ahead of fraudsters will require a more collaborative approach.

Creating a network for better intelligence

Fighting fraud alone rarely pays off. That is why, despite their traditionally competitive nature, there is a need for closer cooperation between banks when it comes to fraud data and intelligence sharing. Through improved collaboration, banks can create a community where real-time information on emerging risks is freely shared between members, including central infrastructure (CI) owners.

Network intelligence capabilities can enable the banking ecosystem to strengthen its fraud management infrastructure—all the way to the core. By sharing fraud intelligence, threats can be more quickly identified, and the wider community alerted to emerging fraud characteristics ever-evolving tactics. What this means, is that banks can incorporate these insights into their machine learning models more quickly and cost-effectively.

An added benefit of this approach is that CIs or governing bodies (RBI in the case of India) can be more actively involved in coordinating or facilitating fraud defence across the community. Without interfering in the decisions that member banks make regarding their individual fraud detection strategies, a community facilitates end-to-end communication between banks and the governing body, which increases transparency without compromising data-sharing compliance. Overall, this collective front is better equipped to respond to new and emerging risks, preventing them from becoming endemic threats.

India’s reality on the ground

The Reserve Bank of India’s bid to create a Central Payment Fraud Registry for monitoring digital payment frauds on a real-time basis, providing customers with periodic aggregated risk data, is a step in the right direction.  According to a recent survey conducted by ACI and YouGov, nearly half of Indian consumers are more worried about fraud while making digital transactions amid the coronavirus crisis. One of the most interesting findings that came out of the research was that when a fraudulent transaction does occur, around 60 percent of respondents would first call their bank to block their account. This indicates that during this time of heightened awareness, consumers still consider their bank the first line of defence.

With the digital payment ecosystem experiencing robust growth, which has barely faltered even as consumer spending dipped due to lockdowns, fraud detection and management have become a top priority. A central fraud registry will ensure a systematic response to fraud while equipping the banking ecosystem with insights and data that serve to strengthen their machine learning algorithms.

Compliant, community information sharing

Compliant information sharing is achieved by enabling the community to share, in real-time and in metadata format, their fraud models and their composite features, along with key performance data that supports their efficacy. Automatically stripping the metadata of any identifiable information resolves the burden and regulatory risks around attempting to extrapolate and submit data externally. This enables the community to share more data, at a lower risk.

While the central body—RBI or NPCI in the case of UPI transactions —controls quality and ensures consistency by pre-aggregating data, it does not decide which models the group adopts—that is done by the community. Furthermore, members can adopt, adapt or combine features with their own models however they see fit—and with no limit on the number of models that can run concurrently. In this way, participants can integrate proven model features into their own customised adaptive machine learning strategies. This drives unparalleled access to all the information needed to assess transaction risk levels.

Agility and adaptability are crucial in a real-time world

While we laud the developments that are leading the banking ecosystem towards the power of networked intelligence, it is crucial to keep in mind that deploying and constantly adapting predictive machine learning models is now the need of the hour. Banks and other financial institutions need solutions that reduce their reliance on specialised resources and allow them to adopt a business-led machine learning strategy that aligned with today’s fast-paced, 24/7 fight against fraud. That means implementing solutions that support the complete model development process, allowing for easy access to examine and analyse data, calculate fraud scenarios, and document key modelling steps.

A robust fraud management mechanism that relies on networked intelligence, provided, and leveraged by multiple stakeholders in the ecosystem, will bring greater efficacy. It will strengthen consumer confidence and protect banks’ reputation, ensuring that India continues to lead the way in digital payments.

—Damon Madden is Principal Fraud Consultant—Payments Risk Management, ACI Worldwide. The views expressed are personal

Market Movers

Grasim1,503.25 44.50 3.05
Tata Motors334.95 9.30 2.86
Bajaj Finserv12,155.65 335.80 2.84
Coal India153.10 4.05 2.72
ONGC125.45 2.95 2.41
Bajaj Finserv12,152.40 331.85 2.81
ONGC125.40 2.75 2.24
Larsen1,537.60 27.35 1.81
Bajaj Finance5,993.55 91.90 1.56
HDFC2,619.45 36.60 1.42
Grasim1,503.25 44.50
Tata Motors334.95 9.30
Bajaj Finserv12,155.65 335.80
Coal India153.10 4.05
ONGC125.45 2.95
Bajaj Finserv12,152.40 331.85
ONGC125.40 2.75
Larsen1,537.60 27.35
Bajaj Finance5,993.55 91.90
HDFC2,619.45 36.60


Rupee-100 Yen0.66690.00600.91