Machine learning is nothing new, but during the pandemic, fraudulent activity hit an all-time high, and its popularity soared. Now, it is the primary tool used for mitigating fraud, and companies like ACI Worldwide are leading the charge in developing algorithms and models to serve each and every one of their customers.

To further discuss the benefits of machine learning and how it can better serve institutions looking to improve their fraud prevention technologies, PaymentsJournal sat down with Patricia Rojas, Senior Manager Data Scientist at ACI Worldwide, and Tim Sloane, VP of Payments Innovation at Mercator Advisory Group.

Machine learning is essential for fraud prevention

It is now clear that machine learning is a valuable tool for fraud prevention, and most experts would agree that it has become essential for mitigating cybercrime. On a high level, detecting fraud is about learning the difference between normal spending behaviors and unusual, fraudulent purchases. With machine learning, the technology can analyze all available data and educate itself on the difference between an honest transaction and a fraudulent one.

“These type[s] of models, when they’re properly trained and get the feel for one specific merchant or one specific sector, they can help increase the fraud detection accuracy in your overall strategy by as much as 40 to 50%,” claimed Rojas. She warns, however, that merchants and PSPs need to understand the specifics when implementing machine learning algorithms, because there are many different techniques and levels of sophistication. It is also important to note that these algorithms are limited by the amount and quality of data within the institution.

There are many different applications of machine learning, and its evolution shows no signs of slowing down. With fraud also occurring in a fast-paced environment, a company like ACI is necessary to correctly apply machine learning to fraud prevention.

Machine learning trumps other fraud prevention tools

Identifying fraudulent behavior can be a complex and time-consuming task, especially for institutions with an abundance of data. In such cases, machine learning models are ideal because of their efficiency and ability to analyze massive amounts of data to identify trends. Not only are they more precise, but they are also exponentially quicker.

“This is very important because different behaviors change very quickly,” said Rojas. “You need to be able to stay on top of that and to adapt your strategy to be able to capture those new fraudulent behaviors.” Overall, machine learning is a tool that can help its users improve their fraud prevention strategy and minimize the ‘false positive’ transactions. It can even assist in reducing friction for customers at checkout.

Tim Sloane breaks down the process to offer a better understanding: “You have data at the merchant location. You have [data] about the account individual, their behavior. You have data coming from the network. You have data at the acquirer. And you have data that, if you’re lucky, you can get from the issuer to be able to tie it all together. [Machine learning can] pull those signals together and learn more than you possibly could any other way.”

All machine learning is not created equal

There are a multitude of machine learning models, as well as many different algorithms that can be used, case-by-case. While tree-based algorithms tend to work best for fraud detection, different use cases might require a different approach. It is crucial to first use the right model, and then to optimize that model for a specific merchant or sector. When models are trained with specificity, they are more effective because they take into account the nuances of customer behavior, fraud trends, and spending patterns.

“At ACI, one of the things we do to improve the performance of our model is to leverage the power of the consortiums by building strong models for our merchants,” explained Rojas. “We do this by identifying similar merchants and then combining all that information to train our models.” This gives ACI a larger set of data to provide information for the model they are building, which then enhances the ability to correctly identify fraudulent behaviors and make more accurate predictions for future transactions. The performance result is significantly increased.

ACI is also developing new incremental learning models. This type of models differs from static models mainly in how they are built and maintained over time. With a static machine learning model, a historical set of data is used to build the model and, over time, that model becomes less efficient as fraudulent behavior evolves and model will need to be retrained to learn the new fraudulent behaviors to be able to make an accurate prediction. With the new learning model, the technology is able to think for itself and adapt to new behaviors without having to relearn everything it already knows which not only makes the training phase more efficient but also a more accurate prediction using more recent and relevant data to prevent future fraudulent transactions.

“These types of models will perform better in production for longer, and it’s reduced the number of retraining[s] that we need to do…it’s a smooth process for the customers,” concluded Rojas.

Mitigating the limitations of machine learning

“Sometimes a merchant has a special offer going out,” explained Sloane. “And that special offer is going to generate new types of traffic that needs to be coordinated with the machine learning tools and the people who are operating them to make sure that that special offer is done in a safe fashion and doesn’t throw off the models.”

Seasonality can significantly impact the performance of models. High sales peak seasons and the launch of a new product can both impact the reading of normal and abnormal behavior.

Everybody has different goals, and merchants are no exception. While one merchant may be looking to reduce false positives, another might want to maximize the fraud detection rate. ACI engages with merchants at a very early stage to understand their goals and offer a multi-layered technology to optimize the overall fraud strategy in a way that best caters to the needs of the merchants. It takes into account seasonality, peak sales seasons, new product launches and other special circumstances to ensure the merchant is protected against fraud and revenue is not impacted.

Part of ACI multi-layered technology is the Rule Intelligence process, which is a machine learning model that generates human readable rules in an automated way that is tailored to merchant-specific needs. The rules generated by this process are a small set of high performing rules, which reduces the false positives, reduces the time needed to create a fraud strategy, and can be refreshed to adapt to changes in behaviors.

The post Machine Learning is the Newest Leader in Fraud Prevention appeared first on Artificial Intelligence.

AI in fraud investigation makes the whole process efficient by generating relevant data

Right from deploying machines to get the work done to the robots assisting the doctors in surgeries, we’ve come a long way – thanks to Artificial Intelligence, truly a remarkable innovation! Today, the business models that we get to see extensive use of technology. Also, the new and complex challenges behind managing the fraud investigation are a strenuous task in itself. Well, it doesn’t end there. Cross-border probe adds to the already existing complexity. Such an investigation could highlight bribery, corruption, data breach, conflict of interest, fraud in financial reporting and IP theft, to name a few.

There are a lot of factors that need to be accounted in case of a cross-border probe. Some of them are –

• Local rules, laws and regulations
• Cultural attributes
• Language barriers
• Different standardization levels, etc.

No wonder why such a probe is complex and full of challenges. With these complexities stepping in, deploying the right tools with a well-laid investigation methodology standardisation is the need of the hour.

The procedure followed by such a probe is as stated –

• Needless to say, the first step has to be getting in as much information as possible. Relying on both – external as well as internal sources for the same yields fruitful results. External sources include media, open source information, etc. whereas internal sources obviously revolves around employees, vendors, business operations, etc. A key point to note is that the information collected should be on a timely basis and also from as many sources as possible.
• Next up, try identifying the relationship between key entities and individuals
• Many tend to ignore this but is equally important – a sound knowledge on the event chronology.
• Transactional data holds a lot of crucial information. Being able to understand what goes in and how to draw necessary insights is the key here. The probe is incomplete when transactional data is not being addressed.

Now that you have a pile of data to analyse, sitting to scan every bit of this makes no sense. Identifying relevant content is the key here. It is here that Artificial Intelligence comes into the picture. Embedded artificial intelligence helps in filtering down the content and classifying it as required. The feature of semantic search is no less than a blessing here for it automatically identifies related concepts and documents.

Artificial intelligence in fraud investigation makes the whole process efficient by generating relevant data and leaves us in a position to draw meaningful insights.

Data extraction is a tedious task and when sensitive data like the data pertaining to banks, financial institutions, hospitals, etc. is involved, one cannot afford being negligent here. With advanced computer vision algorithms in place, it is possible to extract information from bank statements, and various other documents. Natural Language Processing (NLP) techniques help in extracting information and also aid in performing automated verification using digital channels. Artificial intelligence is no less than a saviour for financial institutions as it caters to verification of the critical details and documents.

Analytics also plays a pivotal role in investigation. Two types of analytics, namely advanced analytics and process analytics have a lot to offer to ease the investigation process. Talking about advanced analytics, it performs the following –

1. Network analysis: This is where a relationship between individuals and entities is established.

2. Sentiment analysis: When documents are to be differentiated on the basis of tone, subject, etc., analytics comes into play. Also, any suspicious review, article, conversation, etc. can be effectively identified.

3. Detecting anomalies in the transactional data is easier than ever.

4. Using AI, it is also possible to identify undisclosed entities.

Process analytics: What can get better than getting to know how your processes are performing in addition to what needs to be done to perform better? This is exactly what process analytics has in store for you!

Though these tools and technologies can help in fraud investigation and management, what needs to be understood here is that the process is complex and comes with their own challenges. Sound knowledge about these tools and techniques might help. With frauds and crimes in the world of technology continue to rise, it is high time that the investigators have access to the right AI tools and technologies to tackle these situations.

The post ARTIFICIAL INTELLIGENCE IS PLAYING A BIG ROLE IN FRAUD INVESTIGATION appeared first on Artificial Intelligence.

The Department for Work and Pensions (DPW) is the UK government’s largest department and supports over 22 million people every day. It is currently hiring for a number of senior digital roles, including two lead data scientists, as well as a deputy director of Fraud, Error and Debt. 

DWP has a number of complex technology projects ongoing, but notably it is the lead department for the government’s flagship Universal Credit scheme, which aims to merge a number of welfare payments into one single payment. Universal Credit was created with the intention of making it easier for people to find and get back into work, but has faced high level criticism since launch. 

Last year the National Audit Office said that Universal Credit was not value for money and that many processes “still manual and inefficient”. 

On the Deputy Director of Fraud, Error and Debt role, the Department is looking to hire someone that can support the organisation in preventing fraud and eliminating error through the definition and delivery of a range of “next generation digital solutions into a large scale, business critical live service”. 

The successful candidate will be based in Manchester and receive a salary of £90,000. They will also lead a team of 120 people, with the aim of helping DWP develop an internal enterprise wide capability that will enable the department to “operate at the forefront of this profession”. 

The job description states that the deputy director will be responsible for: 

• Leading the delivery of large scale digital products in major, complex, multi-supplier and in-house development environments. Driving the performance of a number of digital delivery teams operating in an agile environment to deliver business outcomes; 
• Recruiting, building and leading teams to both protect existing live services and to deliver transformative new ones. 
• Ensuring live service performance of existing Fraud, Error, and Debt services and owning a clear roadmap that describes how continuous improvement of these services sits alongside strategic transformation activities already underway; 
• Working with other senior leaders and product managers to remove blockers, manage risks, commercials, budgets, suppliers and people assignments. 
• Leading by example to help transform the culture of the organisation; fostering a high trust, empowered environment. Embedding a culture based on openness and transparency, supporting inclusive values to drive engagement and performance in a matrix-managed system; 
• Advanced stakeholder management, particularly with regard to agile and product orientated approaches used by the team where influential stakeholders may not be bought into the approach and may be lobbying for deviations that would impact our ability to meet desired outcomes.

DWP has said that those interested in applying need to “exceptional and authentic” digital and technology leaders, that are capable of working in complex environments and managing large blended teams. 

Candidates also need an up-to-date awareness of leading-edge technologies and a curiosity to push the boundaries of what’s possible. 

They must also have proven experience of developing and delivering capability at an organisational and team level, as well as a strong track record of building stakeholders and supplier relationships. 

Data science

Elsewhere, DWP Digital, DWP’s internal technology and digital organisation, is looking to hire two lead data scientist roles. 

The first, based in Sheffield, will lead DWP Digital’s Analytics capability, which “discovers, innovates and continuously improves” DWP’s digital services. 

The successful candidate will lead a team that combines software skills with the latest data science techniques, in order to “find and understand patterns in data that can be put into action supporting improvements in customer experience and staff productivity”. 

The second position will be based in London and join DWP’s Risk and Intelligence Service to lead the analytics capability there. 

The salary being offered for both roles is between £70,102 and £83,953. 

DWP has said that the data scientist leads will be overseeing the quality and consistency of multiple analytics-driven initiatives each worth multiple £m including: 

• Building and stretching the data science, machine learning and AI capabilities by keeping up to date with the latest transferable market and industry trends, emerging technologies, delivery accelerators and sharing knowledge and experience with the relevant teams. 
• Building peer and senior relationships across the business including digital, operations and policy in order to ensure teams have the tools they need to de-risk delivery. 
• Building strong relationships between data science peers in DWP, other Government Departments and in industry and academia in the relevant field (e.g. digital service design, cyber security or counter fraud) encouraging innovation, reuse and avoiding duplication. 
• Be seen as a leader by your team in applying a consistent vision, energy and drive that motivates team(s) to meet project, programme or business expectations. 
• Driving quality and impact of analytics strategies, digital solutions and outcomes, balancing the theoretical and practical aspects of technical, business, and time constraints. 
• Ensuring all work adheres to key privacy, security and data protection principles, and also set procedures for ethical considerations such as unintended biases. 
• Playing a key role in the leadership in the community, defining recruitment strategies and career progression as well as contributing to Digital Practices in DWP overall.

The post DWP on the hunt for a fraud chief and leading data scientists appeared first on Artificial Intelligence.