The term artificial intelligence (AI) has always been a source of confusion and controversy. There is no mainstream to guide the discussion. The most prominent type of AI that draws the attention of users, and that is thought to replace the human being in various functions, especially work (where catastrophic scenarios are projected) is the so-called “General Artificial Intelligence”. The objective of this is to create a robot or android that assimilates, speaks and even reacts like human beings and, in this sense, AI-enhanced assistants, such as Siri, Alexa or Cortana, are a first example of this approach.
However, there is still no robot available that can replace human beings in their different aspects and realities, and the truth is, there is a long way to go. Currently most AI programs offer us narrow-minded “special solutions” that can beat humans at chess or can master some discrete tasks to solve specific business problems. This practical type of AI uses machine learning techniques and brings some value in the different industries where it operates.
But what is machine learning? In simple terms, they are tools that use algorithms to learn from and adapt to data, allowing computers to find hidden insights without being told where to look.
Currently specialized software based on Hybrid Artificial Intelligence, which combine the benefits of data-based AI (eg Machine Learning) with knowledge-based AI (OR), are used to identify the risk and fraud that certain organizations are running in the financial market. In this sense, machine learning models have proven to be a powerful tool against this scourge. However, they require a large amount of information and experts behind them, so that they can be used to their full potential.
A hybrid approach, on the other hand, allows machine learning models to be complemented with expert knowledge, achieving immediate and more reliable results.
Current systems that maintain dynamic profiles for different entities (for example, a bank account) can detect “potentially fraudulent” operations in real time through advanced rules. Additionally, through rules based on fuzzy logic, independent rules can be created that help avoid the risk of a particular event. Thus, by being able to manage independent rules for different data within the same transaction, it is possible to create “digital footprints” for customers, identifying “unknown” patterns that, because they are isolated cases, could remain under the radar of a model totally based on in machine learning, such as, for example, a transaction where the amount is 30% higher than the average drawn in the last few months, where there are several new accounts involved for the client, amounts close to the maximum daily limit allowed, unknown IP, country of the operation different from that of the client, etc.
Each of these “anomalies” in the transaction alone can be “strange and suspicious”, but not enough to raise the right alarms. With the Hybrid Artificial Intelligence model, advanced techniques and machine learning work hand in hand, making software that uses this technology substantially more effective alternatives to anticipate these types of problems.
A key factor in making this approach effective is the immediacy with which these models can react. There are few tools that have the capacity to work in real time and be able to give a response in “milliseconds” to identify and prevent fraud before it happens.
Those models that combine all available technologies and techniques have a substantial comparative advantage over models focused only on some technologies. This allows organizations that use them to achieve substantial savings and have a healthier and more satisfied customer base.
Federico dos Reyes
CEO of INFORM in Latam
#Hybrid #artificial #intelligence #scope