Malware Detection in Banking and Financial Sector using Light Gradient Boosting Model

The Internet is an ever growing frontier of information, knowledge and data exchange. Every day approximately 2.5 quintillion bytes of data is created by the 367 million internet users, which can be anything from photos and videos to audios and gifs etc, worldwide. In India alone, the number of users has reached 523.3 million of the 1.35 billion people. This poses a huge responsibility on various companies that handle the sensitive data for hundreds and thousands of millions of people in India, let alone worldwide. One of the most sensitive data is present in the banking and financial institutions which holds the livelihood of such sensitive information. An attack on such data, be it stealing, selling and modification of data, can be harmful for both the trusted parties involved. This attack can be done via malware, for eg, Zeus, Zitmo, GOZ etc. In today's world, modern malware creators have innovative and creative ways to design malware which can evade all the pre-existing traditional signature based malware detection softwares. One of the more advanced ways to detect malware is to use and harness the power of machine learning algorithms like Light Gradient Boosting Model. Such a model will be very efficient on the system resources, along with being scalable and will give better results.