In this day and age of a plethora of information, the importance of information security cannot be emphasized enough. Any threat to confidentiality, integrity or availability of information must be taken seriously. Ignoring such threats can have serious consequences, like misappropriation, modification or encryption of data. Vulnerabilities in information security are a tempting target for malware. Malware are malicious scripts or software, including computer viruses, worms, Trojan-horses, ransomware, spyware, adware, etc. The traditional way of detecting an advanced malware or threat compromise uses a signature-based antivirus. This approach, however, is not foolproof and can be bypassed. The signature-based approach relies on a known list of signatures. The list of signatures is not perfect and also does not contain previously unseen malware signatures. The proposed system uses operational intelligence tools and machine learning to monitor usual user behavior. This is done by collecting system activities like event logs, sysinternal, etc. Once the system learns normal behavior patterns, it can detect anomalies that may be caused by malware. Thus, unlike signature-based approach, the proposed system can detect previously unseen malware as well.