Tech giant Microsoft seeks to harness Artificial Intelligence’s bizarre forecast abilities to identify PC malware attacks before they even happen. For this Redmond-based multinational tech firm has sponsored a contest on Kaggle that challenges data scientists to build models which forecast if a device is likely to become contaminated with malware given a current machine state. Kaggle is an online community of data scientists and machine learners, owned by Google, Inc. that enables users to explore and release datasets, search and build models in a web-based data-science environment.
Microsoft, in a statement, noted that the competition offers academics and researchers with diverse backgrounds a new opportunity to work on a real-world issue utilizing a new dataset from Microsoft. The company further stated that outcomes from the competition will help the company recognize opportunities to further advance Microsoft’s layered defenses, concentrating on preemptive protection. In addition, Microsoft noted that not all machines are equally to get malware; contenders will assist to create a model for detecting devices that have a higher risk of getting malware so that preventative action can be taken. Contestants are supplied with 9.4GB of anonymized data accumulated from 16.8M devices to create their models. For participants, a sum of USD 25,000 prize money will be utilized to incentivize participation that would be spread out as USD 12,000 for 1st place, USD 7,000 for 2nd, USD 3,000 for 3rd, and USD 2,000 and 1,000 for 4th and 5th places respectively.
To date, the best performing entry has gained 68.9 percent precision, though it’s likely this will be improved before the end. For participating in this contest, participants must be submitted their entries before the competition close on March 13, 2019.