Attacking & Preventing an AI Based Disease Identification System

Prasanna Shivaji Shinde, Avinash Shrivas

Abstract: Artificial intelligence in medical (AIM) has reached a period of adolescence in which interactions with outside world are not only natural but mandatory. Furthermore, an adequate appraisal of AIM research requires an understanding of the research motivations, the complexity of problems and a suitable definition of criteria for judging the field-s success. In this proposed system user can communicate to the system for retrieving information about diseases by giving symptoms as input. Also, an algorithm independent approach for mounting poisoning attack across a dataset of system is being stated along with the countermeasure for attack which specifies how long the dataset of system is secure.

Keywords: AIM, Chatbot, Poisoning Attack, machine learning, NLP

How to Cite?: Prasanna Shivaji Shinde, Avinash Shrivas, "Attacking & Preventing an AI Based Disease Identification System", Volume 4 Issue 5, May 2015, International Journal of Science and Research (IJSR), Pages: 2436-2438, https://www.ijsr.net/getabstract.php?paperid=SUB154727, DOI: https://dx.doi.org/10.21275/SUB154727