Accident Identification Using Fuzzy Cognitive Maps with Adaptive Non-Linear Hebbian Learning Algorithm

Abstract

Knowledge Representation (KR) is a medium for communicating our perception of objects, ideas, imaginations, and events within certain contexts, as well as the relations among these perceptions. KR is necessarily an approximation of the real world object, concept or event that focuses on certain properties. So, if the process of KR is done, in a more intelligent way, then almost all the problems, discussed in the research background can be solved in an effective manner. Fuzzy Cognitive Map (FCM) is a powerful tool to engage processes of knowledge acquisition and knowledge representation. The soft computing technique of Fuzzy Cognitive Maps (FCM) for modeling and predicting the reason for the accident has been proposed. For the FCM models, the cause of a complex system is used to develop new knowledge-based system applications. FCM combines the robust properties of the fuzzy rationale and neural systems. To overcome the limitations and to enhance the efficiency of FCM, a good learning technique for unsupervised training could be applied. A decision system with an FCM based on human knowledge and experience, trained using unsupervised Non-linear Hebbian learning algorithm is proposed here. Through this work, the Hebbian algorithm on Non-linear units is used for training FCM to know the reasons for the accident. The investigated model serves as a guide in determining the scenario and in planning the remedial methods for preventing & detecting the accident.