HMM Classifier Object Recognizing System in Brain-Computer Interface

摘要

Machine learning (ML) is the field that adds intelligence to devices providing them with capabilities to process and identify patterns in data just like human beings do. Programming devices in this manner can help in identifying those patterns which human beings often cannot. Machine learning is based on modelling data mathematically. ML has been gaining a lot of attention in the last few decades, especially in fields of interdisciplinary research. Brain-Computer Interface (BCI) is an area where Machine Learning Technology is been rapidly using. Also, Machine Learning techniques have to be used so that one can get a better result and more efficiency. Information Transfer Rate is the best way to measure the performance of the signals. The current research is mainly focused on achieving the systems with higher ITR. The focus of the proposed system is to get better and high Information Transfer Rate by merging two different approaches. The approach used in this work is (SSVEP), Visually Evoked Potential and (SSAEP) Auditory Evoked Potential by using Hidden Markova Model (HMM). The system which is to be developed checks whether the existing system has such facility if it has, does it provides accuracy which is of a higher rate and can put it in the real-world applications.