A Correlation Computing Method for Integrating Passengers and Services in Semantic Anticipation
Motoki Yokoyama, Yasushi Kiyoki, Tetsuya Mita · Frontiers in artificial intelligence and applications · 2019
New information-provision focusing on individual passengers is expected in a railway environment. Our method realizes multiple semantic spaces, which are selected according to passenger's contexts. By using correlation metrics for the causal interaction of different semantic spaces, this method anticipates the passenger's needs and generates a ranking of services and facilities. Experimental study confirms that the ranking of passenger requirements for services and facilities would change appropriately in response to the causal interaction of the semantic space. The experimental results show the feasibility and applicability of this method.