Information Support for Real-Time Decision-Making Based on Big Data: Knowledge-Enabled Machine Activeness and System Efficiency

Real-time decision-making requires efficient information support that dynamically provides information useful to decision makers. Traditional fixed information service mode is challenged by Big Data environment, while search-based methods do not guarantee efficiency. To solve the problem, a method is proposed to improve machine activeness on information collection, allowing user to be focused on decision-making, so as to improve the efficiency of the whole decision-making system. During human decision-making process, machine keeps aware of the decision task context, dynamically recognizes user information requirement, automatically activates search process, based on domain knowledge previously built reflecting the latent relations between decision task types and the information required. It is proved by experiments that the method can effectively save user time cost on information requirement expressing and improve task relevance of collected information.