Commonsense Knowledge Supported Intelligent News Analysis for Portfolio Risk Prediction

In this paper, we present a commonsense knowledge supported intelligent news analysis system for portfolio risk prediction. Different from traditional lexicon based text mining methods, the system aims at simulating common investors´ news analyzing ability with case based reasoning. To automate such a reasoning process by computers, a description logic based knowledge mediator is designed and plugged in the system to dynamically provide the required commonsense knowledge for case learning and case retrieving. The system implementation and evaluation have shown that this commonsense knowledge mediator is able to support multi-agent intelligent systems with an adaptive knowledge managing mechanism. Furthermore, the event level news analysis method is also proved to be suitable for portfolio risk prediction. It is our belief that these two designed architectures could be adopted in other commonsense knowledge involved reasoning problems, and make contributions to event related financial predictions as well.