Cooperative document retrieval making user's Ill‐defined query evolve
Hirohiko Mori, Yousuke Kinoe, Kazuhiko Seto, Yoshio Hayashi · International Journal of Human-Computer Interaction · 1991
Current document retrieval systems assume that a user can represent his/her query using natural language or keywords. The user without much knowledge about the subject being searched for often cannot even phrase the query properly. In this article we propose an interactive document retrieval system to overcome this human‐computer interface difficulty. Our system modifies its responses according to the user's evolving mental state. The system makes inferences about the users’ subject knowledge based on their response to information it presents. This also allows the users’ requests to be more quickly focused and effective. The system proposed in this article is implemented by using the neural network technique, and the user's query is represented by the activation pattern over the neural network.