Engineering design is an information intensive activity. It is reported that designers spent in excess of 50% of their time in handling, e.g., retrieving, organizing, etc., information. Thus the efficiency and the quality of the design process depend considerably on how well designers are able to handle large amounts of information. Design information management has received increasing attention in recent years as a result of these findings and the recognition that lacking or missing key design information may lead to sub-optimal decision-making and design. Much of the research has focused on the capture, storage, indexing and presentation of design information; less work has been done on information retrieval based on an understanding of individual designers, their experience, their skills and the ways in which they use information in the context of their design task. Further, most design information retrieval systems base the context of the designer's information needs on a short phrase or query. This is a severe limitation given the situational and context-dependent nature of design information.

This research focuses on retrieving design information that satisfies designer's specific information needs efficiently and effectively. The key of it lies in the acquisition, representation and utilization of human-like knowledge about information needs. In other word, how could we make people understood by computers and how could we make computers a "real" assistant? Traditionally, human-like knowledge has relied primarily on explicit coding of symbolic facts into computer data structures and algorithms. A serious limitation of this approach is people's inability to access and express the vast reaches of unconscious knowledge on which they rely. Designing a learning mechanism to acquire human-like knowledge from the same source as human is necessary. This research investigates particularly on how to capture/discovery context knowledge in engineering design information retrieval. The hypothesis is to discover and understand tacit and explicit knowledge about designers and their activities in information seeking through data mining usage logs and Latent Semantic Analysis (LSA).