Extracting conceptual relationships from specialized documents

Conceptual modeling has been fundamental to the management of structured data. However, its value is increasingly being recognized for knowledge management in general. In trying to develop suitable conceptual models for unstructured information, issues such as the level of representation and complexity of processing techniques arise. Here, we investigate the use of a conceptual model that is simple enough to allow efficient automatic extraction from two kinds of documents--scientific research papers and patents. Our model focused on the problem-solution relationship that is central to the analysis of scientific papers, while allowing supporting relationships such as methods and claims. We evaluated the utility of the approach by building a prototype system and carrying out experiments that assessed the accuracy level of the techniques used in building the model and the acceptability of the model through preliminary user studies. The feedback from these experiments shows promising results that support our choice in the tradeoffs between the granularity of the model and the processing techniques used. We discuss a variety of issues that arouse from this project and describe several directions for future work.