INFORMEX: An Expert System to Enhance Information Presentation

Abstract

Scope and Method of Study: The proliferation of information systems in organizations has allowed companies to capture and store volumes of operational and strategic data. Unfortunately, most companies lack a systematic way to present this wealth of data into a form useful for decision makers. Artificial intelligence technologies provide a promising vehicle for improving this information flow dilemma. Scientists have conducted studies and experiments to find out how mode of information presentation (graph, table, etc.) and content (raw data, summarized, etc.) affects decision quality. Though much has been learned from this research, the findings have not been consolidated in a form that can be used by information systems professionals who create computer output. The objective of this research is to develop a knowledge-based expert system (knowledge system) for those who are charged with creating the delivery vehicle (report or graph) from the company's information system (i.e., system analysts). This project surveys twenty-nine experiments and integrates their findings in an expert system to recommend presentation format and content. Findings and Conclusions: The INFORMEX expert system has 73 rules in its knowledge base. It uses information about the task type, task actions, task attributes, and individual characteristics to produce weighted recommendations for presentation format and content. The system also has a text-retrieval system with categorized summaries of the experiments used to build the knowledge base. This retrieval system provides an explanation facility for INFORMEX's recommendations. A second benefit of developing the knowledge base has been identifing areas where no or little research has been performed in the information presentation field.