A programmable agent for knowledge discovery on the Web

With the tremendous amount of information that is becoming available on the Web, the ability to develop information agents quickly has become a crucial problem. In this paper, a programmable agent is presented for Internet application to retrieve and extract information from the Web with user's guidance. The agent consists of a retrieval script to identify Web sources, an extraction script based on the document object model for the extraction process and a data translator to export the extracted information into knowledge‐based frame structures. A GUI tool called Script Writer supports the generation of extraction script visually.