A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science, School of Advanced Technologies

Thesis (M.Sc.) - Asian Institute of Technology, 1999

The Internet has long been a large source of information, and it keeps growing every day. Serving diverse audiences around the world, the Internet faces the problem of appropriateness of its content delivered to people - certain materials are appropriate only for certain groups of audiences. Initially, there was an effort to protect children from unfavorable resources. Several working groups had put forth effort to develop methods to filter out the inappropriate materials. The World Wide Web Consortium working group finally invented a metadata specification called 'Platform for Internet Content Selection (PICS)'. Web builders and Software developers thereafter create Web pages and develop filtering software according to this specification. There are several factors influencing the judgement on 'appropriateness' of an Internet material - for example, age and educational level of the user. Embedding a knowledge-based component in the filtering software will indubitably be a means to enhance the software's capability. However, none of the software employs such a component. This thesis study seeks to illustrate another way to exploit metadata for information filtering purpose by combining it with a knowledge-based system. This thesis has also proven that it is possible to build knowledge-based systems that can work with metadata standards. Should there be any new metadata standards in the future, such systems can be developed to exploit them.