Murthy, Ranjit Murthy creator Sadananda, Ramakoti Chairperson Murai, Shunji Examination committee Shrestha, Surendra Man Examination committee Yulu, Qi Examination committee text th Bangkok Asian Institute of Technology 1995 monographic eng

print

93 leaves + 1 online resource There is a world of effort to create large databases for environmental assessment and for all kinds of decision affecting local and global environments. The International organizations such as United nations environment program and many national agencies are collecting all kinds of information going in trillions of units of information. The increasing concerns of environment at all levels will cause the database to multiply and thus together with advances in communication will cause greater interconnection of the databases. Given access to the vast data, it is necessary to discover the knowledge in the corpus. Since the number of databases are proliferating, their sizes ever enlarging and their interconnections globalizing, any contribution, even incremental, towards extracting knowledge would be of great importance. In this thesis Unsupervised learning is used to discover knowledge form databases. Specifically self-organizing maps which maps from a higher dimension to a lower one, are used. The regularity and pattern observed in one dimension gets retained in the other dimension in the mapping process. The knowledge extracted from the databases are indicated in terms of pattern formation and formation of rules. A number of pedagogic databases are taken to demonstrate the methodology in this paradigm for knowledge discovery. A Thesis Report Submitted in the Partial Fulfillment of the requirement for the degree of Master of Engineering Thesis (M.Eng.) - Asian Institute of Technology, 1995 Neural networks (Computer science) Asian Institute of Technology. http://203.159.5.9/ait-thesis/detail.php?q=B14982 http://203.159.5.9/ait-thesis/detail.php?q=B14982 080998 20260818144949.0