Clustering and classification in the ecology of intelligent agents | |
| Author | Padmasiri, Sujeewa Umendra |
| Call Number | AIT Thesis no. CS-97-6 |
| Subject(s) | Computer viruses Intelligence agent (Computer software) |
| Note | A thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering, School of Advanced Technologies |
| Publisher | Asian Institute of Technology |
| Abstract | A clustering and classification approach should produce a clump space in which agents with similar properties agglutinate together to form clusters. These clusters in the clump space should be promising and useful such that they can be used in learning about agents, sharing the domain specific knowledge, reusing of the existing architectures and designing future agent- paradigms. An approach for clustering and classification in the ecology of intelligent agents is proposed. The approach is modeled as an unsupervised learning method by which the similar agents become glued together. In doing so, there should be a way to distinguish the cluster boundaries of the original data. The agent data from the available literature is rather qualitative than quantitative. In the absence of a proper definition for agents, these qualitative data need to be categorized by considering various levels of common, nevertheless significant properties. These categorizations place the agents in a conceptual space. In categorizing, ranking is involved. These ranking values are used to represent the categories of the properties concerned. Since the approach suggested uses quantitative data for learning, it is necessary to derive a suitable conversion method with which the original conceptual distances of the qualitative data are preserved even after the conversion. The overall data organization is used as the input to the learning algorithm. The suggested approach forms distinct clusters of agents. These clusters demonstrate various aspects related to agent paradigm and that augments the importance of these clusters. |
| Year | 1997 |
| Type | Thesis |
| School | School of Advanced Technologies (SAT) |
| Department | Department of Information and Communications Technologies (DICT) |
| Academic Program/FoS | Computer Science (CS) |
| Chairperson(s) | Sadananda, Ramakoti; |
| Examination Committee(s) | Batanov, Dentcho Nikolov ;Pattarasinee Bhattarakosol ;Acharya, Sushil; |
| Scholarship Donor(s) | H.M. the King of Thailand.; |
| Degree | Thesis (M.Eng.) - Asian Institute of Technology, 1997 |