Experience with Rule Induction and k-Nearest Neighbor Methods for Interface Agents that Learn

Mathematics

• Rule Induction 100%
• Nearest Neighbor Method 99%
• K-nearest Neighbor 86%
• Learning Algorithm 50%
• Architecture 47%
• Agent Architecture 45%
• Information Filtering 42%
• User Preferences 40%
• Feature Extraction 31%
• Experimentation 25%
• Training 22%
• Profile 18%
• Knowledge 16%
• Strategy 15%

Engineering & Materials Science

• Learning algorithms 35%
• Information filtering 29%
• Feature extraction 16%