An Effective Inductive Learning Structure to Extract Probabilistic Fuzzy Rule Base from Inconsistent Data Pattern

Bio-signal/behavior pattern acquisition and its use are essential in human-friendly human-robot interaction to recognize human intention. However, it is usually difficult to model and handle such interaction due to variability of the user’s behavior and uncertainty of the environment in human-in-the-loop system. In this paper, we shall show the benefits of a PFR (probabilistic fuzzy rule)-based learning system to handle inconsistent data pattern in view of combining fuzzy logic, fuzzy clustering, and probabilistic reasoning in a single system as an effective engineering solution to resolve inconsistency of the complicated human behavioral characteristics. Moreover, we introduce a PFR-based inductive life-long learning structure for continual adaptation throughout incessant learning and control. The learning system gradually extracts more meaningful/reliable rule-based knowledge in incorporation of learning processes in short-term memory, interim transition memory and long-term memory. To show the effectiveness of the proposed system, we introduce a successful example as a case study in view of probabilistic fuzzy rule-based knowledge discovery to handle TV watching behavior data pattern learning.