Approximate reasoning with fuzzy rule interpolation: background and recent advances

Abstract Approximate reasoning systems facilitate fuzzy inference through activating if–then rules in which attribute values are imprecisely described. Fuzzy rule interpolation (FRI) supports such with sparse bases where certain observations may not match any existing rules, manipulation of that bear similarity an unmatched observation. This differs from classical rule-based requires direct pattern matching between and the given rules. FRI techniques have been continuously investigated for decades, resulting various types approach. Traditionally, it is typically assumed all antecedent attributes equal significance deriving consequents. Recent studies shown significant interest developing enhanced mechanisms associated relative weights, signifying their different importance levels influencing generation conclusion, thereby improving performance. survey presents a systematic review both traditional recently developed methodologies, categorised accordingly into two major groups: non-weighted weighted It introduces, analyses, range commonly used representatives chosen each categories, offering comprehensive tutorial this important soft computing approach to inference. A comparative analysis provided within category two, highlighting main strengths limitations while applying problems. Furthermore, adopted criteria algorithm evaluation outlined, recent developments on methods presented unified pseudo-code form, easing understanding facilitating comparisons.