Business Intelligence in Airline Passenger Satisfaction Study—A Fuzzy-Genetic Approach with Optimized Interpretability-Accuracy Trade-Off

The main objective and contribution of this paper is the application our knowledge-discovery business-intelligence technique (fuzzy rule-based classification systems) characterized by genetically optimized interpretability-accuracy trade-off (using multi-objective evolutionary optimization algorithms) to decision support related airline passenger satisfaction problems. Recently published accessible at Kaggle’s repository passengers data set containing 259,760 records used in experiments. A comparison approach with an alternative method SAS-system’s accuracy-oriented prediction tools determine attribute importance hierarchy) also performed showing advantages terms of: (i) discovering actual hierarchy significance for (ii) system’s optimization. results findings work include: introduction modern fuzzy-genetic solution both high interpretability accuracy support, analysis effect possible "overlapping" some input attributes over other ones order discover real influence particular upon satisfaction, (iii) extended cross-validation experiment confirming effectiveness different learning-test splits considered.