Background/Objectives: Adaptation and personalization of E-learning systems require efficient learner modeling. Attributes of learner are evaluated to classify their knowledge without considering the weight difference with respect to their similarity level in E-learning environment for intuitionistic fuzzy data. Methods/Statistical analysis: This paper proposes an Intuitionistic Fuzzy Weighted Averaging (IFWA) operator. The IFWA operator is combined with Genetic Algorithm (GA) to tune the weight of the attributes of learners with respect to their similarity level. The proposed model tests and evaluates the IFWA algorithm on user knowledge modeling data set taken from UC irvine machine learning repository. Findings: The algorithm measures the performance in terms of the best weight values corresponding to the classification. Intuitionistic fuzzy data set is compared based on mean error for different run of generations' with best weight values. The mean square error .002349 proves the consistent performance of the algorithm to allocate weight to the attributes in intuitionistic fuzzy domain. Applications/Improvements: The proposed Intuitionistic Fuzzy Genetic Weighted Averaging Algorithm (IFGWA) can play an efficient role in various decision making problems.
Keywords
Domain Dependent Data Classifier, Intuitionistic Fuzzy Genetic Weighted Averaging Operator, Multiple Attribute Decision Making
User
Information