Mobile Intelligent Interruption Management: A New Context-Aware Fuzzy Mining Approach

Abstract

In recent days, phones recognized as more significant personal communication device for daily life. Usually, ringing notifications are utilized in notifying users on incoming calls. Notifications of inappropriate incoming calls occasionally cause interruptions for users and surrounding people. These unwanted interruptions have a disruptive effect on productivity, employee concentration, and error rate for tasks. A diversity of recommendation approaches for context-aware (e.g., data mining, decision tree, statistics, besides the soft computing) for limiting mobile phone interruptions was presented. However, a mutual problem for current techniques to minimize the interruptions of the mobile phone isn’t sufficiently coping with noisy or inconsistency instances that may minimize prediction accuracy. Hence, we are motivated to implement an integrated approach depends upon Bays classifier that classifies noisy cases from training the dataset, and fuzzy logic to manage the nebulizer in mobile phone context situations. The integration methodology implemented through feature-in-decision-out level fusion. In these regards, current work thong to extend a commonly utilized context-based data mining approaches that take out individuals unwavering temporal patterns to fuzzy data mining which might recognize social practices patterns, that might change after some time by supporting reinforcement learning. Simulation and evaluation results on real-life datasets cell phone reveal the efficiency of the suggested model. It achieves an improvement of 5%, 7% and 9% for precision, recall, and f-measure respectively contrasted with traditional systems which include decision tree model and Apriori model.