Machine Learning to Calculate Trip Budget

Abstract

Trip planning requires effort. Majority of which is consumed in balancing preferences of travel and stay; with budget. This effort can be minimized using budget estimator. Summing up the total costs to calculate budget is ideally correct. Practically, budget can differ from individual to individual based on their nature. Some prefer to spend more while some less. Machine Learning could help predict human nature using feedback mechanism. Taking feedback about total cost incurred and comparing it to actual estimate could give insight about user nature to the system. In this paper, we have built a budget estimator that considers user preferences and uses regression algorithm to compute costs. It later asks user to input the actual cost incurred, correcting its previous estimate and uses the updated entry to drive data to be more user-specific. The system gives percent classification of 84% and percent recognition of 72.27%.