RECOMMENDATION SYSTEM FOR TOURISTS

Abstract

This project uses statistics from previous tourism experiences to recommend travel areas to travelers. To recommend a traveling area, we use the C4.5 decision tree algorithm with some specific criteria. To promote new locations, all existing algorithms, such as collaboration algorithms and content filtering algorithms, employ the most recent user experience data. If the present user does not have any previous experience data, these algorithms will fail. Another solution to the problem is to utilize the C4.5 decision tree method to build the model based on prior users' experiences. When a new user enters a requirement the decision tree predicts the best and most accurate location. The decision tree does not require any additional user experience data. To use the decision tree model, you'll need a dataset. Pre-processing techniques can be used to eliminate empty or wrong values from this dataset, which can hurt the decision tree model. It is possible to remove the value. To predict or build the model, it may not be necessary to use all of the column values in the dataset. To remove unwanted features a feature selection algorithm is used. The MRMR feature selection algorithm is used to remove extraneous attributes, minimize the building model's execution time, and increase the system's correctness.