The repositary consists of the implementation of naive bayes and peason's coeffecient to make the recommendation for a given user.
Naïve bayes prediction In this method we calculate the gaussian probability density function since the variables are continuous , when the values of the attributes are given we can predict if he would click on the advertisement or not based on his internet usage, time spent, age, income.
Pearson coefficient measure In this method we make use of user based recommender system to suggest the place to new user based on the Pearson measure by comparing his ratings to the other places with other users present in the data set. It helps us to make prediction whether the user would like or dislike the unseen place as the data we are using is the travel ratings.
The language used to write the program is c++. Shell script is used to run the program. Dialog package is used in the script for the graphical windows
For the naïve bayes algorithm we have used the advertisement dataset from the Kaggle which has attributes such as time spent in the web page,age,income,internet usage. Link to the dataset : https://www.kaggle.com/fayomi/advertising/data
For the Pearson measure recommendation we have made use of a dataset from the uci machine learning repository which has the ratings of the users to different place’ Link to the dataset :https://archive.ics.uci.edu/ml/datasets/Tarvel+Review+Ratings
The code of the project is present in the following github repositary - https://github.com/deviprajwala/naive_bayes.git
The working of the project available in the below google drive link - https://drive.google.com/file/d/1O_FVC2YIyfAh_qH_zuCbn4aTmWpu7bGu/view?usp=sharing
References Recommender Systems An Introduction by Dietmar Jannach, Markus Zanker, Alexander Felfernig, Gerhard Friedrich Master Machine Learning Algorithms - Discover how they work by Jason Brownlee Machine Learning for Beginners: Make Your Own Recommender System by Oliver Theobald