Nowadays recommender  system is the hot cake of every E-commerce and Entertainment sites. This project builds the underlying recommender system for a tourism site so that it can customize the list of places viewed by the user and recommend a tour which is expected to be liked by him. Firstly, I have done a survey on different users by asking real users to enter the choice of places and ratings through a Google form. Collecting data from real users will have less chance of biasness and also give the feel of real life dataset which needs to be cleansed and explored. After cleaning the data I have applied a hybrid recommeder algorithm which comprises of both content-based filtering and collaborative filtering. You can refer to the screenshots to get an overview of the algorithm. I have uniquely merged both the filtering algorithms in order to get better results.

Content-based Filtering– Recommends items based on a comparison between the content of the items and a user profile. Recommends items based on comparison of similarity among the different tags associated with every item.

Collaborative Filtering–  A technique that can filter out items that a user might like on the basis of reactions by similar users. It works by searching a large group of people and finding a smaller set of users with tastes similar to a particular user.

The results obtained using hybrid filtering are quite promising. Some of the results have been made available in the screenshot.