Hotel Recommendation System Based On Collaborative Filtering And Implementation

In the age of Web 2.0, the amount of information is increasing far more quickly than our ability to process it. Both the E-Commerce sites and the customers face the serious problem of information overload. A new technology that can help us sift through the available information to find which is most valuable to us is in urgent need. Thus, recommender systems have become an important and one of the hotest research area in the internet research. We study the hotel recommender systems based-on an E-Commerce travelling site.The recommender system has two parts, the search page part and the detail page part, according to how we use the site. We propose a heuristic algorithm based on hotel conversion rate for the search page after we look deep into the user's behavior and the hotel click data. Moreover, we try to apply the collaborative filtering algorithms in detail page recommendations. We also compare the difference and quality between the item to item and item-based collaborative filtering algorithms.Experiments are designed to validate these algorithms. The results prove that the heuristic and the item to item collaborative filtering algorithms were both effective in our hotel data sets. Besides, the website would have big progress in enhancing the conversion rate if these algorithms were implemented.