Comparing accuracy of cosine-based similarity and correlation-based similarity algorithms in tourism recommender systems

• Published in: 2008 4th IEEE International Conference on Management of Innovation and Technology pp. 469 - 474
• Main Authors: Bigdeli, E., Bahmani, Z.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.09.2008
• Subjects: Accuracy; Algorithm design and analysis; Collaboration; Collaborative filtering; correlation-based similarity algorithms; cosine-based similarity algorithm; e-tourism; Filtering; Industries; Prediction algorithms; recommender systems
• ISBN: 9781424423293; 1424423295
• DOI: 10.1109/ICMIT.2008.4654410

Recommender system has a long history as a successful application in artificial intelligence. A growth in the number of products, which has been offered by different e-commerce platforms, leads to a technology which can help customers to choose and buy products. Collaborative filtering or recommender systems use a database about user preferences to predict additional topics or products a new user might like. This paper describes some algorithms designed for this task including cosine-based similarity algorithm and correlation-based similarity algorithm. The predictive accuracy of various methods in tourism recommender domains is compared. On the other hand, we have designed and implemented a recommender system in e-tourism in order to compare performance of these algorithms. Finally, we conclude that correlation based similarity algorithm acts better than Cosine based similarity algorithm.