In continuing to explore more benefits in reading, creating and sharing the interactive fiction, a web application called Vixio has been developed as a platform, where users can develop and distribute interactive fiction. In order to engage and to feed the users with more interactive story, a recommender system is applied to provide the recommendation stories that suitable to the reader interest. This work is focused on developing a recommender system which can generate story recommendations for Vixio web application. This work will determine which techniques is better to be implemented inside the recommender system by conducting a comparative study between five collaborative filtering techniques, which are: Three Matrix Factorizations (SVD, SVD++ and NMF), Slope One and Co-clustering. To compare each technique with one another, 5-fold cross validation and response time were being measured. Based on these two evaluations, it is expected to this research will show the technique that has better accuracy compared to others.

Recommender System. SVD, SVD++, NMF, Slope One, Co-clustering

• Ida Bagus Kerthyayana Manuaba, S.T., Ph.D.