Recommender Systems have been widely used in Information and Communication Technology (ICT). The main reason for this extensive use is to decrease the problem of information explosion. Collaborative Filtering techniques, which attempt to predict what information will meet a user’s needs based on data coming from similar users, are becoming increasingly popular as ways to combat information overload. Collaborative Filtering techniques usage has shown significant advantages in tourism service recommendations. Accuracy improvement of Collaborative Filtering techniques for tourism recommender systems has been an important issue in the previous studies. Therefore, this study aims to improve the recommendation accuracy of Collaborative Filtering techniques for tourism recommender systems. In this study, the method of recommendation is developed using fuzzy C-means algorithm for user-based and item-based models. Two similarity measures, Pearson Correlation and Cosine, are used for similarity calculations of users and items in both user-based and item-based models. Mean Absolute Error (MAE) is then used as an evaluation metric to show the accuracy improvement of proposed method. The experimental results on TripAdvisor dataset with several comparisons are presented to show the enhancement of proposed method predictive accuracy. The experimental results demonstrated that the user-based model of recommendation which uses fuzzy C-means algorithm remarkably improves the recommendation predictive accuracy with MAE=0.72 in relation to the item-based recommendation model with MAE=0.73. Since the proposed recommender system improves the accuracy of Collaborative Filtering techniques, the recommender system will be a promising recommendation method for item recommendation task in tourism domain.