Computing and Informatics

With the growing demand for personalised travel experiences, the development and application of travel point-of-interest (POI) recommendation systems have become increasingly important. However, many existing systems often underperform owing to insufficient integration of social features and contextual information. To address this issue, the S-LMF algorithm is proposed, combining social features with logistic matrix factorisation to improve recommendation accuracy. This approach simulates social influence by incorporating joint check-in similarity and friendship factors, while logistic matrix factorisation leverages check-in frequency data to refine POI recommendations. The effectiveness of social features and logistic matrix factorisation (S-LMF) was tested against five baseline algorithms using publicly available data sets from Yelp and Gowalla. Results demonstrated that S-LMF outperformed the best baseline model by improving \mathrmPrecision@20 by 22.95 % on Yelp and 28.60 % on Gowalla. Moreover, it increased \mathrmRecall@10 by 17.95 % on Yelp and 8.19 % on Gowalla.