This thesis presents a systematic literature review (SLR) of digital game recommender systems, focusing on the current state of recommender system technology within the video game domain. While recommender systems are widely studied in areas such as e-commerce, streaming services and social media, their application to digital games remains comparatively underexplored. Following the PRISMA 2020 guidelines (Page et al., 2021), literature was collected from various databases such as ACM Digital Library, IEEE Xplore and Web of Science. After screening, 255 records were identified and 22 studies were included in the final analysis. 

The findings show that research on game recommender systems is methodologically diverse, with content-based filtering, collaborative filtering and hybrid approaches being most common techniques. Many studies rely on Steam-related datasets, including game metadata, user interaction data, and review data. Model performance is a concern across the literature, with evaluation metrics such as RMSE, MAE, NDCG and Percision@K being one of the most used to compare recommendation quality. However, the review also shows that no single recommendation method performs best in all contexts. Instead, performance depends heavily on the available data, feature extraction/selection and evaluation method. 

The review identified several recurring challenges in digital game recommender systems such as cold start problems, data sparsity and limited user information. Hybrid and more complex models show strong potential for improving recommendation quality however, they also require more complex data integration and platform specific implementation. Overall, the study concludes that digital game recommender systems are developing toward more advanced and context-sensitive approaches, although further research is needed to establish standardized evaluation frameworks and better understand how recommender systems affect players, diversity and game discovery.