Sharing geotagged pictures for an Emotion-based Recommender System. Hitz, A., Naas, S., & Sigg, S. In The 19th International Conference on Pervasive Computing and Communications (PerCom 2021), adjunct, 2021.
abstract   bibtex   
Recommender systems are prominently used for movie or app recommendation or in e-commerce by considering profiles, past preferences and increasingly also further personalized measures. We designed and implemented an emotion-based recommender system for city visitors that takes into account user emotion and user location for the recommendation process. We conducted a comparative study between the emotion-based recommender system and recommender systems based on traditional measures. Our evaluation study involved 28 participators and the experiments showed that the emotion-based recommender system increased the average rating of the recommendation by almost 19%. We conclude that the use of emotion can significantly improve the results and especially their level of personalization.
@inproceedings{hitz2020Emotion,
  title={Sharing geotagged pictures for an Emotion-based Recommender System},
  author={Andreas Hitz and Si-Ahmed Naas and Stephan Sigg},
  booktitle={The 19th International Conference on Pervasive Computing and Communications (PerCom 2021), adjunct},
  year={2021},
  abstract={Recommender systems are prominently used for movie or app recommendation or in e-commerce by considering profiles, past preferences and increasingly also further personalized measures. We designed and implemented an emotion-based recommender system for city visitors that takes into account user emotion and user location for the recommendation process. We conducted a comparative study between the emotion-based recommender system and recommender systems based on traditional measures. Our evaluation study involved 28 participators and the experiments showed that the emotion-based recommender system increased the average rating of the recommendation by almost 19%. We conclude that the use of emotion can significantly improve the results and especially their level of personalization.
},
  group = {ambience}
}


%%% 2020 %%%
Downloads: 0