2nd Emotions in Recommender Systems workshop at UMAP 2014
Sprache des Titels:
Personality and emotions shape our daily lives by having a strong influence on our preferences, decisions and behaviour in general. Hence, personalized systems that want to adapt to end users need to be aware of the user?s personality and/or emotions to perform well. Affective factors may include longterm personality traits or shorterterm states ranging from ?affect dispositions?, ?attitudes? (liking, loving, hating,?), ?interpersonal stances? (distant, cold, warm,?), ?moods? (cheerful, irritable, depressed,?) or ?real emotions?.
Recently, there have been extensive studies on the role of personality on user preferences, gaming styles and learning styles. Furthermore, some studies showed that it is possible to extract personality information about a user without annoying questionnaires, by analyzing the publicly available user?s social media feeds. Also, the affective computing community has developed sophisticated techniques that allow for accurate and unobtrusive emotion detection. Generally, emotions can be used in personalized systems in two ways: (i) either to change the emotion (or mood, e.g. from a negative to a positive) or (ii) to sustain the current emotion (e.g. keep a user ?charged? while doing sports). Recent studies showed that such information can be used in various personalized systems like emotionaware recommender systems.