Social media analytics is the practice of gathering data from user generated content (UGC) such as blogs and social media websites to have an aggregated view of the opinion of the crowd and provide personalized services. The most common use of social media analytics is to mine customer sentiments and customer profiles, in order to support marketing and customer or community service activities.


Opinion mining (or sentiment analysis) is the field of study that analyzes people's opinions, sentiments, evaluations, attitudes, and emotions from written language. It has recently become one of the most active research areas in natural language processing and is also widely studied in data mining, Web mining, and text mining.

Customer profiling analyzes a user's posted content, social activity and social networks and applies data analytics techniques to discover a user's related information such as gender, age, education, personality, interests, etc.


We work on different social media analytics projects and develop systems to address related problems, such as opinion polarity classification, subjectivity detection, aspect based sentiment analysis. For instance, polarity classification helps determining if a text, e.g. a tweet, is positive, negative, or neutral, while aspect based sentiment analysis detects which aspects of a product a user has liked or disliked in an on-line review. We actively participate in international competitions on the topic, such as SemEval.

Examples of applications are text analytics for customer experience, multilingual social media monitoring for customer engagement, e-reputation analysis or public health monitoring.