Al seminario puede asistir cualquier miembro de la facultad. Se realizará en el laboratorio 1.7 y habrá una parte teórica y una pequeña parte práctica. Para la parte práctica se proporcionará una máquina virtual de VirtualBox que tendrá lo necesario para la parte práctica.
RESUMEN: People have to make decisions every day. The demand of the experience of other people through the request of their opinion is a common tool to ease the decision making process. An opinion is the expression of a personal state, which can be related to a thought, a view concerning a specific topic, a past experience or even an emotional state. The NLP task that is concerned with the processing of opinions is Sentiment Analysis. Its aim is is the identification, processing, classification and visualization of opinions published on different media. The advent of Web 2.0 and specifically social networks has favoured the publication and the access to opinions about any topic. Twitter is a microblogging social network in which opinions about topics are posted in real time. The live character of Twitter is its strength and the source of the relevance of the task of analysing and extracting the information about the opinions that are constantly posted. This session will cover the fundamentals of Sentiment Analysis; moreover, a practical exercise will be developed so that students may apply the knowledge acquired in the first part of the session.
PONENTE: Eugenio Martínez Cámara is a postdoctoral researcher at the University of Granada. He obtained his PhD in Computer Science at the University of Jaén (Spain), and thhe topic of his thesis is Sentiment Analysis in Spanish. He worked as postdoctoral researcher in Technishe Universität Darmstadt (Germany), where he worked in the computational treatment of causal relations. Currently, he works on Sentiment Analysis at Aspect Level in the University of Granada. Eugenio is one of the orgnizers of TASS, a workshop on semantic analysis in Spanish, which annualy co-located with the International Conference of the Spanish Society of Natural Language Processing.
DURACIÓN TOTAL ESTIMADA: 180 minutos