ABSTRACT:The web has become an essential resource for obtaining information associated with any topic or domain. The amount of text produced by interactions on social media, blogs, URLs, etc., has made essential to use advanced techniques to be able to understand and obtain valuable information from these large volumes of data. Considering the complexity and richness of web information, the use of graphs in text classication tasks and topics text analysis is a growing area of study. Its aim is to discover novel and insightful knowledge from data that is represented as a graph. The use of this kind of graph techniques on text documents has a wide range of applications like social science, homeland security, finances, healthcare, web analysis, linguistics, etc. This is mainly because graphs are often the most natural way to represent the connections among data and the increasing number of tools available to handle these types of structures. During the talk, the research approach followed to extract valuable knowledge from textual information combining statistical analyis and graph based analysis techniques will be described. A case of study on text related to the American Presidential Elections that took place in Novemebre 2016, will also be presented.
LANGUAGE:English
