Mining text streams

摘要: Natural Language Processing (NLP) and Text Mining (TM) are active research fields that aim to derive information from text and transform the language used by humans to a form that a computer is able to comprehend. The semi-structured nature of text makes the task of mining information from text sources a very challenging task that after several decades is still an open research problem. Some of the most notable applications of text mining are ad hoc information retrieval, text categorization, keyword extraction, and document summarization.During the last decade the rise of the social media, news portals, websites as well as a variety of different applications resulted in massive streams of text data which are continuously produced over time. Many of the traditional text mining techniques cannot handle effectively or efficiently these streams and produce an output in real-time. My research during the last years focused on mining knowledge from text streams in real-time and focused on applications like keyword extraction, event detection, document similarity, and text summarization.