Human languages are complex by nature and efforts in pure symbolic approaches alone have been unable to provide fully satisfying results. Text Mining and Machine Learning techniques applied to texts (raw or annotated) brought up new insights and completely shifted the approaches to Human Language Technologies. Both approaches, symbolic and statistically based, when duly integrated, have shown capabilities to bridge the gap between language theories and effective use of languages, and can enable important applications in real-world heterogeneous environment such as the Web.
The most natural form of written information is raw, unstructured text. The huge amount of this kind of textual information circulating in the Internet nowadays (in an increasing number of different languages) leads us to use and investigate systems, algorithms and models for mining texts. As a consequence, Text Mining is an active research area that is continuously broadening worldwide and fostering reinforced interest in languages other than the most common ones such as English, French, German and now Chinese. This 5th Biannual Track of Text Mining and Applications will provide, as in previous editions of the TeMA Tracks within the EPIA Conferences, a venue for researchers to present and share their work in intelligent computational technologies applied to written human languages. TeMA 2015 is a forum for researchers working in Human Language Technologies i.e. Natural Language Processing (NLP), Computational Linguistics (CL), Natural Language Engineering (NLE), Text Mining (TM) and related areas.