The remarkable development of language technology tools in recent years in terms of robustness, computational speed and volume of processed data, together with the increasing number of languages covered, made possible their usage not only for specific research applications, but also for real world applications which prove useful in everyday life. Automatic correction of text, machine translation, extraction of important information and hands-free devices are just some of these applications. Language technology now has the maturity to be used for addressing societal challenges such as helping people with disabilities, the elderly and migrants.
However, due to the ambiguity and complexity of natural language, its automatic processing is still very challenging and benefits from processing shorter and less ambiguous information. The same is true for people who have difficulties understanding text due to disabilities, or who have to read texts in a language they do not have a good command of. In all these cases, automatic text simplification can prove to be very useful.
In contrast to controlled languages, which practically create a sublanguage by imposing constraints on the grammar rules, discourse style, number of words/sentence etc., text simplification eliminates or replaces parts of sentences or paragraphs, or even reformulates them according to specific requirements of the target user groups. Among the most frequent techniques are: lexical substitution, verb forms replacement (for morphologically rich languages), word order adjustments, deletion of subordinate clauses, replacement of anaphoric pronouns by their reference, usage of synonym expressions with higher frequency as well as compound splitting.
This workshop intends to bring together scientists working in a variety of fields in which text simplification can be applied, computational linguists interested in the research problems of text simplification and of course users who can benefit from the simplified texts.
The innovative aspect of this workshop will be the discussion on text simplification from two perspectives: On the one hand, how computational linguistics applications which simplify texts can be used by people in real world situation, and on the other hand, how to simplify the input for other NLP-based applications in order to improve their accuracy. We also intend to develop a roadmap of activities, tools and resources on text simplification from a multilingual perspective. A special attention will be paid to contributions describing automatic text simplification for less-resourced languages.
Papers are invited on the following directions:
Topics of interest include but are not limited to: