1st International Workshop on
Language Diversity and the Acquisition of Linguistic-Semantic Knowledge - LDALSK
The proportion of Web pages written in English keeps decreasing, but a majority of text-handling tools and techniques have been developed for English and tested on English resources. Many languages have been reasonably well served by the existing systems, but large typological differences (complex morphology, rich inflection, freer-order syntax, not to mention cultural biases) tend to make those tools and resources less than fully adequate. This may be particularly true of deeper processing, including the acquisition of natural language semantics from text.
The workshop seeks to explore methodologies developed from scratch for a variety of languages, and perhaps their reapplication to the processing of English texts. We invite submissions on the topics we list below, and on closely related topics in the general spirit of the workshop. We target the problem of automatic extraction of some form of natural language semantics from corpora and application of the knowledge extracted in that way in systems that deal with natural language.
Work on any language is welcome; we particularly encourage submissions related to "less privileged" languages. The workshop will be held in English, to ensure the broadest dissemination of its contributions. We have room for up to 50 papers, including posters, so there will be ample opportunity to report on work in progress. To suit the host conference, papers on applications of the semantic knowledge extracted from text will also be considered; such applications include machine translation, parsing, sentiment analysis, information retrieval, information extraction, and text mining.
All submissions must present original work. There will be a strict reviewing process.