SADAATL 2014 - Synchronic and Diachronic Approaches to Analyzing Technical Language (SADAATL)

Technology is the application of knowledge to practical pursuits. Information relevant to technology is the subject of various types of documents, including: scholarly publications (journals, conference proceedings, abstracts, grant applications, textbooks); legal documents (patents, contracts, legislation); and more public venues (magazines, webpages, blogs, financial reports). Interest in the automatic classification of technical documents has recently been growing and Natural Language Processing is a major component of such classification systems. On a synchronic level, there has been considerable efforts towards: genre classification, citation sentiment, relation extraction, terminology extraction, and other areas. On the diachronic level, the field of technology forecasting is on the rise because both government agencies and businesses are looking to automatically detect and classify trends in technology, in order to help guide them with resource and financial investments.
SADAATL 2014 workshop aims to bring together natural language processing research applying to technical documents. The goal is to explore techniques which apply across multiple domains and genres (and are not biased towards biomedical, computer science, or other specific genres). The results should either be: a) synchronic in nature, relating to the processing or analysis of text, documents or sets of documents without consideration of time; or b) diachronic in nature, investigating how linguistic features change over time, e.g., by the comparison of documents from different time periods.
Topics
Subject areas include topics specific to the study of technical documents such as:
citation extraction
terminology extraction
citation analysis
technology forecasting
document analysis
as well as other NLP areas applied to technical documents such as:
relation/event extraction
named entity extraction
sentiment analysis
machine translation
Another major focus of this workshop is to explore how synchronic and diachronic topics relate to each other. For example, we encourage synchronic papers to discuss how and why such features may vary over time, e.g., trends in sentiment attributed to citations may indicate changes in the status of a paper. Diachronic papers are encouraged to discuss any synchronic features used that have been tracked over time.