Computational linguistics and beyond: Perspectives at the beginning of the 21st century (review)

The papers in this volume, which represent plenary lectures and other papers from COLING 2002, explore the frontiers of computational linguistics. The book contains five articles written by pioneers whose works play a leading role in this field.

Charles Fillmore, Josef Ruppenhofer, and Collin Baker, in ‘FrameNet and representing the link between semantic and syntactic relations’, explain the goals, procedures, and the final product of FrameNet. They then reveal how the FrameNet data can be utilized as a research tool for a particular derivative task.

‘Computational studies of language evolution’, by William Wang, Jinyun Ke, and James Minett, explores the perspective that claims that language is the shared endowment of the human species. Using various mathematical modeling and simulations, the authors discuss several computational studies in language change and language emergence.

‘New chances for deep linguistic processing’, by Hans Uszkoreit, is concerned with how deep linguistic processing could be successfully integrated and further developed. Disregarding the trends of a statistically based system, the author combines deep processing methods with shallow ones. He claims that information extraction will be the leading research paradigm in computational linguistics and that future development relies on massive international cooperation.

‘The role of natural language and XML in the semantic web’ has six sections, each written independently. As a whole, this article aims primarily to look ahead to future development of the semantic web. The papers demonstrate that how we use the web will be affected by natural-language-processing technology with higher annotation. They also claim that the web with semantically annotated content will play a vital role in the future.

‘Chinese language processing at the dawn of the 21st century’, by Benjamin T’sou, chronologically reviews milestones of Chinese language processing (CLP) development from 1992 to 2002. It starts with three hurdles in CLP that were not dealt with previously: input problems, word segmentation problems, and syntax. While most problems are overcome, some remain a challenge. T’sou claims that solving these problems and issues of user demands will shape CLP in the next decade (2002–2012).

Despite the technical nature of the discussion, the book is informative and well written. The editors’ concise introductions to each article are especially beneficial for readers trying to grasp the background and the general idea of each paper. Also, the biographical sketch of each author helps the reader to understand where their ideas come from, and where related research might be found. Also, the issues raised are undoubtedly of interest to linguists who desire to broaden their knowledge of computational linguistics.