From the perspective of lexical structure, the term 'computational lexicography' is a modifier-head construction. It can be interpreted as the study of lexicographical theory and practice by means of computational technology. Thus, computational lexicography should certainly function within the framework of traditional lexicography, but with the focus put on a new lexicographical methodology based on modern technology. Undoubtedly, the development of information technology and multi-media provide excellent tools for lexicographical study and practice.

In fact, the computer creates favourable conditions for such aspects of lexicography as data storage, extraction, analysis, transmission, and exchange, as well as corpus construction and dictionary compilation. Moreover, large databases or electronic dictionaries are needed to enhance data processing ability in natural language processing (NLP). All these external causes play a key role in the emergence and development of computational lexicography. In this sense, 'computational lexicography' refers both to the development of machine-readable dictionaries based on a printed version and a lexicographic database for computer use.

Many authors (Amsler 1982: 661; Ooi 1998: 1–2; Hartmamn and Gregory 2000; Bennett et al. 1986: 26) provide various definitions of computational lexicography (cf. Zhang Yihua 2004: 13-14). From the historical viewpoint of its evolution, computational lexicography should first deal with the electronization and machine-readability of the lexical knowledge of the printed dictionary, then study of automatic reading, recognition, conversion, and exchange of lexicographic data by computers.

Study must now be done on computer-aided compilation, editing, and revising of dictionaries, aiming at semi-automation in the near future, with the ultimate goal to realize automation throughout the whole process of dictionary-making and publishing. And last, attention should be given to the electronic adaptation of commercial dictionaries, or the design or compilation of electronic dictionaries or online dictionaries for human use.

With respect to lexicographic data processing and compiling digitalization, the analysis of authentic continuous texts, the index, and the extraction of lexical data should be taken into consideration, besides that of computer-aided dictionary compilation. The distribution of related lexical items in natural discourses can be investigated and analysed by the means of an index, so as to examine the functional attributes of various aspects such as grammar, semantics, and pragmatics, and acquire useful features concerning the function and usage of each lexical item. Simultaneously, the data of sense distribution and division can also be obtained through text analysis.

In general, computational lexicography deals with the study of the electronization of corpus processing, semi-automation or even total automation of dictionary compilation, formalization of microstructure arrangement, digitalization of dictionary media, intellectualization of the dictionary query, and integration of multimedia into lexical data representation. The major content of study includes language data collection and processing, sense-division support, comprehensive semantic analysis, illustrative example extraction, computer-aided dictionary compilation, lexicographic database construction, corpus and database management, (semi)-automatic dictionary generation or production, lexicographic data statistics, dictionary compilation management, and data export interface.

责任编辑：admin