>> Download complete programme (.doc)

The Padua group's task can be divided into four main parts:

1) creation of computerized corpora;
2) linguistic analysis of the texts in the corpora;
3) use of corpora and analyses to create hypermedia teaching units for English Language and for Italian-English translation;
4) creation of computerized systems for testing receptive skills (reading and listening), summarizing, translating and textual comprehension skills in the widest sense, as well as metalinguistic and metatextual skills.

The Padua group's first task will be the creation of corpora of texts in English or in English and Italian (parallel and comparable corpora)ranging over many genres and discourse types: proverbs, advertising and other promotion texts, interviews and political statements, business letters, job application letters, technical instructions, recipes, class interaction, different types of texts in the fields of engineering, medicine, psychology (Abstracts, research articles, textbooks), agriculture, journalism and literature. These corpora will be shared with the other units in the project and,where possible, (questions of copyright, etc.) will be made available to the scientific community over the network. Analyses of various kinds will be carried out on the texts in order to highlight their characteristics and genres. The results of the analyses will be made available on line to other members of the project and, where possible, to all interested parties.

Different types of analysis will be performed to refine the method and evaluate the extent to which the results obtained are useful and efficient in the various applications. Analyses based on lexical-grammatical aspects will illustrate grammatical intricacy, lexical density, the presence of and the different types of projection as direct and indirect speech, the theme and information structure, the presence of interpersonal markers, tenses and aspect, use of articles. Other analyses will be concerned with grammatical and lexical cohesion, and generic structure, and, where oral discourse is concerned, functional analysis of speech acts and interactive exchanges. The analyses of parallel and comparable texts in Italian and English will also deal with questions of contrastivity.

One of the main aims of research of the Padova group is to provide valid applications for corpora, seen as a wealth of authentic written, spoken and multimodal language, to language learning and testing and to translation. Other analyses will allow evaluation of textual difficulty (lexical, syntactic, cohesion and content) and this will help to determine at what level these texts should be placed in the course of language learning and testing. We intend to make use of data bases available on the net, the Cobuild corpus and relative systems for concordancing and and research, English and Italian newspapers to integrate and extend our own corpora. Some of the texts analyzed will be inserted into computerized teaching units and made available on the WEB, which will allow access in hypertext to other texts with similar phenomena and to the theory and practice related to the linguistic phenomena in focus. For example the business letter corpus will become the basis for specific units on how to write texts in this genre which focus on the use of tense and aspect in this type of text. Materials from various corpora will be exploited to create hypertext a teaching module on the use of the article in English.

For testing usage of articles in English, rather than the invented sentences that are so often used, we intend to base these tests on the texts of our corpora. There will also be materials created to help summary writing and translation, with hypertextual guides of different types in order to attain the goal from different directions. Short oral tests will be included in the testing as well as written and multimedial programs that can be used on the WEB and which make it possible to match the texts with various types of comprehension questions (multiple choice; gap-filling; matching; true/false; open answers). Some areas of the project can be achieved as parts of degree theses and supervised by members of the project.

The research has been planned to cover four phases of six months each:

>> phase 1 - creation of new corpora, tagging existing ones;

>> phase 2 - analysis and tagging of the texts of new corpora; creation of other hypermedial materials for learning and testing;

>> phase 3 - collection and evaluation of data from text analyses; piloting computerized materials with students from the various Faculties of the University of Padova for the purposes of self-access, or collective learning and testing;

>> phase 4 - elaboration and expansion of hypermedial products created for testing; addition of more texts to corpora or creation of new corpora; new types of analysis and tagging, according to the needs arising as the teaching and testing modules are created and experimented.

At the conclusion of each phase the ongoing results will be discussed with partners from the other universities involve in the project in the course of a joint study day. At the end of the project the results will be presented at meetings and in publications.

Back to top