99814 - CORPORA, LINGUISTICS AND TECHNOLOGY LAB

Learning outcomes

The student is able to independently identify a research problem of relevance to a linguistics- and/or translation- field, with special reference to the use of text corpora; s/he is able to locate and efficiently use the tools and information sources needed to tackle a research problem; s/he is able to acquire further competences related to linguistics and translation, as well as other disciplines of relevance to her/his studies, through interaction with scholars from a range of fields

Course contents

The activities of this seminar revolve around the research carried out as part of the PRIN project "UNITE –UNiversally Inclusive Technologies to practice English".

The aim of UNITE is to explore the applications of Dialogue Systems (DS) as agents to practice English as a Foreign Language (EFL). DS include, among others, chatbots based on large language model such as ChatGPT and Bard. The objective of the project is to produce teaching and learning materials which might favour the uptake of DS as tools for autonomous language learning by university students, including those with disability and specific learning disorders.

Course participants will be actively involved in all the activities carried out within the project, including:

• an overview of existing DS and their classification based on parameters suggested in the literature;
• the design of a strategy and materials to recruit participants in the interaction experiments between learners and DS;
• piloting of interaction experiments;
• creation of an annotated corpus representative of written interactions between learners and DS;
• linguistic analysis of the annotated corpus.

The course includes frontal lectures, part of which will be delivered by guest speakers, focusing on topics such as Universal Design for Learning, studies on DS for language learning, and corpus-based analysis of learner language.

The course is taught jointly with Daniele Polizzi.

Readings/Bibliography

• Ayedoun, E., Hayashi, Y., & Seta, K. 2015. A conversational agent to encourage willingness to communicate in the context of English as a Foreign Language. International Conference on Knowledge-Based Intelligent Information & Engineering Systems.
  

• CAST. 2018. Universal Design for Learning Guidelines (version 2.2). http://udlguidelines.cast.org
  
• Gilquin, G. 2020. Learner corpora. In M. Paquot and S.Th. Gries (ads.), A Practical Handbook of Corpus Linguistics. Springer, 283-304.

Teaching methods

The seminar combines frontal lectures and practice-based sessions.

Theoretical and methodological contents are delivered through presentations by the lecturer.

The practical sessions consist of hands-on activities, which are carried out collaboratively in class or individually as assignments, and are followed by group discussion. The activities are aimed at constantly monitoring progress in the development of the research and technological skills that make the object of the seminar.

All students must attend Module 1 and 2 on Health and Safety online.

Assessment methods

Assessment will be partly in itinere and partly based on a final project.

The in itinere assessment is based on the quality of students' assignments carried out on a weekly basis, as well as the continuity with which they are handed in. For those who do not hand in the minimum number of assignments (3 out of 5), the final mark is entirely based on the final project.

The final project, whose contents and format will be agreed upon with the teacher, will focus on one of the topics addressed in class (see "Course contents" section) and can be written in the format that is deemed most suitable (e.g. a technical report, an oral presentation with slides, or a blog post aimed at a lay audience).

Assessment criteria:

• 30-30L: excellent project work, displaying very high technical quality, presented in a very clear and detailed way.
• 28-29: excellent project work, displaying high technical quality, presented in a clear and detailed way.
• 26-27: good project work, displaying adequate technical quality; some aspects of the presentation could be improved.
  
• 23-25: fair project work, displaying technical shortcomings; some aspects of the presentation could be improved.
• 20-22: project work displays technical shortcomings; presentation has gaps or is unclear at several points.
• 18-20: project work displays several major technical shortcomings; presentation has many gaps and/or is mostly unclear. 
• <18: insufficient or severely insufficient project work.

Teaching tools

Hardware: PC and overhead projector.

Software: Web-based and mobile Dialogue Systems; text editors to process textual data; graphics software to create materials for participant recruitment; corpus consultation software.

Office hours

See the website of Adriano Ferraresi