Analysing and visualising free-text comments:a corpus-based toolkit

Knight, Dawn and Rayson, Paul and Khallaf, Nouran and Morris, Steve and El-Haj, Mahmoud and Ezeani, Ignatius (2023) Analysing and visualising free-text comments:a corpus-based toolkit. In: Corpus Linguistics, 2023-07-022023-07-06, Lancaster University.

Full text not available from this repository.

Abstract

Free-text qualitative comments (e.g. from surveys and questionnaires), pose a particular challenge to a range of companies/institutions, who may not have the expertise to analyse these comments with ease. Following the Welsh Language (Wales) Measure (2011), survey respondents in Wales should be given the opportunity to respond to surveys in English or Welsh, posing even more of a challenge when analysing the resultant data, if adequate Welsh language expertise do not exist. Although a range of sophisticated tools for the analysis of text-based data are already available, many of these tools are not necessarily affordable, quick and easy to use, and/or accessible to non-expert user, nor do they fully support the task of systematically processing free-text responses in Welsh and English. This presentation reports on the developments of a unique open-source online free-text analysis tool that has been designed to respond to this need: FreeTxt. Funded by the AHRC, and co-designed/co-developed in collaboration with project partners National Museum Wales, National Trust Wales and Cadw, FreeTxt is a unique corpus-based analysis toolkit that is designed to enable the quick and easy analysis of English and Welsh language data, and to engage new user groups with corpus-based methods in new ways. In this presentation we will: - underline the importance of user feedback, and articulate the key challenges of tackling such data, - present a novel corpus-based approach to the analysis of FreeTxt data, which can be adapted to multiple languages and contexts, - outline the key functionalities of the tool, which include: KWIC, POS tagging, semantic tagging, summarisation and sentiment analysis utilities, a novel n-gram frequency tool, text visualisation and multilingual support, and - provide a demonstration of an early version of the FreeTxt tool in action, using data from survey responses and online feedback forums.

Item Type:
Contribution to Conference (Paper)
Journal or Publication Title:
Corpus Linguistics
ID Code:
199799
Deposited By:
Deposited On:
13 Sep 2023 10:15
Refereed?:
Yes
Published?:
Published
Last Modified:
13 Sep 2023 10:15