Research in language teaching/learning has factored in many parameters like age, input mechanism and their role in ultimate attainment. Over time, specific details of the learners, like motivation, social entrenchment etc. have also been looked at. In today’s time, unsupervised, online learning environment demands customization of input as well. This customization can be arrived at through a thorough knowledge of the learners’ background and utilizing the same to classify learners into categories. A lesson plan built on this holistic approach will ensure higher success rate among students.

In this regard, this work investigates linguistic correlates of language learner diversity i.e., important factors that influence new language learning. The study comprises of analysis of data, collected through a questionnaire, on a number of linguistic factors like: learner’s language background, language usage pattern, parents' language background, community language use, along with other factors like aptitude, motivation, cognitive style, attitudes, personality and extracurricular activities. Based on the dataset generated from approximately 350 participants, a total of 54 features are extracted. Principal Components Analysis (PCA) used for feature reduction to 20 features, followed by K-mean clustering for grouping the learners into four groups: novice, beginner, intermediate and proficient. This grouping maps on the implicit study material grouping. The main contribution of the current work is the use of machine learning algorithms on a range of linguistic parameters to arrive at learner classification, thus minimizing human error and subjectivity in the process.

Authors:
Shalu Kumari, Indian Institute of Technology Guwahati, India
Sukumar Nandi, Indian Institute of Technology Guwahati, India
Bidisha Som, Indian Institute of Technology Guwahati, India

About the Presenter(s)
Shalu Kumari is currently a doctoral candidate at the CLST, IIT Guwahati. Her research interest lies in the domain of cognitive linguistics & language learning Her doctoral work is aimed at creating customizable language teaching tools.

Connect on Linkedin
https://www.linkedin.com/in/shalu-kumari-9bbb0a235

Additional website of interest
https://www.instagram.com/shalusingh.96

See this presentation on the full schedule – Monday Schedule

Bookmark