Thanks for visiting...

Dr Le An Ha

L/SL in Commercial Exploration

Research Institute in Information and Language Processing

RIILP - RGCL - University of Wolverhampton, UK

Ha.L.A@wlv.ac.uk

The subject of my PhD thesis is automatic terminology processing. In my PhD, “knowledge patterns”, which are extracted (semi)automatically from resources, are used to improve the performance of a terminology processing system. I have been working and publishing on a variety of subjects, including automatic terminology extraction, both monolingual and multilingual, multiple-choice question generation, analysis of multiple-choice test items, and multilingual preprocessing. I have extensive experience in developing commercial Natural Language Processing (NLP) vertical solutions. I have acted as an acting coordinator of an EU Leonardo project (TELLME) which developed a range of products including work-related language exercises, showcasing NLP technologies such as automatic term extraction and MCQ generation. I have developed a Computer-Aided Patient Notes Scoring System for a well-known US medical examiners organisation. I have been leading research activities in the domain of applying NLP technologies for licensing testing, funded by an US organization on a yearly rolling research contract, including American-British transliteration, information extraction, item difficulty prediction, item response time prediction, and item distractor prediction. I have experiences on utilizing eye tracking data for various tasks. I have extensive experiences in machine learning techniques algorithms, and platforms, both traditional (svm, linear regression, random forest, weka etc.) and recently developed (deep learning, tensorflow, keras).

I have taken part in the teaching of an European Master Programme in Computational Linguistics and currently supervising two PhD student (I has supervised three completed PhD projects). I also have a degree in Economy (specialised in International Trade).

Research interests

Multiple choice question generation; Multiple choice question analysis; Machine learning; Deep learning; NLP applications;