Mrs. Charini Nanayakkara

I am currently working as a PhD student at the Australian National University, where the focus of my research work is, Record Linkage of Complex Historical Birth, Marriage, Death and Census Data. Record linkage is the task of identifying records in a dataset which relate to the same entity. It is an important step in cleaning datasets (for duplicate removal) and extensively used for tasks such as population reconstruction. In the absence of unique identifiers per records, a record linkage method must be used to identify whether two records relate to the same entity. This task is particularly complicated for historic datasets due to the many mistakes introduced to them at the time of data entry and transcription. The Scottish dataset which I am currently experimenting with is a typical example for the challenges posed by historic datasets, due to the errors in names, dates, the imbalance in name distribution (many people sharing common names and few very uncommon names), etc. My focus is on developing scalable record linkage techniques using which high quality historical record linkage can be achieved.

 

Education:

  • BSc. (Hons) Computer Science with first class from University of Colombo School of Computing, SriLanka (2016).

Work Experience:

  • Software Engineer at WSO2 (Jan 2016 - Jan 2018)

Awards & Scholarships:

  • ANU PhD Scholarship (International) (2018)
  • ANU HDR Merit Scholarship (2018)

Publications:

  • Robust temporal graph clustering for group record linkage, Charini Nanayakkara, Peter Christen, and Thilina Ranbaduge, Springer PAKDD 2019, Macau, China, DOI: https://doi.org/10.1007/978-3-030-16145-3_41
  • Temporal graph-based clustering for historical record linkage, Charini Nanayakkara, Peter Christen, and Thilina Ranbaduge, 14th International Workshop on Mining and Learning with Graphs (MLG) Workshop, held at ACM SIGKDD 2018, London, UK. Available: https://arxiv.org/pdf/1807.02262.pdf
  • Identification of Musically Induced Emotion: A Machine Learning Based Approach, Charini Nanayakkara and Amitha Caldera, 3rd International Conference on Data Mining, Internet Computing, and Big Data, Konya, Turkey 2016
  • Music Emotion Recognition with Audio and Lyrics Features, Charini Nanayakkara and Amitha Caldera, International Journal of Digital Information and Wireless Communications (IJDIWC), Vol. 6, No. 4, 260-273 (Oct 2016)
  • Data Mining and Pattern Recognition
  • Record Linkage
  • Machine Learning
  • Robust temporal graph clustering for group record linkage, Charini Nanayakkara, Peter Christen, and Thilina Ranbaduge, Springer PAKDD 2019, Macau, China, DOI: https://doi.org/10.1007/978-3-030-16145-3_41
  • Temporal graph-based clustering for historical record linkage, Charini Nanayakkara, Peter Christen, and Thilina Ranbaduge, 14th International Workshop on Mining and Learning with Graphs (MLG) Workshop, held at ACM SIGKDD 2018, London, UK. Available: https://arxiv.org/pdf/1807.02262.pdf
  • Identification of Musically Induced Emotion: A Machine Learning Based Approach, Charini Nanayakkara and Amitha Caldera, 3rd International Conference on Data Mining, Internet Computing, and Big Data, Konya, Turkey 2016
  • Music Emotion Recognition with Audio and Lyrics Features, Charini Nanayakkara and Amitha Caldera, International Journal of Digital Information and Wireless Communications (IJDIWC), Vol. 6, No. 4, 260-273 (Oct 2016)

Updated:  1 June 2019/Responsible Officer:  Dean, CECS/Page Contact:  CECS Marketing