Nilani Rangika

Academic Summary

MSc in Computer Science
University of Moratuwa, Sri Lanka

BSc in Management and Information Technology
University of Kelaniya, Sri Lanka


Research Summary

My PhD research focuses on fusing multiple modalities of data such as text, visual and audio using deep learning in real-time to provide structured information for disaster response. With the increased use of mobile devices, people are now generating more data during disasters (e.g., data sharing through social media applications). Therefore, besides physical sensors and many other sources, people acting as social sensors generate a huge amount of data in different modalities during a crisis contributing to big data. However, it is beyond the capacity of the human brain to combine different modalities of data in real-time and process them to form meaningful information in a complex crisis situation. As a result, emergency responders are still mainly dependent on single sources for their decision-making which are mostly text-based (e.g., reports prepared by field officers and emails). Therefore, they don’t get access to a lot more information that could support their decision making.

I have selected Emergency Traffic Management as a particular application area to implement a multimodal deep learning framework, that could fuse data in real-time to improve situation awareness of emergency responders. We have signed a Data sharing agreement with Christchurch City Council to obtain traffic-related CCTV footage for this research.

Supervisors

  • Raj Prasanna, Joint Centre for Disaster Research, Massey University
  • Kristin Stock, Institute of Natural and Mathematical Sciences, Massey University
  • Emma Hudson-Doyle, Joint Centre for Disaster Research, Massey University
  • David Johnston, Joint Centre for Disaster Research, Massey University

Publications

[1] Nilani Algiriyage, Sanath Jayasena, and Gihan Dias. Web user profiling using hierarchical clustering with improved similarity measure. In 2015 Moratuwa Engineering Research Conference (MERCon), pages 295–300. IEEE, 2015.
DOI 10.1109/MERCon.2015.7112362

[2] Nilani Algiriyage, Sanath Jayasena, Gihan Dias, Amila Perera, and Kushan Dayananda. Identification and characterization of crawlers through analysis of web logs. In 2013 IEEE 8th International Conference on Industrial and Information Systems, pages 150–155. IEEE, 2013.
DOI 10.1109/ICIInfS.2013.6731972

[3] Nilani Algiriyage, Raj Prasanna, Emma E H Doyle, Kristin Stock, & David Johnston. (2020). Traffic Flow Estimation based on Deep Learning for Emergency Traffic Management using CCTV Images. In Amanda Hughes, Fiona McNeill, & Christopher W. Zobel (Eds.), ISCRAM 2020 Conference Proceedings – 17th International Conference on Information Systems for Crisis Response and Management (pp. 100–109). Blacksburg, VA (USA): Virginia Tech.

[4] Nilani Algiriyage, Raj Prasanna, Kristin Stock, Emma Hudson-Doyle, and David Johnston. Identifying research gap and opportunities in the use of multimodal deep learning for emergency management. 2019.

[5] Nilani Algiriyage, Rangana Sampath, Chamli Pushpakumara, and Gamini Wijayarathna. A simulation approach for reduced outpatient waiting time. In 2014 14th International Conference on Advances in ICT for Emerging Regions (ICTer), pages 128–133. IEEE, 2014.
DOI 10.1109/ICTER.2014.7083891

[6] Nilani Algiryage, Gihan Dias, and Sanath Jayasena. Distinguishing real web crawlers from fakes: Googlebot example. In 2018 Moratuwa Engineering Research Conference (MERCon), pages 13–18. IEEE, 2018.
DOI 10.1109/MERCon.2018.8421894

[7] N Algiriyage. Detecting access patterns through analysis of web logs. MSc thesis, 2015.