Skip to main content

Notice

The new RDA web platform is still being rolled out. Existing RDA members PLEASE REACTIVATE YOUR ACCOUNT using this link: https://rda-login.wicketcloud.com/users/confirmation. Please report bugs, broken links and provide your feedback using the UserSnap tool on the bottom right corner of each page. Stay updated about the web site milestones at https://www.rd-alliance.org/rda-web-platform-upcoming-features-and-functionalities/.

RDA / CODATA Data Systems, Tools, and Services for Crisis Situations WG – Mentoring Pilot

  • Creator
    Discussion
  • #136911

    Stefanie Kethers
    Participant

    Pilot Mentorship of Masters Projects

    • Summary

      • Masters Project  @ UTS iLabs: 3 Groups (9 students) are engaged in DSTSCS related areas for 10 week project. Projects pertain to data science application to disaster datasets

      • Final outputs will be uploaded to DSTSCS page on RDA website by mid-Nov 2023

      • One of the requirements is that all outputs conform to open science principles, and be made FAIR

    • Projects are: 

      • Group A (CrisisUnity): Employs LLM (Llama LLM) to extract structured sentiment score form unstructured data (CrisisLexT26 dataset).

        • Team: Raymond Tsui, William Kent, Marco Sulser

      • Group B (Rescue.ai): Emotional Analysis using GenAI for Disaster Response using LLMs and Chatbot Development.

        • Team: Savinay Singh, Divgun Singh , Bharani Surya Sivasubramanian, Ragul Raj Sasikalaa Pandian Rajah Pandian

      • Group C (PyroVision): Created a Baseline Model for Smoke Detection using Deep Learning

        • Team: Nutan Thapa, Shivani Nandkishor Nipane, Ashutosh Jayant Patil, Ahmed Khursheed

    • Implications of Pilot Mentorship Projects

      • Projects currently focus on pre-processing, exploration, analysis and model development on existing open source structured and unstructured datasets (time constraints). 

      • Two teams focus on unstructured data and employ LLMs to enable fast access to social media posts on crises. Useful for crisis response team to quickly scan and understand the relevance of tweets. One team carried out exploratory analysis of structured data, but aspires to complete smoke detection using images. 

      • Provides nuclei/ artefact case studies for ELSI and FAIR considerations

      • Provides a template for expanding mentorship at other places

    • Program Info: 

      • Course through which mentorship is run: Data Science and Innovation Masters @UTS TD School.

      • Course Coordinator: Dr Shibani Antonette

      • Project Mentors: Drs Md Sarwar Kamal, Jayan Kurian, Flavio Pileggi, Gnana Bharathy

    Group20A20Crisitunity20LLM20in20Crisis20Interim20Presentation.pdf

    Group20B20RescueAI20LLM20in20Crisis20EmotionalAnalysisSupport20Interim20Presentation_final.pdf

    Group20C20Pyrovision20Analytics20Wildfire20Interim20Presentation.pdf

Log in to reply.