SAS Hacker's Hub

Where your Curiosity Leads to Innovation
BookmarkSubscribeRSS Feed

Council Insights Heidelberg  - Data Analytics on Council Work using LLM

Started ‎08-21-2024 by
Modified ‎11-21-2024 by
Views 6,519
 
PITCH VIDEO
Pitch - Council Insights Heidelberg.mov
Video Player is loading.
Current Time 0:00
Duration 0:00
Loaded: 0%
Stream Type LIVE
Remaining Time 0:00
 
1x
    • Chapters
    • descriptions off, selected
    • captions off, selected
      (view in My Videos)
      JURY VIDEO
      Jury - Council Insights Heidelberg.20fps.mov
      Video Player is loading.
      Current Time 0:00
      Duration 0:00
      Loaded: 0%
      Stream Type LIVE
      Remaining Time 0:00
       
      1x
        • Chapters
        • descriptions off, selected
        • captions off, selected
          (view in My Videos)
           
          TEAM PCITURE
          Hack24_CityCouncilHeidlberg_TeamPicture.jpg

          Team Name 

          Council Insights Heidelberg  

          Track 

          Public Sector 

          Use Case 

          Implementing Lage Language Models LLM for a strong Democracy

          Technology 

           SAS Viya Model Manager, Intelligent Decisioning, Visual Text Analytics, Visual Analytics, Python, LLM, NLP, SAS Viya

          Region 

          Heidelberg, Germany 

          Team lead 

          Benjamin Gärtner @benj_gaertner 

          Team members 

          Stephan Frenzel,   Ahsan Ali, Maria Shcheglova, Christoph Blattgerste (Student), Harsh Patel (Student) , Silvio Martin (Scientific Staff, KIT), Lutz Berger,  Ian Doig, Slava Kuliger, Stephan Verclas, Patrik Bannholzer,  Andreas Koch,   Sebastian Warkentin, Maik Röder
          Sundaresh Sankaran (Mentor), Ulrich Reincke (Mentor)

          Social media handles 

           See the team's  LinkedIn Page specific for the #SASHackathonhttps://www.linkedin.com/company/hack-team-heidelberg/posts/?feedView=all

          Is your team interested in participating in an interview? 

          yes 

          Optional: Expand on your technology expertise 

           

          ============== 

           

          Plan 

          1.  Create automated scripts to regularly retrieve and download relevant documents from the local council website (https://www.gemeinderat.heidelberg.de/info.asp). 
          2. Indexation and vectorization of the scraped data 
          3. Data Analytics on the vectorized data using NLP, LLM and SAS Viya 

          Motivation 

          Our aim is to improve the transparency and accessibility of the work of Heidelberg Municipal Council. We want to enable citizens, the press and other interested parties to find out quickly and easily about the resolutions and activities of the local council. By visualizing and better presenting the work of the committees, we can strengthen citizens' commitment and trust in local government. 

           

          Analysis of strategies 

           

          We would like to compare the strategic objectives of the city of Heidelberg with the activities and resolutions of the municipal council in order to identify specific resolutions that contribute to the implementation of strategic objectives. 

           

          Relevant strategies can be strategies developed by the city of Heidelberg, such as the Heidelberg Biodiversity Strategy, as well as strategies that are valid nationwide, such as the practical guide ‘Climate Protection in Municipalities’ from the German Institute of Urban Studies. 

           

          Possible outputs 

           

          • Analytics on stringency, consistency or legality of council work  
          • Location of resolutions on a map (e.g. by districts, streets, etc.) where possible 
          • Filter function according to districts, topics, committees, time reference (last 6 months or similar) 
          • Search functions (full text, language, locations/regions/environment) 
          • Summary of resolutions in understandable language 
          • Create a chatbot or integration of an intelligent search 
          • Possibly other types of visualizations – Mindmaps, Knowledge Graphs or similar according to keywords or districts (e.g. which topics have been discussed most frequently in the last 6 months) 
          Comments

          Link to the Team Page on LinkedIn with the making of the videos: https://www.linkedin.com/company/hack-team-heidelberg/posts/?feedView=all

          Hack24_CityCouncilHeidlberg_TeamPicture.jpg

          Fantastic work, great team effort - and so nice videos! I am impressed - and as a citizen of Heidelberg proud of those ambitions!!

          Great job ! you did something very useful for citizen as well 

          Congratulations team, great use case, applicable to every state, municipality and country. Also, great approach to solve it using LLMs, AI, analytics and SAS capabilities.

          Very cool project, great use of text analytics, and overall excellent team effort.  Well done!

          Version history
          Last update:
          ‎11-21-2024 09:24 AM
          Updated by:

          sas-innovate-white.png

          Our biggest data and AI event of the year.

          Don’t miss the livestream kicking off May 7. It’s free. It’s easy. And it’s the best seat in the house.

          Join us virtually with our complimentary SAS Innovate Digital Pass. Watch live or on-demand in multiple languages, with translations available to help you get the most out of every session.

           

          Register now!

          Ready to join fellow brilliant minds for the SAS Hackathon?

          Build your skills. Make connections. Enjoy creative freedom. Maybe change the world. Join us at the 2025 SAS Hackathon Sept. 15 – Oct 10. Visit the SAS Hackathon homepage.

          Check it out!