Utilizing Lithuanian Public Sector Big Data to Make the Country Run Smarter
summary
Concrete deployments across Lithuania: call-center assistant using Whisper and wav2vec for live speech recognition and LLM guidance with Llama 70B, transport dashboards using 5G feeds for lane violations and passenger counts, OCR plus clustering to flag agricultural and social-benefit fraud, CNNs that tag damage on 26,000 heritage buildings, IoT energy work that targets 15–20% savings, and an FTIR classifier that cuts lab analysis time by 80% time reduction — most models run on-premises for GDPR compliance and projects take about 3–6 months.
AI-written summary · about
chapters (14)
- 0:00AI in public sector
- 2:36Konas clinic assistant
- 6:09Vilnius transport with 5G
- 7:54Duskin municipality automation
- 9:21Heritage building AI
- 10:46Agricultural fraud detection
- 13:18FTIR for kidney stones
- 15:30Energy efficiency schools
- 17:16Lessons from AI projects
- 20:51AI as assistant in government
- 21:55Future of AI in government
- 25:11AI agents productivity
- 28:49AI adoption challenges
- 30:12Q&A
related talks
32mIsolated Artificial Intelligence: Now or Tomorrow?
22mUrban Artificial Intelligence in Amsterdam
32mHarnessing Large Language Models for Enhanced Customer Experience
35mAI and Advanced Analytics applied at
29mML-Powered IoT: From Warehouse Data to Prod Intelligence in Real Time
52mMachine Learning in IoT Applications
41mSpotlight on AI use cases and innovators around the world
38mUnraveling Multimodality with Large Language Models