AI based decision support for LSS
| Reference number | |
| Coordinator | Mölndals kommun - Biståndsenheten LSS |
| Funding from Vinnova | SEK 375 000 |
| Project duration | May 2020 - January 2022 |
| Status | Completed |
| Venture | AI - Competence, ability and application |
| Call | Start your AI-journey! For public organizations |
Important results from the project
The overall goal of the project was to develop and implement machine learning methods based on natural language processing (NLP) that can identify and summarize historical cases about the Act on support and Service for Certain Disabled Persons (LSS) at the Health and Care Administration in Mölndal city. The project analyzed, developed, and implemented multiple algorithms that could search and identify cases in a database of historical LSS cases. A large part of the project was spent on data security issues and how to ensure that sensitive data is not exposed.
Expected long term effects
The project was a feasibility study where the goal was for the city of Mölndal to start its journey towards more AI-based decision support tools. In the project, the FCC showed that it is possible to use more data-driven methods to simplify the work for people at Mölndal stad. They are now well prepare when it comes to planning and procuring future decision support tools.
Approach and implementation
The project was carried out by the Mölndal stad and Fraunhofer-Chalmers Centre (FCC). Mölndal stad acted as problem owners and requirement setter for the project and FCC acted as experts in machine learning and AI. One large part of the project was to transfer data regarding historical LSS cases to the FCC for analysis and development. Due to the pandemic, this caused some difficulties due to sensitive personal data in the raw data.