Webbläsaren som du använder stöds inte av denna webbplats. Alla versioner av Internet Explorer stöds inte längre, av oss eller Microsoft (läs mer här: * https://www.microsoft.com/en-us/microsoft-365/windows/end-of-ie-support).

Var god och använd en modern webbläsare för att ta del av denna webbplats, som t.ex. nyaste versioner av Edge, Chrome, Firefox eller Safari osv.

8

jun

CS MSc Thesis Presentation 8 June 2026

8 juni 2026 13:00 till 14:00 | Exjobbspresentationer

One Computer Science MSc thesis to be presented on 8 June

Monday, 8 June there will be a master thesis presentation in Computer Science at Lund University, Faculty of Engineering.

The presentation will take place in E:2116.

Note to potential opponents: Register as an opponent to the presentation of your choice by sending an email to the examiner for that presentation (firstname [dot] lastname [at] cs [dot] lth [dot] se). Do not forget to specify the presentation you register for! Note that the number of opponents may be limited (often to two), so you might be forced to choose another presentation if you register too late. Registrations are individual, just as the oppositions are! More instructions for opponents are found here on the LTH thesis project page.

13:00-14:00 in E:2116 N.B. No more opponents for this presentation

  • Presenters: Hanna Rosenberg, Love Hedelius
  • Title: Developing an LLM-Based Assistant for Database Interaction
  • Examiner: Alma Orucevic-Alagic
  • Supervisors: Elizabeth Bjarnason (LTH), Douglas Persson (Axis Communications AB), Damir Knezevic (Axis Communications AB)

Analysing data is a valuable way for companies to gain insights, but can be difficult due to large volumes, complex database structures and the need for proficiency in query languages or analytics tools. With their combination of accessibility and capability, LLMs offer a promising means of lowering this barrier. In this thesis, we develop a chatbot assistant to help users utilise a large analytics database. We implemented a solution in several parts, and evaluated them systematically to assess the effectiveness of each. The final product is an LLM-based agent system connected to an MCP server and with access to knowledge base files. Results show that it successfully answers questions about the database and is considered useful, but fails on some tasks that are advanced or involve subtle pitfalls. We conclude that LLM-based assistants have potential for making data analytics more accessible, and that the solution can be valuable in practice.
 

Om evenemanget

Plats:

E:2116

Kontakt:

birger [dot] swahn [at] cs [dot] lth [dot] se