👷 Introduction to Information Retrieval

Information retrieval by question answering by large language models and Clustering 26. 4. 2023

Lecture

Question Answering and Beyond
Martin Fajčík's lecture slides

Readings

Chyba: Odkazovaný objekt neexistuje nebo nemáte právo jej číst.
https://is.muni.cz/el/fi/jaro2023/PV211/um/readings/16flat.pdf
Chyba: Odkazovaný objekt neexistuje nebo nemáte právo jej číst.
https://is.muni.cz/el/fi/jaro2023/PV211/um/readings/17hier.pdf

Question Answering Decomposed
Colab notebook showing how to build an extractive Question Answering system based on word embeddings (like Word2Vec) and Recurrent neural network.

Seminar

Clustering
Exercise solution for seminars in the eleventh week
Clustering
Google Colaboratory code for seminars in the eleventh week

Chyba: Odkazovaný objekt neexistuje nebo nemáte právo jej číst.
https://is.muni.cz/el/fi/jaro2023/PV211/um/whiteboards/

Second term project

Second term project assignment
CQADupStack Collection and the ARQMath Collection

Below, you can find the homework vaults for submitting the second term project:

Second term project (CQADupStack Collection)
Homework vault for the second term project (a ranked supervised retrieval system for the CQADupStack Collection).
Alternative second term project (ARQMath Collection)
Homework vault for the alternative second term project (a ranked supervised retrieval system for the ARQMath Collection).

As announced in the course introductory lecture, to verify the originality of your approach and to help the reviewers to assess the quality of your solution, you were asked to accompany your Python notebook with a short Technical Report (max 400 words) for up to 6 points. You may (optionally) use the Technical Report Template. Due to Michal's misunderstanding and his late and ambiguous information, we have allowed resubmitting the Technical report to the new homework vaults for the second term project of a) ARQMath and b) CQADupStack.