Abstract
Optimising Information Retrieval (IR) models is vital for improved performance. While various strategies, including parameter estimation and tuning, can contribute to these improvements, our focus lies in utilising a statistical modelling approach, particularly through residual analysis. This paper investigates the application of the Logistic Regression approach within Generalised Linear Models (GLMs) to enhance BM25 IR models. Notably, the utilisation of residual analysis and GLMs for IR model enhancement remains largely unexplored in the domain of Information Retrieval research.
| Original language | English |
|---|---|
| Title of host publication | Proceedings of the 38th International Workshop on Statistical Modelling, Durham, July 14-19, 2024 |
| Editors | Jochen Einbeck, Reza Drikvandi, Georgios Karagiannis, Konstantinos Perrakis, Qing Zhang |
| Place of Publication | U.K. |
| Publisher | University of Durham |
| Pages | 265-270 |
| Number of pages | 6 |
| ISBN (Print) | 9780907552444 |
| Publication status | Published - 2024 |
| Event | International Workshop on Statistical Modelling - Durham, United Kingdom Duration: 14 Jul 2024 → 19 Jul 2024 Conference number: 38th |
Conference
| Conference | International Workshop on Statistical Modelling |
|---|---|
| Country/Territory | United Kingdom |
| City | Durham |
| Period | 14/07/24 → 19/07/24 |