Student research opportunities
Predictive models for Information Retrieval
Project Code: CECS_922
This project is available at the following levels:
Honours, Masters, PhD
Keywords:
information retrieval, text analysis
Supervisor:
Dr Wray BuntineOutline:
In the world of Information Retrieval, BM25, a variant of TF-IDF is king. "Language models for information retrieval" have been developed as an alternative but is an incremental improvement at best, primarily because the models are mostly unigram. Richer predictive models would look at word interactions and could offer improvements. An initial study would abandon computational considerations and test out models for retrieval performance ignoring cost.
Goals of this project
To explore richer predictive models of text in the language modelling style and evaluate their performance on some standard collections.
Requirements/Prerequisites
COMP4650 and COMP4670 or graduate equivalents.
Links
The Probabilistic Relevance Framework: BM25 and BeyondOkapi BM25
Language models for IR
