- Published a large-scale, text-based search engine in Java leveraging Apache Lucene for 33,258 astronomical publications
- Utilized machine learning functionality such as ranking diversification and learning-to-rank with an SVM classifier and integrated multiple information-retrieval models to analyze data for classification and regression analyses
- Constructed a predictive model for document relevance based on Cornell SVM-Rank with 18 features which improved ranking precision by 20%
- unranked/ranked booleans
- BM25
- Indri
Reference: https://boston.lti.cs.cmu.edu/classes/11-642/ 2019 version