Skip to main content

Table 1 Supervised query expansion for biomedical information retrieval

From: A supervised term ranking model for diversity enhanced biomedical information retrieval

Algorithm 1 supervised query expansion pipeline
Training the SQE model M
1: For each training query q, select k candidate terms via PRF
2: Label each term based on the diversity-oriented strategy
3: Represent each term as a feature vector using different term features
4: Train term ranking model M using the modified loss function
Testing the model M in query expansion retrieval
1: For each testing query q, select k candidate terms via PRF
2: Represent each term as a feature vector using the term features
3: Apply M to obtain the top m terms for query expansion