Table 1.

Term ranking algorithms and their formulas

p_t: the probability of occurrence of term t in the set of pseudo-relevant documents, q_t: the probability of occurrence of term t in the set of non-relevant documents, c_t: the probability of occurrence of term t in the whole document collection, w_t: the weight to be assigned to term t, EMIM: expected mutual information measure, F4: F4MODIFIED, IDF: inverse document frequency, KLD: Knullback-Leibler divergence, LCA: local context analysis, RSV: Robertson selection value.