Statistical Modeling of Relevance Judgments for Probabilistic Retrieval of American Case Law

Studies the use of Logistic regression methods to meld relevance clues into a single probability of relevance estimate. The model is applied to the domain of American case law.