Figure 5

Precision/recall (PR) curves for three models of text (LDA, mixture of unigrams, random) and the task of retrieving a set of aging-related documents (842 CGC items that refer to one or more of the genes listed in Table 1). Precision is the fraction of documents in a list that are relevant (related to aging) whereas recall is the fraction of relevant documents in the list. For a desired level of recall, for example 70%, there is a corresponding precision. The graph shows average precision against average recall. Although each point is a mean of 842 pairs of precision and recall values, the standard error is negligible and so not depicted.