Figure 5From: Uncovering protein interaction in abstracts and text using a novel linear model and word proximity networksVTT decision surface for a training partition. Decision boundary for VTT on the space of P(a)/N(a) and np(a), for one of the training k-fold partitions. Red and blue dots represent negative and positive abstracts. Dotted line represents surface that optimizes training data alone. VTT, variable trigonometric threshold.Back to article page