Figure 3
From: Protein interaction sentence detection using multiple semantic kernels
The vector space representation in a word-based semantic space. An example of the vector space representation in a word-based semantic space, where the context consists of all the words co-occurring with the target within the sentence. The columns represent the basis words that make up the contexts, while the rows are the target words. In this model the co-occurrence matrix is symmetric. The stop words (a, the, on) are ignored.