Improving Classification of Protein Interaction Articles Using Context Similarity-Based Feature Selection
Table 5
The comparison of common and special selected features on the . denotes the measure of the based on the common features selected by the frequency-based method and the context similarity-based method. The integer in parentheses is the number of the common features; denotes the measure obtained by the context similarity-based method based on the top 1000 features. The number in parentheses is the increments compared with ; denotes the measure obtained by the frequency-based method based on the top 1000 features. The number in parentheses is the increments compared with .
