Textual Knowledge Representation through the Semantic-Based Graph Structure in Clustering Applications

To represent the textual knowledge more expressively, a kind of semantic-based graph structure is proposed for this issue and thereafter applied to clustering problems. Such graph structure for textual representation consists of nodes and directed edges, which stand for the feature terms derived from the texts and the semantic relationships between them, respectively. Moreover, the weight is assigned to each edge so that the strength of relationship between two terms can be measured. For this weighted directed graph structure, a novel graph similarity algorithm is developed by extracting the maximum common subgraph between two concerned graphs, which can therefore be used to measure the distance between two graph structures, i.e. two texts, and finally be used to sort the texts into different clusters. Some experiments have been done through the proposed semantic graph structure in clustering applications and the results have proved the high performance of our textual knowledge representation model.