A Study of Constructing Topic Person Interaction Network

The development of a topic in a set of topic documents is constituted by a series of person interactions at a specific time and place. Knowing the interactions of the persons mentioned in these documents is helpful for readers to better comprehend the documents. To discover person interactions, we need a detection method that can identify text segments containing information about the interactions. Information extraction algorithms then analyse the segments to extract interaction tuples and construct an interaction network of topic persons. In this dissertation, we define interaction detection as a classification problem. We first recognize person interactions from topic documents by exploring various types of knowledge. We present a feature-based approach called FISER, exploits 19 features covering syntactic, context-dependent, and semantic information in text to detect person interactions. Then, we design the rich interactive tree structure to represent syntactic, context, and semantic information of text, and this structure is incorporated into a tree-based convolution kernel to identify interactive segments. Experiment results based on real world topics demonstrate that effective incorporation of divers features enable our system recognize person interactions efficiently. Moreover, the proposed rich interactive tree structure effectively detects the topic person interaction and that our method outperforms many well-known relation extraction and protein-protein interaction methods.