A Semantic Aspect-Based Vector Space Model to Identify the Event Evolution Relationship within Topics

Abstract

Understanding how the topic evolves is an important and challenging task. A topic usually consists of multiple related events, and the accurate identification of event evolution relationship plays an important role in topic evolution analysis. Existing research has used the traditional vector space model to represent the event, which cannot be used to accurately compute the semantic similarity between events. This has led to poor performance in identifying event evolution relationship. This paper suggests constructing a semantic aspect-based vector space model to represent the event: First, use hierarchical Dirichlet process to mine the semantic aspects. Then, construct a semantic aspect-based vector space model according to these aspects. Finally, represent each event as a point and measure the semantic relatedness between events in the space. According to our evaluation experiments, the performance of our proposed technique is promising and significantly outperforms the baseline methods.

Challenges of event detection from social media streams

“…Supervised classification methods have obtained good results, but they are time-consuming and require a large amount of data to be trained, and required a lot of human effort [45]. On the other hand, despite the popularity of clustering techniques for ED and the fact that they do not require labeled data, it is still a challenging task to build an automated unsupervised method which can deal with high dimensionality of news stream data without human effort and cost [46]. In practice, various ED approaches are available in the literature which affects the quality of results [12].…”

“…In practice, not all UGC contains useful information; in fact, a substantial amount contains meaningless contents that are not relevant to real-world events [46]. Consequently, this generates a lot of noise, and eventually, affects negatively on ED accuracy and performance.…”