Knowledge Map Construction Using Text Mining and Artificial Bee Colony Algorithm
Tsai Chieh-Yuan, Ji Wei-Zhong · 2018
With the rapid development of the information technology, information overload becomes a serious problem during the information acquisition process. To relieve this difficulty, knowledge map is a systematic approach to reveal the underlying relationships between abundant knowledge sources. However, few studies focused on how to optimize the coordinates of knowledge items in the map to help users easily understand complicated relatedness among knowledge topics. To bridge this gap, this paper proposes a novel knowledge map construction approach using text mining and artificial bee colony (ABC) algorithm. First, the textural documents related to a certain domain are represented as a term vector in m-dimensional space with the term frequency-inverse document frequency (TF-IDF) analysis. Second, hierarchical clustering is applied to identify important topics. Third, high-dimensional relationships among knowledge items are transformed into a 2-dimensional space optimized by the ABC algorithm. A set of experiments shows that setting appropriate number of clusters is important for visual perception. In addition, a practical example in topic trend analysis using the proposed approach is demonstrated at the end of this paper.