Text Document Topical Recursive Clustering and Automatic Labeling of a Hierarchy of Document Clusters

Abstract

Abstract. The overwhelming amount of textual documents available nowadays highlights the need for information organization and discovery. Effectively organizing documents into a hierarchy of topics and subtopics makes it easier for users to browse the documents. This paper borrows community mining from social network analysis to generate a hierarchy of topically coherent document clusters. It focuses on giving the document clusters descriptive labels. We propose to use betweenness centrality measure in networks of co-occurring terms to label the clusters. We also incorporate keyphrase extraction and automatic titling in cluster labeling. The results show that the cluster labeling method utilizing KEA to extract keyphrases from the documents generates the best labels overall comparing to other methods and baselines.

Apropiación social del conocimiento investigación, innovación y extensión

“…Así, este escenario está formado por conceptos que describen constructos diferenciables entre sí. Los hallazgos sobre Directed Louvain son afines a los reportados en (Li et al, 2013), donde se agrupan semánticamente los documentos mediante un algoritmo de optimización de modularidad, produciendo grupos de textos con baja superposición y muy similares al listado de referencia manual. Igualmente, en (Ping & Chen, 2018) los autores emplearon un algoritmo Louvain para la construcción de conceptos dado que las unidades resultantes facilitaron la narración visual de artículos científicos.…”

Cluster labelling using chi-square-based keyword ranking and mutual information score: a hybrid approach

“…Comparative study is done on the proposed method to evaluate the relative performance with respect to the Lingo and Suffix tree clustering. Similar to the community mining in the social network, Li et al (2013) in their paper have generated a hierarchy of document clusters which are topically coherent. Cluster labelling is done using the betweenness centrality measure of the term which co-occur in the network.…”

Labeling of Web Search Result Clusters Using Heuristic Search and Frequent Itemset

“…There are many clustering engines available like Kartoo, Carrot2, Vivisimo etc. As search result clustering is being widely researched, research on cluster labeling is also going hand in hand 1,2,3,4 . Labeling of clusters is equally important.…”