ACONS: A New Algorithm for Clustering Documents

Abstract

In this paper we present a new algorithm for document clustering called Condensed Star (ACONS). This algorithm is a natural evolution of the Star algorithm proposed by Aslam et al., and improved by them and other researchers. In this method, we introduced a new concept of star allowing a different star-shaped form; in this way we retain the strengths of previous algorithms as well as address previous shortcomings. The evaluation experiments on standard document collections show that the proposed algorithm outperforms previously defined methods and obtains a smaller number of clusters. Since the ACONS algorithm is relatively simple to implement and is also efficient, we advocate its use for tasks that require clustering, such as information organization, browsing, topic tracking, and new topic detection.

An incremental nested partition method for data clustering

“…The complete-link and single-link were overcame by the star algorithms and ACONS algorithm, as we expect from the experiments carried out by Aslam et al [34] and results showed in [36]. On the other hand, the results of our algorithm are better than the results of all algorithms in the case of AFP database whereas that in TDT database are comparable with the results obtained by the star algorithms and better than all remaining algorithms.…”

“…Several algorithms such as star algorithm (S) [34], the extended star algorithm (ES) [35], ACONS algorithm (AC) [36], algorithms based on compact sets [27], among other, are in the state of the art. Also, classical algorithms such as k-means, single and complete-link have been used in this task.…”

A New Graph-Based Algorithm for Clustering Documents

“…The ACONS algorithm [6], as the Generalized Star algorithm, represents the document collection through its associated thresholded similarity graph G β and was proposed to reduce the drawbacks presented in Star and Extended algorithms.…”

“…Another algorithm that improves Star and Extended Star algorithms is ACONS [6]. The experimentation on standard documents collections showed that ACONS outperforms previous Star-like algorithms regarding to external evaluation measures.…”

Categorizing Overlapping Regions in Clustering Analysis Using Three-Way Decisions

“…In recent years, clustering has been widely used in many areas such as business intelligence, image recognition and biological information [1]. However, there are some applications like social network analysis [2], wireless sensor networks [3], document clustering [4], and so on, where it is common that objects may belong to more than one cluster. In those areas, discovering overlapping regions is important and meaningful.…”