Minimum spanning tree based split-and-merge: A hierarchical clustering method

Zhong, Chongquan; Miao, Duoqian; Fränti, Pasi

doi:10.1016/j.ins.2011.04.013

Cited by 97 publications

(46 citation statements)

References 44 publications

(50 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Alternative clustering algorithms, such as the divisive method of [28] may allow more diversely shaped clusters of documents to be found more easily, while different methods of merging or splitting clusters during the construction of the hierarchy can significantly affect the final hierarchy [29]. As our focus was on the associative network, not the effect of different clustering algorithms, examining this effect is left as future work.…”

Section: Discussionmentioning

confidence: 99%

Hierarchical Document Categorization using Associative Networks

Bloom¹,

Theune²,

Jong³

2013

Artificial Intelligence and Applications / 794: Modelling, Identification and Control / 795: Parallel and Distributed Computing

View full text Add to dashboard Cite

Associative networks are a connectionist language model with the ability to handle dynamic data. We used two associative networks to categorize random sets of related Wikipedia articles with only their raw text as input. We then compared the resulting categorization to a gold standard: the manual categorization by Wikipedia authors and used a neural network as a baseline. We also determined a human rating by having a group of judges rank the four categorization methods by correctness and by usefulness with regards to finding information. Based on these tests, we determined that associative networks produce results that are clearly better than the neural network baseline, coming close to the gold standard in terms of usefulness and correctness. Furthermore, automated testing suggests these results continue to hold for large datasets.

show abstract

Section: Discussionmentioning

confidence: 99%

Hierarchical Document Categorization using Associative Networks

Bloom¹,

Theune²,

Jong³

2013

Artificial Intelligence and Applications / 794: Modelling, Identification and Control / 795: Parallel and Distributed Computing

View full text Add to dashboard Cite

show abstract

“…First, the given dataset X is divided into k partitions using K-means algorithm, where k is set to √ n [7]. Then, the local neighborhood graph G LN is obtained by considering the points in the neighboring partitions.…”

Section: Mst-based Clustering On K-means Graphmentioning

confidence: 99%

“…In order to demonstrate that the local neighborhood graphs KMLNG and BMLNG will not miss any information that is significant for clustering, we test the performance of the clustering results on the approximate MST obtanied from KMLNG and BMLNG using Zahn's clustering algorithm [3]. The results are validated using the external quality indices such as Rand, FM, Jaccard and Adjusted Rand [7]. Table 3 shows the quality indices of clustering on various datasets.…”

Section: Algorithm 2 Bi-means Algorithmmentioning

confidence: 99%

See 1 more Smart Citation

Fast Minimum Spanning Tree Based Clustering Algorithms on Local Neighborhood Graph

Jothi

Mohanty

Ojha

2015

Graph-Based Representations in Pattern Recognition

View full text Add to dashboard Cite

Abstract. Minimum spanning tree (MST) based clustering algorithms have been employed successfully to detect clusters of heterogeneous nature. Given a dataset of n random points, most of the MST-based clustering algorithms first generate a complete graph G of the dataset and then construct MST from G. The first step of the algorithm is the major bottleneck which takes O(n 2 ) time. This paper proposes two algorithms namely MST-based clustering on K-means Graph and MST-based clustering on Bi-means Graph for reducing the computational overhead. The proposed algorithms make use of a centroid based nearest neighbor rule to generate a partition-based Local Neighborhood Graph (LN G). We prove that both the size and the computational time to construct the graph (LN G) is O(n 3/2 ), which is a O( √ n) factor improvement over the traditional algorithms. The approximate MST is constructed from LN G in O(n 3/2 lg n) time, which is asymptotically faster than O(n 2 ). The advantage of the proposed algorithms is that they do not require any parameter setting which is a major issue in many of the nearest neighbor finding algorithms. Experimental results demonstrate that the computational time has been reduced significantly by maintaining the quality of the clusters obtained from the MST.

show abstract

“…This is the case of agglomerative hierarchical clustering methods [22,25,34]. These algorithms work in a bottom-up fashion, starting out by considering each instance as a cluster, and iteratively combining the two most similar clusters in terms of a similarity function.…”

Section: Introductionmentioning

confidence: 99%

Learning data structure from classes: A case study applied to population genetics

Coz

Díez

Bahamonde

2012

Information Sciences

View full text Add to dashboard Cite

a b s t r a c tIn most cases, the main goal of machine learning and data mining applications is to obtain good classifiers. However, final users, for instance researchers in other fields, sometimes prefer to infer new knowledge about their domain that may be useful to confirm or reject their hypotheses. This paper presents a learning method that works along these lines, in addition to reporting three interesting applications in the field of population genetics in which the aim is to discover relationships between species or breeds according to their genotypes. The proposed method has two steps: first it builds a hierarchical clustering of the set of classes and then a hierarchical classifier is learned. Both models can be analyzed by experts to extract useful information about their domain. In addition, we propose a new method for learning the hierarchical classifier. By means of a voting scheme employing pairwise binary models constrained by the hierarchical structure, the proposed classifier is computationally more efficient than previous approaches while improving on their performance.

show abstract

Minimum spanning tree based split-and-merge: A hierarchical clustering method

Cited by 97 publications

References 44 publications

Hierarchical Document Categorization using Associative Networks

Hierarchical Document Categorization using Associative Networks

Fast Minimum Spanning Tree Based Clustering Algorithms on Local Neighborhood Graph

Learning data structure from classes: A case study applied to population genetics

Contact Info

Product

Resources

About