Range-Based Clustering Supporting Similarity Search in Big Data

Phan, Trong Nhan; Jäger, Markus; Nadschläger, Stefan; Küng, Josef

doi:10.1109/dexa.2015.41

Cited by 3 publications

(7 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The author is grateful to John E. Baumler for inspiring discussions that initiated the idea of range‐clustering, and to Marjan Baghaie for her contributions to the writing of this paper and pointing out reference . This is to thank the referee for detailed and perceptive comments as well as providing reference . The referee's numerous suggestions contributed significantly to improving the presentation and are much appreciated.…”

Section: Acknowledgmentsmentioning

confidence: 91%

Algorithms and complexity of range clustering

Hochbaum

2018

Networks

View full text Add to dashboard Cite

We introduce a novel criterion in clustering that seeks clusters with limited range of values associated with each cluster's elements. In clustering or classification the objective is to partition a set of objects into subsets, called clusters or classes, consisting of similar objects so that different clusters are as dissimilar as possible. We propose a number of objective functions that employ the range of the clusters as part of the objective function. Several of the proposed objectives mimic objectives based on sums of similarities. These objective functions are motivated by image segmentation problems, where the diameter, or range of values associated with objects in each cluster, should be small. It is demonstrated that range-based problems are in general easier, in terms of their complexity, than the analogous similarity-sum problems. Several of the problems we present could therefore be viable alternatives to existing clustering problems which are NP-hard, offering the advantage of efficient algorithms.

show abstract

Section: Acknowledgmentsmentioning

confidence: 91%

Algorithms and complexity of range clustering

Hochbaum

2018

Networks

View full text Add to dashboard Cite

show abstract

“…In this section, we inherit and develop the basic idea of range-based clustering from the work [32] to perform a spiral clustering scheme with its improved version and variants. When given a query object ‫ܦ‬ , the spiral clustering process also maps ‫ܦ‬ into a cluster ‫ܥ‬ , where ‫ܦ‬ are verified against the objects in the cluster ‫ܥ‬ for its similarity.…”

Section: Spiral Clustering Schemementioning

confidence: 99%

“…There are lots of metrics or distance functions to estimate the similarity scores [24]. In this paper, we use Jaccard coefficient [2,3,6,29,[31][32][33], a well-known metric for fast set-based similarity.…”

Section: A Similarity Searchmentioning

confidence: 99%

“…Lots of literatures have been employing MapReduce for many of large computational problems [2,8,18,21,25,27,28,[30][31][32][33]. A MapReduce paradigm, however, suffers high I/O costs due to the fact that it has to write both intermediate and final key-value pairs into its distributed file system.…”

Section: Introductionmentioning

confidence: 99%

“…x We develop the spiral clustering scheme from the basic idea of the work [32] and introduce its improved version for similarity search. In addition, our proofs guarantee the correctness of our variants.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

eHSim: An Efficient Hybrid Similarity Search with MapReduce

Phan

Küng

Dang

2016

2016 IEEE 30th International Conference on Advanced Information Networking and Applications (AINA)

Self Cite

View full text Add to dashboard Cite

In this paper, we study the problems of scalability and performance for similarity search by proposing eHSim, an efficient hybrid similarity search with MapReduce. More specifically, we introduce clustering schemes that partition objects into different groups by their length. Additionally, we equip our proposed schemes with pruning strategies that quickly discard irrelevant objects before truly computing their similarity. Moreover, we design a hybrid MapReduce architecture that deals with challenges from big data. Furthermore, we implement our proposed methods with MapReduce and make them compatible with the hybrid MapReduce architecture. Last but not least, we evaluate the proposed methods with real datasets. Empirical experiments show that our approach is considerably more efficient than state-of-the-arts in terms of query processing, batch processing, and data storage.

show abstract

Development of KNN-EIM based hybrid model for brain MRI density range determination

Rani

Gladis

2022

ijhs

View full text Add to dashboard Cite

Data mining techniques are used in the medical data analysis to predict the hidden knowledge from the existing data. This work attempted to apply the data mining techniques to analysis the medical image focus to brain tumor evaluation. The affected tumor region range values. The work focused to evaluate the performance of clustering algorithms for medical image analysis via predicting the tumor affected area. It is achieved through analysing Brain MRI to identify the high-density neuron using the integrated approach of K-Means and Equal Interval clustering. The tumour affected area and its range values are identified using a clustering algorithm such as Equal Interval Method and K-Means approach. To enhance the accuracy of the affected region, the common affected high-density pixels are identified using a hybrid model through the Integration of K-Means and Equal Interval Method. It is evaluated according to time complexity and accuracy.

show abstract

Range-Based Clustering Supporting Similarity Search in Big Data

Cited by 3 publications

References 9 publications

Algorithms and complexity of range clustering

Algorithms and complexity of range clustering

eHSim: An Efficient Hybrid Similarity Search with MapReduce

Development of KNN-EIM based hybrid model for brain MRI density range determination

Contact Info

Product

Resources

About