Self-splitting competitive learning: a new on-line clustering paradigm

Zhang, Yajun; Liu, Zhiqiang

doi:10.1109/72.991422

Cited by 84 publications

(3 citation statements)

References 42 publications

(56 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Many clustering models have been proposed in the literature such as partitioning [4], hierarchical [5] and spectralbased models [6], which have applications in many fields including image segmentation [7], [8], social networks analysis and community discovery [9], [10], recommender systems [11]- [13] and so on [14], [15]. Clustering in the artificial neural networks (ANNs) literature is usually based on a competitive learning (CL) paradigm [16]- [18] where codebook weight vectors (prototypes) compete in order to elect the best matching unit (BMU), i.e., a neuron unit whose weight vector has the minimum distance to an input vector. Afterward, the selected prototype is updated to get closer to the input vector.…”

Section: Introductionmentioning

confidence: 99%

“…To address the limitations related to (1) sensitivity to cluster centers initialization and selection of the number of clusters k (i.e., requiring multiple restarts) and ( 2) optimization of non-convex clustering quality measures, we represent the problem of clustering multiple transaction databases as a quasi-convex optimization problem solvable without specifying the number of clusters beforehand. In contrast to competitive learning paradigm [16]- [18], we have adopted a gradient-based learning approach [27] with back-propagation [28] to minimize a clustering quasi-convex loss function L(θ) which guarantees convergence to the global minimum. We also discover the number of clusters (denoted by f θ (D)) in the input space by incorporating f θ (D) into our objective.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Gradient-Based Clustering for Multi-Database Mining

2021

View full text Add to dashboard Cite

Multinational corporations have multiple databases distributed throughout their branches, which store millions of transactions per day. For business applications, identifying disjoint clusters of similar and relevant databases contributes to learning the common buying patterns among customers and also increases the profits by targeting potential clients in the future. This process is called clustering, which is an important unsupervised technique for big data mining. In this article, we present an effective approach to search for the optimal clustering of multiple transaction databases in a weighted undirected similarity graph. To assess the clustering quality, we use dual gradient descent to minimize a constrained quasi-convex loss function whose parameters will determine the edges needed to form the optimal database clusters in the graph. Therefore, finding the global minimum is guaranteed in a finite and short time compared with the existing non-convex objectives where all possible candidate clusterings are generated to find the ideal clustering. Moreover, our algorithm does not require specifying the number of clusters a priori and uses a disjoint-set forest data structure to maintain and keep track of the clusters as they are updated. Through a series of experiments on public data samples and precomputed similarity matrices, we show that our algorithm is more accurate and faster in practice than the existing clustering algorithms for multi-database mining.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A Gradient-Based Clustering for Multi-Database Mining

2021

View full text Add to dashboard Cite

show abstract

“…To tackle the latter problems, it is suggested to keep the transactional data stored locally and only forward the local patterns mined at each branch database to a central site where they will be clustered into disjoint cohesive pattern-base groups for knowledge discovery. In fact, analyzing the local patterns present in each individual cluster of the multiple databases (MDB) enhances the quality of aggregating novel relevant patterns, and also facilitates the parallel maintenance of the obtained database clusters.Various clustering algorithms and models have been introduced in the literature, namely spectral-based models [ 2 ], hierarchical [ 3 ], partitioning [ 4 ], competitive learning-based models [ 5 , 6 , 7 ] and artificial neural networks (ANNs) based clustering [ 8 , 9 , 10 ]. Additionally, clustering could be applied in many domains [ 11 , 12 ] including community discovery in social networks [ 13 , 14 ], image segmentation [ 15 , 16 ] and recommendation systems [ 17 , 18 , 19 ].…”

Section: Introductionmentioning

confidence: 99%

An Improved Similarity-Based Clustering Algorithm for Multi-Database Mining

Miloudi

Wang

Ding

2021

Entropy

View full text Add to dashboard Cite

Clustering algorithms for multi-database mining (MDM) rely on computing (n2−n)/2 pairwise similarities between n multiple databases to generate and evaluate m∈[1,(n2−n)/2] candidate clusterings in order to select the ideal partitioning that optimizes a predefined goodness measure. However, when these pairwise similarities are distributed around the mean value, the clustering algorithm becomes indecisive when choosing what database pairs are considered eligible to be grouped together. Consequently, a trivial result is produced by putting all the n databases in one cluster or by returning n singleton clusters. To tackle the latter problem, we propose a learning algorithm to reduce the fuzziness of the similarity matrix by minimizing a weighted binary entropy loss function via gradient descent and back-propagation. As a result, the learned model will improve the certainty of the clustering algorithm by correctly identifying the optimal database clusters. Additionally, in contrast to gradient-based clustering algorithms, which are sensitive to the choice of the learning rate and require more iterations to converge, we propose a learning-rate-free algorithm to assess the candidate clusterings generated on the fly in fewer upper-bounded iterations. To achieve our goal, we use coordinate descent (CD) and back-propagation to search for the optimal clustering of the n multiple database in a way that minimizes a convex clustering quality measure L(θ) in less than (n2−n)/2 iterations. By using a max-heap data structure within our CD algorithm, we optimally choose the largest weight variable θp,q(i) at each iteration i such that taking the partial derivative of L(θ) with respect to θp,q(i) allows us to attain the next steepest descent minimizing L(θ) without using a learning rate. Through a series of experiments on multiple database samples, we show that our algorithm outperforms the existing clustering algorithms for MDM.

show abstract

References

2008

Clustering

View full text Add to dashboard Cite

Self-splitting competitive learning: a new on-line clustering paradigm

Cited by 84 publications

References 42 publications

A Gradient-Based Clustering for Multi-Database Mining

A Gradient-Based Clustering for Multi-Database Mining

An Improved Similarity-Based Clustering Algorithm for Multi-Database Mining

References

Contact Info

Product

Resources

About