Evolutionary spectral clustering by incorporating temporal smoothness

Chen, Yun; Xue, Song; Zhou, Dengyong; Hino, Koji; Tseng, Belle L.

doi:10.1145/1281192.1281212

Cited by 300 publications

(243 citation statements)

References 14 publications

Supporting

Mentioning

240

Contrasting

Order By: Relevance

“…In the evolving environment [9], [10], the data changes at each time step. The number of terms, operations and services that define the web service may increase or decrease or the same words might be used in different domains to define a new web service.…”

Section: B Handling the Evolving Web Services Datamentioning

confidence: 99%

Discovering Heterogeneous Evolving Web Service Communities Using Semi-Supervised Non Negative Matrix Factorization

Andhale¹,

Rege²,

Yu³

2013

IJMLC

View full text Add to dashboard Cite

Abstract-There has been a paradigm shift in the design of software applications. Each offering is designed as a service that can easily integrate with other services. As a result, there are large numbers of web services that are constantly created and consumed. One of the key challenges is to create relevant sub-space domain for web services that can identify the best suitable service to perform a task. Web service discovery process addresses the problem of selecting the best service. In this paper, we propose semi-supervised service community discovery on heterogeneous evolving web services data. The semi-supervised knowledge about the current time step is incorporated into the heterogeneous evolving environment. The evolving changes in the web services are captured and service community is created based on their prior and current knowledge of the web services. Also, the heterogeneous model helps us to create a highly relevant evolving sub space taking into account the operations and services performed by the web services along with the terms used to define the web service simultaneously.

show abstract

Section: B Handling the Evolving Web Services Datamentioning

confidence: 99%

Discovering Heterogeneous Evolving Web Service Communities Using Semi-Supervised Non Negative Matrix Factorization

Andhale¹,

Rege²,

Yu³

2013

IJMLC

View full text Add to dashboard Cite

show abstract

“…This is an NP-hard problem as noted in [6]. The spectral clustering solution involves first relaxing constraint (2), solving the resulting continuous optimization problem, and finally, discretizing the solution to obtain a near global-optimal graph partition [5]. We represent the partition by an n × k partition matrix Y where yij = 1 if vertex i is in cluster j and yij = 0 otherwise.…”

Section: Spectral Clusteringmentioning

confidence: 99%

“…Chakrabarti et al [1] proposed evolutionary extensions of K-means and agglomerative hierarchical clustering. Chi et al [2] proposed two evolutionary frameworks for spectral clustering, one of which we adopt in this paper. [1,2] both make use of a fixed smoothing parameter to control the amount of weight to be applied to past data.…”

Section: Related Workmentioning

confidence: 99%

“…Frameworks for evolutionary clustering have been proposed in previous studies [1,2,3]. We adopt an evolutionary extension for spectral clustering proposed in [2] that takes a convex combination of current and past affinities between data points as the input to the traditional spectral clustering algorithm.…”

Section: Introductionmentioning

confidence: 99%

“…We adopt an evolutionary extension for spectral clustering proposed in [2] that takes a convex combination of current and past affinities between data points as the input to the traditional spectral clustering algorithm. The weights in the convex combination act as a forgetting factor applied to past affinities.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Evolutionary spectral clustering with adaptive forgetting factor

Kliger²,

Hero

2010

2010 IEEE International Conference on Acoustics, Speech and Signal Processing

View full text Add to dashboard Cite

Many practical applications of clustering involve data collected over time. In these applications, evolutionary clustering can be applied to the data to track changes in clusters with time. In this paper, we consider an evolutionary version of spectral clustering that applies a forgetting factor to past affinities between data points and aggregates them with current affinities. We propose to use an adaptive forgetting factor and provide a method to automatically choose this forgetting factor at each time step. We evaluate the performance of the proposed method through experiments on synthetic and real data and find that, with an adaptive forgetting factor, we are able to obtain improved clustering performance compared to a fixed forgetting factor.

show abstract

Extracting evolutionary communities in community question answering

Zhang

Zeng

et al. 2014

Asso for Info Science & Tech

View full text Add to dashboard Cite

With the rapid growth of Web 2.0, community question answering (CQA) has become a prevalent information seeking channel, in which users form interactive communities by posting questions and providing answers. Communities may evolve over time, because of changes in users' interests, activities, and new users joining the network. To better understand user interactions in CQA communities, it is necessary to analyze the community structures and track community evolution over time. Existing work in CQA focuses on question searching or content quality detection, and the important problems of community extraction and evolutionary pattern detection have not been studied. In this article, we propose a probabilistic community model (PCM) to extract overlapping community structures and capture their evolution patterns in CQA. The empirical results show that our algorithm appears to improve the community extraction quality. We show empirically, using the iPhone data set, that interesting community evolution patterns can be discovered, with each evolution pattern reflecting the variation of users' interests over time. Our analysis suggests that individual users could benefit to gain comprehensive information from tracking the transition of products. We also show that the communities provide a decision-making basis for business.

show abstract

Evolutionary spectral clustering by incorporating temporal smoothness

Cited by 300 publications

References 14 publications

Discovering Heterogeneous Evolving Web Service Communities Using Semi-Supervised Non Negative Matrix Factorization

Discovering Heterogeneous Evolving Web Service Communities Using Semi-Supervised Non Negative Matrix Factorization

Evolutionary spectral clustering with adaptive forgetting factor

Extracting evolutionary communities in community question answering

Contact Info

Product

Resources

About