Maximal Quasi-Bicliques with Balanced Noise Tolerance: Concepts and Co-clustering Applications

Li, Jinyan; Sim, Kelvin; Liu, Guimei; Wong, Limsoon

doi:10.1137/1.9781611972788.7

Cited by 33 publications

(21 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Semiclique mining of the co-occurrence graph, as we have described, is just one way of inducing candidate sets and partial sets. 11 One can envision methods that, similar to ngram extraction, require a few passes over the corpus, and mine for statistically significant sets of co-occurring terms.…”

Section: Discussion On Semiclique Generationmentioning

confidence: 99%

“…Here, it is very important to allow for overlapping groupings to support multiple senses and events, unlike much (graph) clustering work. On the other hand, unlike some applications [11], we do not require enumerating all the maximal semicliques. As we mentioned, the co-occurrence graph of terms is best viewed as hierarchical, reflecting term frequencies, not flat.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Discovery of numerous specific topics via term co-occurrence analysis

Madani

2010

Proceedings of the 19th ACM International Conference on Information and Knowledge Management

View full text Add to dashboard Cite

We describe techniques for the construction of term co-occurrence graphs and explore an application of such graphs to the discovery of tens of thousands of fine-grained, that is specific rather than broad, topics. A topic corresponds to a small dense subgraph in our work. We discover topics by randomized local searches (constrained random walks) initiated at each term (node) in the graph. The mined topics are highly interpretable, and reveal the different meanings of a term in the corpus. We explore document tagging via the induced topics, and demonstrate the information-theoretic utility of the topics when they are used as features in supervised learning. Such features lead to consistent improvements in text classification accuracy over the standard bag-of-words tfidf representation, using SVM classification, even at high training proportions when it is difficult to improve over the tfidf representation. We investigate the effect of various options and parameters, including window size during graph construction and variants of a tagging strategy, on the accuracy of classification. We explain how a layered pyramidal view of the term distribution helps in understanding the algorithms and in visualizing and interpreting the topics.

show abstract

Section: Discussion On Semiclique Generationmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Discovery of numerous specific topics via term co-occurrence analysis

Madani

2010

Proceedings of the 19th ACM International Conference on Information and Knowledge Management

View full text Add to dashboard Cite

show abstract

“…Thus, these methods cannot be used to find concept graphs since, e.g., the members of a concept graph need not be connected at all. Methods [11] have been proposed for quasi-biclique detection, which explicitly distinguish between members and aspects. The existing methods, however, are not aware of the concept of cue validity and are not able to determine representative vertices for each concept.…”

Section: Related Workmentioning

confidence: 99%

Fault-Tolerant Concept Detection in Information Networks

Kötter

Günnemann

Berthold

et al. 2014

Advances in Knowledge Discovery and Data Mining

View full text Add to dashboard Cite

Abstract. Given information about medical drugs and their properties, how can we automatically discover that Aspirin has blood-thinning properties, and thus prevents heart attacks? Expressed in more general terms, if we have a large information network that integrates data from heterogeneous data sources, how can we extract semantic information that provides a better understanding of the integrated data and also helps us to identify missing links? We propose to extract concepts that describe groups of objects and their common properties from the integrated data. The discovered concepts provide semantic information as well as an abstract view on the integrated data and thus improve the understanding of complex systems. Our proposed method has the following desirable properties: (a) it is parameter-free and therefore requires no user-defined parameters (b) it is fault-tolerant, allowing for the detection of missing links and (c) it is scalable, being linear on the input size. We demonstrate the effectiveness and scalability of the proposed method on real, publicly available graphs.

show abstract

“…In [22], we introduce an alternate version of maximal quasibiclique whose error tolerance is percentage based. As this alternate version does not have anti-monotone property, there is no efficient algorithm to mine it.…”

Section: A Graphmentioning

confidence: 99%

Mining maximal quasi‐bicliques: Novel algorithm and applications in the stock market and protein networks

Sim¹,

Li²,

Gopalkrishnan³

et al. 2009

Statistical Analysis

Self Cite

View full text Add to dashboard Cite

Abstract-Several real world applications require mining of bicliques, as they represent correlated pairs of data clusters. However, the mining quality is adversely affected by missing and noisy data. Moreover, some applications only require strong interactions between data members of the pairs, but bicliques are pairs that display complete interactions. We address these two limitations by proposing maximal quasi-bicliques. Maximal quasibicliques tolerate erroneous and missing data, and also relax the interactions between the data members of their pairs. Besides, maximal quasi-bicliques do not suffer from skewed distribution of missing edges that prior quasi-bicliques have. We develop an algorithm MQBminer, which mines the complete set of maximal quasi-bicliques from either bipartite or non-bipartite graphs. We demonstrate the versatility and effectiveness of maximal quasibicliques to discover highly correlated pairs of data in two diverse real world datasets. Firstly, we propose to solve a novel financial stocks analysis problem by using maximal quasi-bicliques to cocluster stocks and financial ratios. Results show that the stocks in our co-clusters usually have significant correlations in their price performance. Secondly, we use maximal quasi-bicliques on a mining protein network problem and we show that pairs of protein groups mined by maximal quasi-bicliques are more significant than those mined by maximal bicliques.

show abstract

Maximal Quasi-Bicliques with Balanced Noise Tolerance: Concepts and Co-clustering Applications

Cited by 33 publications

References 29 publications

Discovery of numerous specific topics via term co-occurrence analysis

Discovery of numerous specific topics via term co-occurrence analysis

Fault-Tolerant Concept Detection in Information Networks

Mining maximal quasi‐bicliques: Novel algorithm and applications in the stock market and protein networks

Contact Info

Product

Resources

About