A framework for simultaneous co-clustering and learning from complex data

Deodhar, Meghana; Ghosh, Joydeep

doi:10.1145/1281192.1281222

Cited by 58 publications

(85 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Deodhar and Ghosh (2007) 1 stated that researchers most often do partitioning a priori based on domain knowledge or a separate segmentation routine.…”

Section: Introductionmentioning

confidence: 99%

Applying CHAID for logistic regression diagnostics and classification accuracy improvement

Antipov

Pokryshevskaya²

2010

J Target Meas Anal Mark

View full text Add to dashboard Cite

In this study a CHAID-based approach to detecting classification accuracy heterogeneity across segments of observations is proposed. This helps to solve some important problems, facing a model-builder:1. How to automatically detect segments in which the model significantly underperforms?2. How to incorporate the knowledge about classification accuracy heterogeneity across segments to partition observations in order to achieve better predictive accuracy?The approach was applied to churn data from the UCI Repository of Machine Learning Databases. By splitting the dataset into 4 parts, which are based on the decision tree, and building a separate logistic regression scoring model for each segment we increased the accuracy by more than 7 percentage points on the test sample. Significant increase in recall and precision was also observed. It was shown that different segments may have absolutely different churn 2 predictors. Therefore such a partitioning gives a better insight into factors influencing customer behavior.

show abstract

“…Deodhar and Ghosh (2007) 1 stated that researchers most often do partitioning a priori based on domain knowledge or a separate segmentation routine.…”

Section: Introductionmentioning

confidence: 99%

Applying CHAID for logistic regression diagnostics and classification accuracy improvement

Antipov

Pokryshevskaya²

2010

J Target Meas Anal Mark

View full text Add to dashboard Cite

show abstract

“…Two of these approaches are SCOAL (Simultaneous Co-clustering and Learning) [8] and PDLF (Predictive Discrete Latent Factor Modeling) [4]. Both approaches partition the users and items into a grid of blocks (co-clusters) of related users and items, while simultaneously learning a predictive model on each co-cluster.…”

Section: Related Workmentioning

confidence: 99%

“…The organic emergence of these predictive models is coupled with the formation of the co-clusters that define the domain of the models. Such coupling of the models and co-clusters improves both the interpretability and the accuracy when modeling predictively heterogeneous dyadic datasets, as this mechanism can effectively exploit both local neighborhood patterns as well as the globally available attributes [8,4].…”

Section: Related Workmentioning

confidence: 99%

Learning multiple models for exploiting predictive heterogeneity in recommender systems

Jones

Ghosh

Sharma

2011

Proceedings of the 2nd International Workshop on Information Heterogeneity and Fusion in Recommender Systems

Self Cite

View full text Add to dashboard Cite

Collaborative filtering approaches exploit information about historical affinities or ratings to predict unknown affinities between sets of "users" and "items" and make recommendations. However a model that also incorporates heterogeneous sources of information that may be available on the users and/or items can become a much more effective recommender, in terms of both increased relevance of the predictions as well as explainability of the results. In this paper, we propose a Bayesian approach that exploits not only such "side-information", but also a different kind of heterogeneity that captures the variations in the mapping from user/item attributes to the affinities of interest. Such predictive heterogeneity is likely to occur in large recommender systems that involve a diverse set of users, and can be mitigated by using multiple localized predictive models rather than a single global one that covers all user-item pairs. The scope or coverage of each local model is determined simultaneously with the model parameters. The proposed approach can incorporate different types of inputs to predict the preferences of diverse users and items. We compare it against well-known alternative approaches and analyze the results in terms of both accuracy and interpretability.

show abstract

“…Clustering on such data is useful in many applications including product recommendations, customer and product segmentation, and identifying various customer and market trends. However, typically only a small subset of customers show statistically significant coherent buying behavior and that too when one focuses only a small subset of products [Strehl and Ghosh 2003;Deodhar and Ghosh 2007;Wedel and Steenkamp 1991]. Therefore, a clustering algorithm for such datasets should have the ability to prune out (potentially large) sparse and noisy portions of the data to uncover the highly coherent clusters.…”

Section: Introductionmentioning

confidence: 99%

Bregman Bubble Clustering: A Robust Framework for Mining Dense Clusters

Ghosh¹,

Gupta²

2012

Intelligent Systems Reference Library

Self Cite

View full text Add to dashboard Cite

In classical clustering, each data point is assigned to at least one cluster. However, in many applications only a small subset of the available data is relevant for the problem and the rest needs to be ignored in order to obtain good clusters. Certain nonparametric density-based clustering methods find the most relevant data as multiple dense regions, but such methods are generally limited to low-dimensional data and do not scale well to large, high-dimensional datasets. Also, they use a specific notion of "distance", typically Euclidean or Mahalanobis distance, which further limits their applicability. On the other hand, the recent One Class Information Bottleneck (OC-IB) method is fast and works on a large class of distortion measures known as Bregman Divergences, but can only find a single dense region. This article presents a broad framework for finding k dense clusters while ignoring the rest of the data. It includes a seeding algorithm that can automatically determine a suitable value for k. When k is forced to 1, our method gives rise to an improved version of OC-IB with optimality guarantees. We provide a generative model that yields the proposed iterative algorithm for finding k dense regions as a special case. Our analysis reveals an interesting and novel connection between the problem of finding dense regions and exponential mixture models; a hard model corresponding to k exponential mixtures with a uniform background results in a set of k dense clusters. The proposed method describes a highly scalable algorithm for finding multiple dense regions that works with any Bregman Divergence, thus extending density based clustering to a variety of non-Euclidean problems not addressable by earlier methods. We present empirical results on three artificial, two microarray and one text dataset to show the relevance and effectiveness of our methods.

show abstract

A framework for simultaneous co-clustering and learning from complex data

Cited by 58 publications

References 22 publications

Applying CHAID for logistic regression diagnostics and classification accuracy improvement

Applying CHAID for logistic regression diagnostics and classification accuracy improvement

Learning multiple models for exploiting predictive heterogeneity in recommender systems

Bregman Bubble Clustering: A Robust Framework for Mining Dense Clusters

Contact Info

Product

Resources

About