Impact of Similarity Measures on Web-page Clustering

By Alexander Strehl, Joydeep Ghosh, Raymond J. Mooney. 2000

863 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/e20ad680666a4e02a1b0a99d0b6c974f7390bc64

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Efficient phrase-based document indexing for Web document clustering (2004, background): “The cosine measure was chosen due toitswideuseinthedocumentclusteringliterature,andsince it is described as being able to capture human categorization behavior well [30].”
Cluster ensembles --- a knowledge reuse framework for combining multiple partitions (2003, methodology): “), or techniques (graph-based, agglomerative, k-means) (Strehl et al., 2000).”
Introduction to the special topic section on mining Web resources for enhancing information retrieval (2007, background): “Thus, we say that a positive-path web FD x → y holds if M(x,y) > T , where M is a similarity metric and T is a threshold between 0 and 1. With this more general definition , a host of similarity metrics from IR, such as cosine or Jaccard’s ( Srehl, Ghosh, & Mooney, 2000 ), each with different proper”
Why is Tanimoto index an appropriate choice for fingerprint-based similarity calculations? (2015, background): “texture image retrieval [8], webpage clustering [9] or event identification in social media [10].”
Semi-supervised graph clustering: a kernel approach (2005, methodology): “…Mutual Information (NMI): it is a standard technique for determining the quality of clusters, by measuring the amount of statistical information shared by the random variables representing the cluster distribution and the underlying class distribution of the data points (Strehl et al. 2000).”
Exploiting hierarchical domain structure to compute similarity (2003, methodology): “Other such measures include the Inclusion Measure, the Overlap Coef.cient, and the Extended Jaccard Coef.cient [Strehl et al. 2000; van Rijsbergen 1979].”
Frequency-sensitive competitive learning for scalable balanced clustering on high-dimensional hyperspheres (2004, methodology): “clusters [52] can be used if external information such as class labels for all data points can be obtained [2].”
Similarity Measures for Text Document Clustering (2008, methodology): “Our results broadly agree with Strehl et al’s [17].”
A generalized maximum entropy approach to bregman co-clustering and matrix approximation (2004, unknown): “Since the co-clustering objective functions based on the different divergences are not comparable and sometimes not even well-defined 14, we measure the co-clustering quality in terms of the average of the normalized mutual information (Strehl et al., 2000) between the clustering and true class labe”
Intelligent Code Completion with Bayesian Networks (2015, methodology): “To determine the distance between two object usages, the cosine similarity [Strehl et al. 2000] is used, which is also a common choice in the research area of information retrieval.”
Clustering with Bregman Divergences (2005, background): “However, in many data mining applications, this distortion function is not a good match with the data, and consequently kmeans performs poorly as compared to other approaches [25].”
Deep Neural Networks for Learning Graph Representations (2016, methodology): “5.2.4 Importance of Deep Structures To understand necessity of deep structures, we further measured the layerwise NMI value of 20-NewsGroup.”
Text Mining Infrastructure in R (2008, methodology): “Common similarity measures in text mining are Metric Distances, Cosine Measure, Pearson Correlation and Extended Jaccard Similarity (Strehl et al. 2000).”
Finding Clusters of Different Sizes, Shapes, and Densities in Noisy, High Dimensional Data (2003, background): “For example, the Southern Oscillation Index (SOI) measures the sea level pressure (SLP) anomalies between Darwin, Australia and Tahiti and is associated with El Nino, the anomalous warming of the eastern tropical region of the Pacific that has been linked to climate phenomena such as droughts in Aus”