An ensemble-clustering-based distance metric and its applications

Abdallah, Loai; Shimshoni, Ilan

doi:10.1504/ijbidm.2013.059052

Cited by 5 publications

(9 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The main goal of their algorithm is to develop a selective sampling methodology for nearest-neighbor (NN) classification learning algorithms. Later, Abdallah and Shimshoni (2013) developed a selective sampling algorithm based on ensemble clustering. Dasgupta and Hsu (2008) use hierarchical clustering.…”

Section: Related Workmentioning

confidence: 99%

“…The first one is based on selective sampling for the nearest neighbor classifier (LSS) (Lindenbaum et al, 2004), and the second one is based on selective sampling using ensemble the clustering based distance metric (LSS-EC) (Abdallah and Shimshoni, 2013). In this section we will describe the LSS algorithm for incomplete datasets.…”

Section: Lookahead Algorithm For Selective Sampling For Incomplete Damentioning

confidence: 99%

“…One version is for the original LSS described by Lindenbaum et al (2004) while the other for LSS-EC described by Abdallah and Shimshoni (2013).…”

Section: Introductionmentioning

confidence: 99%

“…As a result, in this research we generalize this distance function to deal with other types of missing values. Abdallah and Shimshoni (2013) show that LSS using ensemble clustering (EC) performed better than LSS using the Euclidean distance. As a result, in this research we also decided to extend this algorithm (i.e., LSS-EC) for datasets with missing values.…”

mentioning

confidence: 99%

“…This new space has two important advantages. Firstly, there is a distance metric defined by Abdallah and Shimshoni (2013) that can reflect the similarity between the objects better than the Euclidean distance. Moreover, it is a new way to represent the incomplete data as complete categorical data using cluster labels.…”

mentioning

confidence: 99%

See 4 more Smart Citations

Lookahead selective sampling for incomplete data

Abdallah

Shimshoni

2016

International Journal of Applied Mathematics and Computer Science

Self Cite

View full text Add to dashboard Cite

Missing values in data are common in real world applications. There are several methods that deal with this problem. In this paper we present lookahead selective sampling (LSS) algorithms for datasets with missing values. We developed two versions of selective sampling. The first one integrates a distance function that can measure the similarity between pairs of incomplete points within the framework of the LSS algorithm. The second algorithm uses ensemble clustering in order to represent the data in a cluster matrix without missing values and then run the LSS algorithm based on the ensemble clustering instance space (LSS-EC). To construct the cluster matrix, we use the k-means and mean shift clustering algorithms especially modified to deal with incomplete datasets. We tested our algorithms on six standard numerical datasets from different fields. On these datasets we simulated missing values and compared the performance of the LSS and LSS-EC algorithms for incomplete data to two other basic methods. Our experiments show that the suggested selective sampling algorithms outperform the other methods.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Lookahead Algorithm For Selective Sampling For Incomplete Damentioning

confidence: 99%

“…One version is for the original LSS described by Lindenbaum et al (2004) while the other for LSS-EC described by Abdallah and Shimshoni (2013).…”

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

mentioning

confidence: 99%