Performance Analysis of Hard and Soft Clustering Approaches For Gene Expression Data

Banu, P. K. Nizar; Andrews, S.

doi:10.4018/ijrsda.2015010104

Cited by 15 publications

(9 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…has been trying to mine gene expression to predict gene expression patterns and assist clinicians in early diagnosis of tumor formation [13]. Using clustering techniques from data mining perspective is trying to reveal the similarity between genes or a set of genes with similar conditions.…”

Section: Related Workmentioning

confidence: 99%

Supervised Machine Learning Model for High Dimensional Gene Data in Colon Cancer Detection

Chen

Zhao

Shen

et al. 2015

2015 IEEE International Congress on Big Data

View full text Add to dashboard Cite

With well-developed methods in gene level data extraction, there are huge amount of gene expression data, including normal composition and abnormal ones. Therefore, mining gene expression data is currently an urgent research question, for detecting a corresponding pattern, such as cancer species, quickly and accurately. Since gene expression data classification problem has been widely studied accompanying with the development of gene technology, by far numerous methods, mainly neural network related, have been deployed in medical data analysis, which is mainly dealing with the high dimension and small quantity. A lot of research has been conducted on clustering approaches, extreme learning machine and so on. They are usuallly applied in a shallow neural network model. Recently deep learning has shown its power and good performance on high dimensional datasets. Unlike current popular deep neural network, we will continue to apply shallow neural network but develop an innovative algorithm for shallow neural network. In the supervised model, we demonstrate a shallow neural network model with a batch of parameters, and narrow its computational process into several positive parts, which process smoothly for a better result and finally achieve an optimal goal. It shows a stable and excellent result comparable to deep neural network. An analysis of thealgorithm is also presented in this paper. Disciplines Engineering | Science and Technology Studies Publication DetailsChen, H., Zhao, H., Shen, J., Zhou, R. & Zhou, Q. (2015). Supervised machine learning model for high dimensional gene data in colon cancer detection. Abstract-With well-developed methods in gene level data extraction, there are huge amount of gene expression data, including normal composition and abnormal ones. Therefore, mining gene expression data is currently an urgent research question, for detecting a corresponding pattern, such as cancer species, quickly and accurately. Since gene expression data classification problem has been widely studied accompanying with the development of gene technology, by far numerous methods, mainly neural network related, have been deployed in medical data analysis, which is mainly dealing with the high dimension and small quantity. A lot of research has been conducted on clustering approaches, extreme learning machine and so on. They are usuallly applied in a shallow neural network model. Recently deep learning has shown its power and good performance on high dimensional datasets. Unlike current popular deep neural network, we will continue to apply shallow neural network but develop an innovative algorithm for shallow neural network. In the supervised model, we demonstrate a shallow neural network model with a batch of parameters, and narrow its computational process into several positive parts, which process smoothly for a better result and finally achieve an optimal goal. It shows a stable and excellent result comparable to deep neural network. An analysis of the algorithm is also presented in this paper.

show abstract

Section: Related Workmentioning

confidence: 99%

Supervised Machine Learning Model for High Dimensional Gene Data in Colon Cancer Detection

Chen

Zhao

Shen

et al. 2015

2015 IEEE International Congress on Big Data

View full text Add to dashboard Cite

show abstract

“…Following that, in [26], Kulvaibhav et al propose an enhanced genetic clustering method to diagnose and predict diseases using patient's history, symptoms, and existing medical condition. Afterwards, Banu et al [7] utilize Self Organizing Maps based clustering [27] for brain tumor gene expression dataset. Due to recent technological advances in wireless body area networks (WBANs), there also emerges numerous research work for clustering analysis of the massive context-aware data generated by various WBANs sensors in health monitoring systems, aiming at providing personalized healthcare services for patients in a timely and appropriate manner.…”

Section: Introductionmentioning

confidence: 99%

“…This has paved the ways for cloud data mining technologies deployment in telemedicine and mobile healthcare services [1,2], which enables long-term continuous medical monitoring for many purposes: for outpatients with chronic medical conditions (eg, diabetes, arthritis, stroke, Ischaemic heart disease, etc.) [3,4], individuals seeking to change behavior (such as being less sedentary, playing sports, losing weight, or sleeping more) [5,6], or physicians needing to quantify and detect disease or behavioral aberrations for early diagnosis or alert generation [7,8]. In spite of the huge potential to improve healthcare quality, improve efficiency and reduce cost, these rapidly evolving cloud mining technologies also produce the inevitable problem of the privacy revelation, which arouses increasing attentions by the field of information security and social aspects [9,10].…”

Section: Introductionmentioning

confidence: 99%

A semi-supervised privacy-preserving clustering algorithm for healthcare

Huang

Chen

et al. 2015

Peer-to-Peer Netw. Appl.

View full text Add to dashboard Cite

With the proliferation of healthcare data, the cloud mining technology for E-health services and applications has become a hot research topic. While on the other hand, these rapidly evolving cloud mining technologies and their deployment in healthcare systems also pose potential threats to patient's data privacy. In order to solve the privacy problem in the cloud mining technique, this paper proposes a semisupervised privacy-preserving clustering algorithm. By employing a small amount of supervised information, the method first learns a Large Margin Nearest Cluster metric using convex optimization. Then according to the trained metric, the method imposes multiplicative perturbation on the original data, which can change the distribution shape of the original data and thus protect the privacy information as well as ensuring high data usability. The experimental results on the brain fiber dataset provided by the 2009 PBC demonstrated that the proposed method could not only protect data privacy towards secure attacks, but improve the clustering purity.

show abstract

“…However, in real life, data objects may naturally belong to either single or multiple groups, which limits the application of hard clustering [4]. Therefore, fuzzy clustering techniques have been developed [5]. Fuzzy clustering is more flexible because it allows every object to belong to more than one cluster [6].…”

Section: Introductionmentioning

confidence: 99%