An enhanced topologically significant directed random walk in cancer classification using gene expression datasets

Seah, Choon Sen; Kasim, Shahreen; Fudzee, Mohd Farhan Md; Ping, Jeffrey Mark Law Tze; Mohamad, Mohd Saberi; Saedudin, Rd. Rohmat; Ismail, Mohd Arfian

doi:10.1016/j.sjbs.2017.11.024

Cited by 12 publications

(9 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thus, one uses eigenvector centrality as a refinement of degree centrality. Generally, eigenvector centrality can be used to measure the influence of a node in a network . A recursive definition of eigenvector centrality, in which the eigenvector centrality of a vertex is defined as being proportional to the sum of the eigenvector centrality of the vertices to which it is connected, is also helpful for understanding the importance of eigenvector centrality .…”

Section: Resultsmentioning

confidence: 99%

From Infection Clusters to Metal Clusters: Significance of the Lowest Occupied Molecular Orbital (LOMO)

Tsuji

Yoshizawa

2021

ACS Omega

View full text Add to dashboard Cite

In this paper, the nature of the lowest-energy electrons is detailed. The orbital occupied by such electrons can be termed the lowest occupied molecular orbital (LOMO). There is a good correspondence between the Huckel method in chemistry and graph theory in mathematics; the molecular orbital, which chemists view as the distribution of an electron with a specific energy, is to mathematicians an algebraic entity, an eigenvector. The mathematical counterpart of LOMO is known as eigenvector centrality, a centrality measure characterizing nodes in networks. It may be instrumental in solving some problems in chemistry, and also it has implications for the challenge facing humanity today. This paper starts with a demonstration of the transmission of infectious disease in social networks, although it is unusual for a chemistry paper but may be a suitable example for understanding what the centrality (LOMO) is all about. The converged distribution of infected patients on the network coincides with the distribution of the LOMO of a molecule that shares the same network structure or topology. This is because the mathematical structures behind graph theory and quantum mechanics are common. Furthermore, the LOMO coefficient can be regarded as a manifestation of the centrality of atoms in an atomic assembly, indicating which atom plays the most important role in the assembly or which one has the greatest influence on the network of these atoms. Therefore, it is proposed that one can predict the binding energy of a metal atom to its cluster based on its LOMO coefficient. A possible improvement of the descriptor using a more sophisticated centrality measure is also discussed.

show abstract

Section: Resultsmentioning

confidence: 99%

From Infection Clusters to Metal Clusters: Significance of the Lowest Occupied Molecular Orbital (LOMO)

Tsuji

Yoshizawa

2021

ACS Omega

View full text Add to dashboard Cite

show abstract

“…The Bioconductor will analyse the expression value and further arrange the dataset using normalization which narrows the range of data to be studied. In this study, the dataset will undergo 3 pre-processing stages before being applied into the real classification algorithms such as genetic algorithm [10], pathway based cancer classification [11], and significant directed random walk (sDRW) [12]. Figure 1 illustrate the phases in data pre-processing stage.…”

Section: Methodsmentioning

confidence: 99%

“…Data pre-processing not only clears up the dataset to be ready for the implementation purpose but also allow the researchers to select the right attributes that would be the key influences for the study. In sDRW [12], Seah believes the weight of genes plays an important role in affecting the tumour formation and hence, during data pre-processing stage, he was focusing on those attribute that are related to gene's weight.…”

Section: Figure 3 Step 2 Normalizationmentioning

confidence: 99%

An Effective Pre-Processing Phase for Gene Expression Classification

Seah¹,

Kasim²,

Fudzee³

et al. 2018

IJEECS

Self Cite

View full text Add to dashboard Cite

A raw dataset prepared by researchers comes with a lot of information. Whether the information is usefull or not, completely depends on the requirement and purposes. In machine learning, data pre-processing is the very initial stage. It is a must to make sure the dataset is totally suitable for the requirement. In significant directed random walk (sDRW), there are three steps in data pre-processing stage. First, we remove unwanted attributes, missing value and proper arrangement, followed by normalization of the expression value and lastly, filtering method is applied. The first two steps are completed by Bioconductor package while the last step is works in sDRW.

show abstract

“…The wrapper evaluates and selects attributes based on accuracy estimates by the target-learning algorithm. A filter usually works alone with statistical correlation to determine the importance of features data with low complexity [17].…”

Section: Classification Algorithmmentioning

confidence: 99%

Analysis of Attribute Selection and Classification Algorithm Applied to Hepatitis Patients

Samsuddin

Shah

Saedudin

et al. 2019

International Journal on Advanced Science, Engineering and Information Technology

Self Cite

View full text Add to dashboard Cite

Data mining techniques are widely used in classification, attribute selection and prediction in the field of bioinformatics because it helps to discover meaningful new correlations, patterns and trends by sifting through large volume of data, using pattern recognition technologies as well as statistical and mathematical techniques. Hepatitis is one of the most important health problem in the world. Many studies have been performed in the diagnosis of hepatitis disease but medical diagnosis is quite difficult and visual task which is mostly done by doctors. Therefore, this research is conducted to analyse the attribute selection and classification algorithm that applied to hepatitis patients. In order to achieve goals, WEKA tool is used to conduct the experiment with different attribute selector and classification algorithm . Hepatitis dataset that are used is taken from UC Irvine repository. This research deals with various attribute selector namely CfsSubsetEval, WrapperSubsetEval, GainRatioSubsetEval and CorrelationAttributeEval. The classification algorithm that used in this research are NaiveBayesUpdatable, SMO, KStar, RandomTree and SimpleLogistic. The results of the classification model are time and accuracy. Finally, it concludes that the best attribute selector is CfsSubsetEval while the best classifier is given to SMO because SMO performance is better than other classification techniques for hepatitis patients.

show abstract

An enhanced topologically significant directed random walk in cancer classification using gene expression datasets

Cited by 12 publications

References 37 publications

From Infection Clusters to Metal Clusters: Significance of the Lowest Occupied Molecular Orbital (LOMO)

From Infection Clusters to Metal Clusters: Significance of the Lowest Occupied Molecular Orbital (LOMO)

An Effective Pre-Processing Phase for Gene Expression Classification

Analysis of Attribute Selection and Classification Algorithm Applied to Hepatitis Patients

Contact Info

Product

Resources

About