Impact of preprocessing on medical data classification

AlMuhaideb, Sarab; Menaï, Mohamed El Bachir

doi:10.1007/s11704-016-5203-5

Cited by 30 publications

(14 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Tujuan dari pra-pemrosesan ini adalah untuk menentukan karakteristik data medis yang mencakup noise, data tak lengkap, dan fitur tak relevan. Metode pra-pemrosesan dilakukan dengan operasi diskretisasi atribut numerik, pemilihan atribut subset dan penanganan nilai hilang [6]. Termasuk dalam hal pra-pemrosesan data, [7] mengusulkan perbaikan algoritma kNN dengan menggunakan matrik korelasi untuk merekonstruksi titik-titik data tes untuk mendapatkan nilai k yang paling tepat mengacu kepada matrik korelasi kNN [7].…”

Section: Pendahuluanunclassified

Perbandingan Kinerja Metode Pra-Pemrosesan Dalam Pengklasifikasian Otomatis Dokumen Paten

Nugroho

Denih

2020

KOMPUTASI JIIKM

View full text Add to dashboard Cite

This paper presents a performance analysis and comparison of several pre-processing methods used in automatic patent classification with graph kernels for Support Vector Machine (SVM). The pre-processing methods are based on the data transform techniques, namely data scaling, data centering, data standardization, data normalization, the Box-Cox transform and the Yeo-Johnson transform. The automatic patent classification is designed to classify an input of patent citation graphs into one of 10 possible classes of the International Patent Classification (IPC). The input is taken with various background conditions. The experiments showed that the best result is achieved when the pre-processing method is data normalization, achieving a classification accuracy of up to 85.33.15% for the KEHL and 93.80% for the KVHL. In contrast, for the KEHG, the preprocessing method application decreased the accuracy.

show abstract

Section: Pendahuluanunclassified

Perbandingan Kinerja Metode Pra-Pemrosesan Dalam Pengklasifikasian Otomatis Dokumen Paten

Nugroho

Denih

2020

KOMPUTASI JIIKM

View full text Add to dashboard Cite

show abstract

“…Data preprocessing tasks are necessary to transform the original raw information with incomplete, inconsistent, and noisy data into a high-quality and cleaned dataset for subsequent analysis. The classification performance can be improved mainly by selecting the right combination of preprocessing methods [62]. There is no predefined sequence of preparation steps.…”

Section: Data Preprocessingmentioning

confidence: 99%

“…Step 5: Discretization This process is performed on the numerical features to partition values into a finite number of non-overlapping intervals. Finding the optimal discretization of a feature is NP-hard [62]. There are two main techniques of discretization, namely supervised method, where the class feature is considered, and the unsupervised method, where the class feature is not considered.…”

Section: Step 3: Outlier Detection and Preventionmentioning

confidence: 99%

A Comprehensive Medical Decision–Support Framework Based on a Heterogeneous Ensemble Classifier for Diabetes Prediction

et al. 2019

View full text Add to dashboard Cite

Early diagnosis of diabetes mellitus (DM) is critical to prevent its serious complications. An ensemble of classifiers is an effective way to enhance classification performance, which can be used to diagnose complex diseases, such as DM. This paper proposes an ensemble framework to diagnose DM by optimally employing multiple classifiers based on bagging and random subspace techniques. The proposed framework combines seven of the most suitable and heterogeneous data mining techniques, each with a separate set of suitable features. These techniques are k-nearest neighbors, naïve Bayes, decision tree, support vector machine, fuzzy decision tree, artificial neural network, and logistic regression. The framework is designed accurately by selecting, for every sub-dataset, the most suitable feature set and the most accurate classifier. It was evaluated using a real dataset collected from electronic health records of Mansura University Hospitals (Mansura, Egypt). The resulting framework achieved 90% of accuracy, 90.2% of recall = 90.2%, and 94.9% of precision. We evaluated and compared the proposed framework with many other classification algorithms. An analysis of the results indicated that the proposed ensemble framework significantly outperforms all other classifiers. It is a successful step towards constructing a personalized decision support system, which could help physicians in daily clinical practice.

show abstract

“…The preprocessing method includes a number of stages such as: data extraction, data cleaning, data fusion, data reduction, and feature construction [9]. The preprocessing method can be done through the process of discretization of numeric attributes, subset attribute selection, and handling missing value [10]. Studies conducted by Krawczyk modified hybrid ensemble methods and incorporated feature selection processes at the bagging stage by using a genetic algorithm, but the method they proposed failed to address the problem of data accuracy and their results showed that diversity measurements need attention [11].…”

Section: Related Workmentioning

confidence: 99%

Hybrid Approach Redefinition (HAR) model for optimizing hybrid ensembles in handling class imbalance: a review and research framework

et al. 2018

View full text Add to dashboard Cite

The purpose of this research is to develop a research framework to optimize the results of hybrid ensembles in handling class imbalance issues. The imbalance class is a state in which the classification results give the number of instances in a class much larger than the number of instances in the other class. In machine learning, this problem can reduce the prediction accuracy and also reduce the quality of the resulting decisions. One of the most popular methods of dealing with class imbalance is the method of ensemble learning. Hybrid Ensembles is an ensemble learning method approach that combines the use of bagging and boosting. Optimization of Hybrid Ensembles is done with the intent to reduce the number of classifier and also obtain better data diversity. Based on an iterative methodology, we review, analyze, and synthesize the current state of the literature and propose a completely new research framework for optimizing Hybrid Ensembles. In doing so, we propose a new taxonomy in ensemble learning that yields a new approach of sampling-based Ensembles and will propose an optimization Hybrid Ensembles using Hybrid Approach Redefinition (HAR) Model that combines the use of Hybrid Ensembles and Sampling Based Ensembles methods. We further provide an empirical analysis of the reviewed literature and emphasize the benefits that can be achieved by optimizing Hybrid Ensembles.

show abstract

Impact of preprocessing on medical data classification

Cited by 30 publications

References 55 publications

Perbandingan Kinerja Metode Pra-Pemrosesan Dalam Pengklasifikasian Otomatis Dokumen Paten

Perbandingan Kinerja Metode Pra-Pemrosesan Dalam Pengklasifikasian Otomatis Dokumen Paten

A Comprehensive Medical Decision–Support Framework Based on a Heterogeneous Ensemble Classifier for Diabetes Prediction

Hybrid Approach Redefinition (HAR) model for optimizing hybrid ensembles in handling class imbalance: a review and research framework

Contact Info

Product

Resources

About