Combination of PCA with SMOTE Oversampling for Classification of High-Dimensional Imbalanced Data

Abstract

Imbalanced data classification is a common issue in data mining where the classifiers are skewed towards the larger data class. Classification of high-dimensional skewed (imbalanced) data is of great interest to decisionmakers as it is more difficult to. The dimension reduction method, a process in which variables are reduced, allows high dimensional datasets to be interpreted more easily with a certain loss. This study, a method combining SMOTE oversampling with principal component analysis is proposed to solve the imbalance problem in high dimensional data. Three classification algorithms consisting of Logistic Regression, K-Nearest Neighbor, Decision Tree methods and two separate datasets were utilized to evaluate the suggested method's efficacy and determine the classifiers' performance. Respectively, raw datasets, converted datasets by PCA, SMOTE and SMOTE+PCA (SMOTE and PCA) methods, were analyzed with the given algorithms. Analyzes were made using WEKA. Analysis results suggest that almost all classification algorithms improve their classification performance using PCA, SOMTE, and SMOTE+PCA methods. However, the SMOTE method gave more efficient results than PCA and PCA+SMOTE methods for data rebalancing. Experimental results also suggest that K-Nearest Neighbor classifier provided higher classification performance compared to other algorithms.

Artificial Intelligence-based Colon Cancer Prediction by Identifying Genomic Biomarkers

“…In datasets with class imbalance problem, most machine learning techniques ignore minority class performance and therefore underperform in minority class. One approach to these datasets is to oversample the minority class and is called the Synthetic Minority Oversampling Technique, or SMOTE for short (9). In order to eliminate the class imbalance problem in the colon cancer gene expression dataset (22 normal and 40 tumor tissues), the SMOTE method was applied before feature selection.…”

Analysis and Classification of Autism Data Using Machine Learning Algorithms

“…It is the process of recognizing patterns, concepts, and other objects in order to better comprehend them and classify them based on incoming data [5]. Classification can help uncover abnormalities when developing a learning model from prior data [4]. There are various classification algorithms, each of which builds a prediction model in a different way.…”

“…It can be used to create guesses regarding category variable names [34]. Each branch might be relegated to the training sample category [4]. The decision tree is formulated as follows:…”

“…In this paper, we studied and analyzed autism data among children affected in the Dohuk governorate using Machine Learning (ML) methods. ML, as a branch of artificial intelligence, is the process of applying computers to real-world problems, to better analyze, train, and model data, and hence perform faster and provide better predictions [4]. For this purpose, a new dataset of autistic children in Duhok was created by collecting data from 515 cases in different centers.…”

“…ML algorithms are categorized into two main types, supervised (with labeled data) and unsupervised (unlabeled data) [4]. Classification, which is the main core of this study, is one of the most popular methods of supervised learning, which is based on predicting the output from a set of input variables [5].…”