Model optimisation of class imbalanced learning using ensemble classifier on over-sampling data

Abstract

<span lang="EN-US">Data imbalance is one of the problems in the application of machine learning and data mining. Often this data imbalance occurs in the most essential and needed case entities. Two approaches to overcome this problem are the data level approach and the algorithm approach. This study aims to get the best model using the pap smear dataset that combined data levels with an algorithmic approach to solve data imbalanced. The laboratory data mostly have few data and imbalance. Almost in every case, the minor entities are the most important and needed. Over-sampling as a data level approach used in this study is the synthetic minority oversampling technique-nominal (SMOTE-N) and adaptive synthetic-nominal (ADASYN-N) algorithms. The algorithm approach used in this study is the ensemble classifier using AdaBoost and bagging with the classification and regression tree (CART) as learner-based. The best model obtained from the experimental results in accuracy, precision, recall, and f-measure using ADASYN-N and AdaBoost-CART.</span>

Optimasi Metode Supervised Learning Dengan Menggunakan Particle Swarm Optimization Untuk Deteksi Malware

“…Salah satu Teknik untuk mengurangi kelas tidak seimbang digunakan Teknik data sampling yang dapat membagi kelas mayoritas menjadi kelas yang seimbang [10]. Teknik data sampling berhasil meningkatkan akurasi yang dihaslkan Ensemble Classifier [11], Teknik data sampling dapat mengatisipasi data kelas yang tidak seimbang [12] Selain data sampling untuk meningkatkan nilai akurasi dari klasifikasi adalah metode pemilihan fitur dengan memilih subset fitur yang dipilih dari jumlah dataset [13], seleksi fitur adalah proses dimana Sebagian ruang fitur dipilih sesuai relevansinya dengan mempertimbangkan keluaran dari klasifikasi [14]. PSO merupakan algoritma Optimasi fitur dapat meningkatkan nilai akurasi yang dihasilkan PSO telah digunakan untuk optimasi peningkatan akurasi untuk Random forest, decision tree, naïve bayes dan KKN untuk klasifikasi dataset diabetes [15].…”

Efforts of Performance Optimization: The Experiment on Ten Accounting Datasets

“…Moreover, [17] observed that the Adaboost method could be applied to improve classifier performance. In addition, [18], [19], and [20] also conducted experiments by assessing comparisons among SMOTE, Adaboost, and bagging techniques to increase the accuracy level of a prediction.…”