Machine Learning and Ensemble Learning Based Method Using Online Employee Assessments to Identify and Analyze Job Satisfaction Factors

Authors : Ali ÖZDEMİR, Aytuğ ONAN, Vildan ÇINARLI ERGENE

Pages : 19-28

Doi:10.31590/ejosat.1173627

View : 91 | Download : 14

Publication Date : 2022-09-30

Article Type : Research Paper

Abstract :In this paper it was emphasized that machine learning techniques can achieve high performance in classification and work effectively and scalably with large data sets. The dataset used in this study was obtained from www.kaggle.com. A total of 67529 comments collected from people working at Google, Amazon, Netflix, Facebook, Apple and Microsoft were evaluated. The N-gram model is an important representation scheme in text mining. N-gram models are the unigram model insert ignore into journalissuearticles values(N = 1);, bigram insert ignore into journalissuearticles values(N = 2);, and trigram insert ignore into journalissuearticles values(N = 3);. Three different weighting schemes as TP, TF, and TF-IDF, and three different weighting schemes for traditional machine learning-based analysis as N-gram model insert ignore into journalissuearticles values(bigram, unigram and trigram); was used. Five supervised learning algorithm was used to train models: Naive Bayes, Support Vector Machines insert ignore into journalissuearticles values(SVM);, Logistic Regression insert ignore into journalissuearticles values(LR);, K-Nearest Neighbor insert ignore into journalissuearticles values(KNN); and Random Forest insert ignore into journalissuearticles values(RF);.
Keywords : Makine öğrenmesi, metin sınıflandırma, yapay zeka, topluluk öğrenmesi

ORIGINAL ARTICLE URL