?? Memnuniyeti Faktörlerini Belirlemek ve Analiz Etmek için Çevrimiçi Çal??an De?erlendirmelerini Kullanan ilgili Makine Ö?renmesi ve Topluluk Ö?renmesi Tabanl? Yöntem

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