A preliminary task of sentiment analysis aims to detect polarities of a text either positive or negative. These texts vary from movie reviews to customer comments on electronic devices. The polarity detector trained in one domain may not achieve remarkable results in another domain. In this study, we provide a training and test dataset generator for domain-specific sentiment analysis in which machine learning methods can be trained without any human labor. To do it, we extract comments and polarity scores from a popular e-commerce website for electronic devices in Turkey. Also, we translate a well-known sentiment lexicon into Turkish and use this lexicon in a lexicon-based polarity detection. We compare well-known machine learning methods trained by automatically labeled dataset and lexicon-based method for Turkish texts in the electronics domain. The experimental setup is conducted by the generated evaluation dataset for the e-commerce domain. The test set contains 30% randomly selected from the generated dataset. Our lexicon-based method has achieved 0.62 F1 score and other three supervised learning methods achieved the best results. Further, logistic regression gets the highest score when it uses count vectorizer as a feature extraction mechanism.