TY - JOUR AU - Coban, Onder AU - Ozyildirim, Buse Melis AU - Ozel, Selma Ayse PY - 2018/09/26 Y2 - 2024/03/28 TI - An Empirical Study of the Extreme Learning Machine for Twitter Sentiment Analysis JF - International Journal of Intelligent Systems and Applications in Engineering JA - Int J Intell Syst Appl Eng VL - 6 IS - 3 SE - Research Article DO - 10.18201/ijisae.2018644774 UR - https://ijisae.org/index.php/IJISAE/article/view/677 SP - 178-184 AB - <p>Extreme Learning Machine (ELM) method is proposed for single hidden layer feed-forward networks (SLFNs). The ELM<br />employs feed-forward neural network architecture and works with randomly determined input weights. In this aspect, ELM depends on<br />principle that enables to determine weights and biases in the network. In the first phase of ELM that can be named as feature mapping,<br />the usage of random values differs the ELM from other methods that employ a kernel function for feature mapping such as Support<br />Vector Machines (SVM) and Deep Neural Networks. After the feature mapping, the main goal of the ELM is to learn weights between<br />hidden and output layers by minimizing the error. The ELM has gained much more popularity recently; and can be utilized for<br />classification, regression, and dimension reduction. In literature, Twitter sentiment analysis is generally considered as a classification<br />task. Therefore, in this study, the basic ELM is utilized for Twitter sentiment analysis and compared with the SVM which is one of the<br />most successful machine learning algorithms used for sentiment analysis. Experiments are conducted on two different Turkish datasets.<br />Experimental results show that the performance of the two methods are slightly different, but SVM outperforms basic ELM.</p> ER -