Abstract
Sentiment classification has received increasing attention in recent years. Supervised learning methods for sentiment classification require considerable amount of labeled data for training purposes. As the number of domains increases, the task of collecting data becomes impractical. Therefore, domain adaptation techniques are employed. However, most of the studies dealing with the domain adaptation problem demand a few amount of labeled data or lots of unlabeled data belonging to the target domain, which may not be always possible. In this work, a novel method for sentiment classification, which does not require labeled and/or unlabeled data from the target domain, is proposed. The propose method mainly consists of two stages. At first, the target domain is predicted even if it is not among the source domains in hand. Then, sentiment is classified as either positive or negative using the sentiment classifier specifically trained for the predicted domain. Extensive experimental analysis on two different datasets with distinct languages and domains verifies that the proposed method is superior to the domain independent sentiment classification approach at each case considered.DOI: http://dx.doi.org/10.5755/j01.eie.22.2.14599
Highlights
Nowadays, most people share their opinions through the internet on which the products and services they buy or use
As mentioned in the related work, most of the methods developed for domain adaptation problem in sentiment classification require labeled or unlabeled data on the target domain
Naïve Bayesian (NB) and Support Vector Machine (SVM) classification algorithms [18] were employed for the domain prediction and sentiment classification purposes
Summary
Most people share their opinions through the internet on which the products and services they buy or use. Instead, existing data rather than collecting new one may be used to minimize the need for labeled data to train the classifier on a new domain. This approach is called as domain adaptation [4]. A novel method for sentiment classification, which does not require neither labeled nor unlabeled data on the target domain, is proposed. In this two-stage method, the target domain is first predicted. Experimental analysis verifies that the proposed method beats the domain independent sentiment classification approaches.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have