Aspect based feature extraction and sentiment classification of review data sets using Incremental machine learning algorithm
Rajalaxmi Hegde, S. Seema · 2017
As there is a exponential growth of social networks and due to large usage of social media, there is a increasing demand for data in the web for the users which leads to recent trends and ideas in the field of research. The users will be eagerly using these data for the future purpose and get information about the opinions of others thus there is a need of automatic summarization of opinion of the web users. Opinion Mining is a process of extracting and analyzing people's opinion regarding the given object and Sentiment Analysis explains about the hidden sentiments in the opinions. A most important challenge in this is to identify the sentiments and aspects and then perform the data classification based on these features and this process is called aspect based feature extraction. Opinion summarization can be done using machine learning such as maximum entropy, naive bayes, Support vector machine (SVM) model and, Random forest technique. The data sources to be taken for the proposed method are from customer review of product. In this paper experiments were conducted to compare the performance of proposed iterative decision tree method against other machine learning algorithms like SVM, Baseline, and Naive Bayes.