An Review of Student Sentimental Analysis For Educational Database Using Unsupervised Machine Learning Approaches
European Journal of Molecular & Clinical Medicine,
2020, Volume 7, Issue 9, Pages 2151-2165
Abstract
With the World Wide Web's growth at a relatively high rate, a huge increase in network communications has become inevitable. Online communication data includes feedback posted by students. Sentiment analysis, natural language processing, computational linguistics and text analysis, application recognition and text retrieval of information.Social media platforms are considered to be the most popular form of online communication. Through platforms such as the e-learning education system, multiple pieces of information that reflect student opinions and attitudes are published and shared among users every day. These platforms have recently focused on tracking and monitoring the reputation of their study time and student review and helping decision-makers and politicians assess their public opinion on policy and political issues. The opinions and emojis of understanding its important role in influencing student study opinion decisions are becoming more important for the educational field. It is possible to consider how emotions are used to advance e-learning intelligence tasks so that organizations can gain access to analyze education and detect unfavorable rumor risk management capabilities. Machine Learning is widely used for reputation analysis, and many sentiment analysis systems use unsupervised learning methods. This existing sentimental analysis for dictionary-based approaches is to provide a higher time complexity. The words are utilizing unsupervised learning based lexicon pattern methodology. The educational, sentimental pattern dependency factor between aspects and the sentiment word isconsidered. This approach evaluates the sentimental information to find the aspect and provides less complexity and a higher classification rate than existing methods.
Keywords:
- Article View: 346
- PDF Download: 901