Opinion mining from social media using Fuzzy Inference System (FIS)
| Autor: | K. Nivetha, P. Ajitha, G. Ragavi Ram |
|---|---|
| Rok vydání: | 2016 |
| Předmět: |
Adaptive neuro fuzzy inference system
business.industry Computer science Fuzzy set Sentiment analysis 020207 software engineering 02 engineering and technology Machine learning computer.software_genre Fuzzy logic Naive Bayes classifier Bag-of-words model 0202 electrical engineering electronic engineering information engineering Feature (machine learning) 020201 artificial intelligence & image processing Artificial intelligence business Set (psychology) computer |
| Zdroj: | 2016 International Conference on Communication and Signal Processing (ICCSP). |
| DOI: | 10.1109/iccsp.2016.7754566 |
| Popis: | The Emotion recognition of the speaker can impact in the commercial sector to know the valuable feedback. Though many systems have appeared to perform similar task, they do not provide feasible solutions for various parameters, especially have lesser accuracy. Thus an integrated Fuzzy Inference System (FIS) with naive Bayes classification that provided crisp outputs with greater accuracy. Fuzzy set theory is applied over the selected feature input and maps to the classified output. The rules are formulated to make suitable judgments on whether the uttered text is a negative or non-negative class of emotion. The defined set of database helps to correlate and identify similar texts. With this the opinions of the group of people can be interpreted. The bag of words are predicted with the help of wordlist database. Experimental analysis is made for comparing the existing hybrid machine learning approach and the proposed algorithm. The results showed that the proposed system has greater accuracy and improved recognition ratio. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|