Rule Based Approach for Contextual Classification of Twitter Dataset.
Mr. L.K. Ahire, S. D. Babar, Dr. P. N. Mahalle. · International Journal of Advanced Networking and Applications · 2024
The amount of data on social networks and the number of users has been growing quickly in recent years.Any time an event or activity occurs nearby, nearby individuals express their thoughts and reactions on social media.When a new product is introduced, users on social media platforms also comment on it.It is challenging to ascertain the genuine state of emotions because of sophisticated ways of presenting various perspectives.Sarcasm is the use of words to convey a negative emotion in a humorous way.Machines have an extremely difficult time comprehending and recognizing these caustic remarks when trying to discern sarcasm from text, it helps to know the context of the content.In this research, we propose a novel method called the Rule Based Approach for Contextual Classification (RBACC), which uses the context of tweets to identify sarcasm using a variety of already available methods.RBACC uses four features that were taken from tweets that were acquired using the tweeter API, and rule-based evaluation is done using the linguistic data of the four characteristics.The RBACC technique ensures flexibility and energy efficiency, according to experimentation.The RBACC technique is also scalable because performance and functionality are unaffected by an increase in the quantity of tweets.Results demonstrate that RBACC accurately identifies the context of text when given a variety of datasets that contain both type of data balanced as well as imbalanced.