To search, Click below search items.


All Published Papers Search Service


Applying IoT for Sentiment Classification and Tone Analysis of Urdu Tweets


Shakeel Ahmad, Yasser D. Al-Otaibi


Vol. 19  No. 11  pp. 166-173


Social media sites such as Facebook, Twitter and Flicker are becoming integral parts of social network users. The sentiment classification and tone analysis are two important applications of social media analytics. Most of the existing studies pertaining to the two aforementioned tasks are based on the machine learning approaches. However, a more recent trend is to take advantage of the Internet-of-Things (IoT) based platforms for efficient classification of users’ sentiments and tones (emotions), which could be of a great assistance to the business community aiming to monitor and satisfy the needs their customers. The IBM Watson Cloud equipped with Tone Analyser and Natural Language Understating services could provide a best avenue for the sentiment classification and tone analysis tasks. Some works are performed on IoT-based sentiment analysis tasks. However, there is lack of such applications for sentiment analysis in Urdu language, which is widely spoken in major parts of the Indian sub-continent. In this work, we propose an IoT-based Urdu sentiment classification and tone analysis framework, capable of crawling Urdu tweets from Twitter in real time and then performing sentiment classification and tone analysis using IBM Watson Cloud services. The results obtained show the efficacy of the proposed system with respect to baseline methods.


Twitter, IoT, Sentiment Classification, Tone Analysis, Emotion Classification, Urdu Sentiment Analysis, IBM Watson