“…Researchers have also found that various factors such as cost performance (Chen and Xu, 2016), customer's sentiment polarity (Geetha et al, 2017), social influence bias and ratings bubbles (Aral, 2014) affect the overall ratings of products or services. Researchers have also used different techniques like, tensor factorization method (Chambua et al, 2018), latent factor models (Gu et al, 2020), convolution neural networks (Khan and Niu, 2021), etc., to predict ratings from textual reviews. Since online reviews and ratings serve as great information sources (Engler et al, 2015) for users as well as providers (Chatterjee, 2019), determining ratings from SM posts can be really useful from the business perspective.…”