Data mining is a method to refine raw data to useful information. In education, data mining is a significant research part used to progress the value of education by observing students’ performance and understanding their learning patterns. Real-time student feedback would empower faculty and students to comprehend the teaching and learning problems in the most user-friendly way for the students. This paper uses a Lexicon based sentimental analysis technique to analyze students’ feelings and emotions through their feedback by correlating learning analytics to grounded theory. The sentiment analysis technique is a computational process to identify and classify subjective information such as positive, negative, and neutral from the source material. It can extract feelings and emotions from a piece of a sentence. Hence this paper aims to recognize the students’ positive or negative feelings and distinguished emotions, towards online teaching. The methodology undertakes four processes. The first process is data extraction from the feedback collected from the students through open-ended questions (Text) and is used as source material and imported to R studio. The second process is data cleaning /data preprocessing, removal of annoying data, and separation of data. The third process is sentimental analysis, which divides the data into positive, negative, and neutral categories/groups. This lexicon-based method of sentimental analysis is used to classify the sentiments. The results were estimated using sentiment scores and emotional variance. The sentiment scores result found that students have positive sentiments/emotions towards online teaching and emotions vary concerning the online class timing.