Spectral Pattern Learning Approach‐based Student Sentiment Analysis Using Dense‐net Multi Perception Neural Network in E‐learning Environment

E‐learning, a form of distance education in the realm of educational services, has witnessed significant growth, particularly with the pervasive influence of internet technologies especially during the pandemic, the demand for distance learning support has become notably intensive. Analyzing the interest factor is crucial for assessing student learning capabilities, given the distinction between distance learning and traditional method. Evaluating performance through sentiment analysis becomes essential in the context of distance learning. The main problem is understanding the student capability, and its performance through feedback analysis has uncertainty in sentiment term evaluation. Because the non‐dependability term of text evolution in feature extraction leads to low learning capability to predict the result. To resolve this problem, we propose Spectral Pattern learning approach (SPLA) based student sentiment analysis using Dense‐net multi perception neural network (DMPNN) for identifying student performance in E‐learning Environment. Initially, the preprocessing was carried out and find the sentiment terms through Word vector correlation extraction (WVCE) based on predominant defined words. Then the sequence pattern of word relation is extracted using Spectral pattern learning (SPL). Depending on the feature weights, Spider genetic sequence feature elevation (SGSFE) is used to predict the features terms and trained to classify the student learning capabilities using Dense net multi‐perception neural network (DMPNN). The proposed system produces high feature evaluation based on sentiment extraction terms to produce high sensitivity and precision rate to produce best result compared to the other system.