The COVID-19 pandemic has ushered in an era of unprecedented challenges for students at various educational levels, from schools to colleges. This study delves into the pervasive issue of academic stress experienced by students during the pandemic and employs a data-driven approach to examine its underlying causes. Leveraging Python-based data analysis techniques, we apply a suite of classification algorithms, including K-Nearest Neighbours (K-NN), Decision Tree, Support Vector Machine (SVM), Random Forest, and Logistic Regression, to pre-process and analyse the data. As the pandemic prompted a shift towards online education in India, we sought to understand its ramifications on student mental health. Our research unfolds a notable finding: a significant upsurge in mental health issues among students, inextricably linked with elevated stress levels. To gauge stress levels and classify data effectively, we employed diverse classification methods, accompanied by accuracy assessment algorithms. Emerging from our comprehensive analysis is a compelling insight: the primary driver of academic stress during the pandemic is the palpable lack of effective communication between students and their educators. This finding underscores the pivotal role that communication channels play in mitigating stress and bolstering student well-being. In conclusion, this research paper illuminates the multifaceted impact of the COVID-19 pandemic on students' mental health, particularly in the context of academic stress. By harnessing machine learning algorithms, author provides a nuanced understanding of the stressors faced by students, offering valuable insights to educational institutions, policymakers, and mental health professionals. The imperative for improved communication between students and teachers is a salient takeaway, emphasizing the need for proactive interventions to support students' emotional well-being during these challenging times.