International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 11 Issue: 03 | Mar 2024
p-ISSN: 2395-0072
www.irjet.net
SENTIMENT ANALYSIS OF COLLEGES IN INDIA Venkata Siddarth Gullipalli1, Milan Kumar Dholey 2 1 Department Of Computer Science , GITAM University,
Visakhapatnam, India
2 Department Of Computer Science ,GITAM University,
Visakhapatnam, India ---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - This research on Sentimental Analysis of Colleges
Sentiment analysis refers to the computational process of extracting sentiment and opinions from textual data [7]. It allows researchers to categorize reviews as positive, negative, or neutral, providing a quantitative measure of student sentiment towards different aspects of college life [8]. By employing sentiment analysis, educators and college administrators can gain valuable insights into student concerns and tailor their offerings to better meet student needs [9].
in India explores the application of sentiment analysis to understand student perceptions of colleges in India, focusing on reviews found online. Analyzing these reviews can provide valuable insights into student satisfaction with various aspects of college life, such as placements, infrastructure, and faculty interaction. However, the unstructured nature of textual reviews presents challenges. To address this, we developed a sentiment analysis system utilizing a fine-tuned BERT model. This model is pre-trained on a massive dataset of text and code, allowing it to effectively capture semantic relationships within student reviews. Our system goes beyond basic positive/negative sentiment by classifying reviews into categories like placement, campus life, and academics, quality of education, research collaboration, outreach program, collaboration with different countries, examination pattern. This allows for a more nuanced understanding of student concerns and areas for improvement within colleges. The finetuned BERT model achieved an accuracy of 91% in sentiment classification, demonstrating its effectiveness in analyzing student reviews from an Indian context.
However, analyzing student reviews presents challenges due to the unstructured nature of textual data [10]. Traditional sentiment analysis methods often rely on lexicon-based approaches, which involve matching words in the review with pre-defined lists of positive and negative sentiment words [11]. While effective for basic sentiment classification, these methods struggle to capture the nuances of language and the context of student reviews [12]. Recent advancements in deep learning have led to the development of more sophisticated techniques for sentiment analysis, such as Bidirectional Encoder Representations from Transformers (BERT) models [13, 14]. BERT models are pre-trained on massive datasets of text and code, allowing them to capture complex semantic relationships within language [15]. This ability to understand context makes BERT models particularly well-suited for sentiment analysis tasks involving student reviews, which often contain complex language and implicit sentiment [16].
Key Words: Sentiment Analysis, Fine-Tuned BERT Algorithm, Colleges in India, Data Driven Ranking, College Selection, Student Experience
1.INTRODUCTION Higher education plays a pivotal role in shaping individual careers and contributing to a nation's overall development [1]. In today's competitive landscape, choosing the right college is crucial for students seeking a fulfilling academic experience and successful future careers [2]. Traditionally, students relied on factors like college rankings, faculty reputation, and program offerings to make informed decisions [3]. However, with the rise of the internet, a new source of information has emerged – online student reviews.
This research explores the application of a fine-tuned BERT model for sentiment analysis of online student reviews in the Indian context. Our primary objective is to develop a robust system that can accurately classify student sentiment towards various aspects of college life in India. Additionally, we aim to contribute to the growing body of research on sentiment analysis in higher education by providing insights into student perceptions specific to the Indian educational landscape.
These online reviews provide valuable, real-world insights into student experiences at various colleges [4]. They offer a glimpse into student satisfaction with various aspects of college life, including academics, campus infrastructure, placements, and faculty interaction [5]. Analysing these reviews using sentiment analysis techniques allows colleges to understand student perceptions and identify areas for improvement [6].
© 2024, IRJET
|
Impact Factor value: 8.226
2. RELATED WORK Several researchers have explored sentiment analysis techniques to understand student perceptions from online college reviews. Al-Harbi et al. (2017) analyzed reviews on a university website using both lexicon-based methods and machine learning approaches. Their Support Vector Machine
|
ISO 9001:2008 Certified Journal
|
Page 1040