International Journal of INTELLIGENT SYSTEMS AND APPLICATIONS IN ENGINEERING ISSN:2147-67992147-6799 www.ijisae.org Original Research Paper Internaonal Journal of Intelligent Systems and Applicaons in Engineering IJISAE, 2024, 12(4), 2260–2271 | 2260 Qubits and Sentiments: Unveiling New Perspectives in Hindi Textual Data Vaibhav Prakash Vasani 1 * Asha Ambhaikar 2 Submitted:10/03/2024 Revised: 25/04/2024 Accepted: 02/05/2024 Abstract. Sentiment analysis (SA) is a critical component of Natural Language Processing (NLP), particularly for automatic text classification. However, previous approaches have shortcomings in capturing nuances such as negation, word pairings, and contextual understanding, particularly in languages like Hindi. To solve these issues, this research offers a novel Quantum Neural Network (QNN) technique designed specifically for sentiment analysis in Hindi text data, which employs quantum computing concepts to capture linguistic nuances and context better. This study uses a Quantum Variational Auto Encoder to encode classical data into a quantum form, capturing diverse sentiments like sarcasm and colloquial expressions. A Tuned Quantum Convolutional Neural Network architecture is introduced to capture complex linguistic syntax. A novel Sequential-based hyperband optimization technique is used to enhance model performance. The hybrid approach significantly improves accuracy and efficiency in handling quantum data, contributing to the advancement of SA, particularly in Hindi Movie Reviews. The findings demonstrate that the proposed strategy performs best with accuracy 97.64 %, precision 85.93%, recall 99.17%, F-1 score 92.20%, than other accepted strategies. Keywords. Quantum Neural Network (QNN), Quantum Variational Autoencoder (Q-VAE), Tuned Quantum CNN, Sequential-based hyperband optimization technique, Hindi movie reviews. 1. Introduction: SA, a prominent branch of NLP, involves identifying and extracting sentiments or opinions expressed in textual data. SA has become essential for understanding public perception, customer feedback, and social trends as digital content on social media, product reviews, and online forums has exploded [1]. This process involves determining whether a text conveys a positive, negative, or neutral sentiment. In recent years, [2] machine learning (ML) and deep learning (DL) techniques have significantly advanced sentiment analysis methodologies. ML approaches typically employ algorithms that learn patterns and relationships in labelled training data to predict unseen text. DL, particularly through neural networks, has gained prominence for its ability to capture intricate features and hierarchical representations in text data, resulting in more nuanced sentiment predictions [3], [4]. Extending sentiment analysis to languages such as Hindi reflects the global need to comprehend various linguistic contexts [5]. Due to the rich morphology, varying sentence structures, and the presence of code-switching, analyzing sentiment in Hindi text presents unique challenges [6], [7]. Creating sentiment analysis models for Hindi entails adapting and fine-tuning existing approaches to account for the complexities of the language. ML-based sentiment analysis frequently employs techniques such as Support Vector Machines (SVM), Naive Bayes, and Random Forests [8]-[11]. These models predict sentiment classes by extracting features from text data such as word frequencies or n-grams. While these approaches are effective, they may struggle to capture semantic nuances and context-dependent sentiments [12]. DL models, particularly recurrent neural networks (RNNs) and transformer architectures such as BERT (Bidirectional Encoder Representations from Transformers), have demonstrated outstanding performance in SA tasks [13], [14]. These models capture dependent context and long term dependencies, improving their ability to detect subtle emotions. DL methods, on the other hand, may necessitate large amounts of labelled data and computational resources. Researchers have investigated techniques such as machine translation for preprocessing, domain-specific lexicons, and sentiment lexicons tailored for the language in the context of sentiment analysis in Hindi. Despite these advances, challenges remain, such as the scarcity of labelled datasets for Hindi sentiment analysis and the need for additional research to address linguistic variations [15]. The limitations of different approaches in predicting sentiment classes include potential bias in training data, difficulties in dealing with sarcasm or irony, and the model's sensitivity to context changes [16]. Furthermore, when _______________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________ 1* Department of Computer Science and Engineering Kalinga University, Kotni, Near Mantralaya, Naya Raipur Chhattisgarh – 492101 https://orcid.org/0000-0001-6498-553X *Corresponding Author Email: vaibhav.prakash@kalingauniversity.ac.in 2 Department of Computer Science and Engineering Kalinga University, Kotni, Near Mantralaya, Naya Raipur Chhattisgarh – 492101 https://orcid.org/0000-0002-6814-5949 Co author Email: asha.ambhaikar@kalingauniversity.ac.in