Browsing by Author "Rabeya, Tapasy"

Now showing 1 - 6 of 6

A Machine Learning Approach for Sentiment Analysis of Book Reviews in Bangla Language
(Daffodil International University, 2022-05-30) Khatun, Mst. Eshita; Rabeya, Tapasy
With the advent of technology, Sentiment polarity detection has recently piqued the interest of NLP researchers. Sentiment analysis determines the profound meaning of an article. Due to COVID-19 pandemic, online shopping is the safest way of shopping. Moreover, there are product quality and service issues. Our target is to analyze the book reviews which provide positive and negative reviews in Bangla language. For this, a total of 5500 user generated Bengali reviews are collected from various book review pages of social media. In order to get the best possible result, sentiment analysis is used. Thereafter, five different algorithms are applied to predict with almost high accuracy. Among them, the Random Forest provides us the maximum accuracy which is 98.39%.
A survey on emotion detection: A lexicon based backtracking approach for detecting emotion from Bengali text
(IEEE, 2018-02-08) Rabeya, Tapasy; Ferdous, Sanjida; Ali, Himel Suhita; Chakraborty, Narayan Ranjan
Emotion recognition ability has been introduced as a core component of emotional competence. Every emotion has different ways to be expressed such as text, speech, lyrics etc. This paper reflects the current experimental study and their outcomes on emotion detection from different textual data. In case of lexicon-based analysis, the position of emotional lexicons really varies the state of an emotion. In this empirical study, our focus was to find how people use the emotional keywords to express their emotions. We have presented an emotion detection model to extract emotion from Bengali text at the sentence level. In order to detect emotion from Bengali text, we have considered two basic emotion `happiness' and `sadness'. Our proposed model detects emotion on the basis of the sentiment of each sentence associated with it. A lexicon based backtracking approach has been introduced for recognizing the sentiments of sentences to show how frequently people express their emotion in the last part of a sentence. Proposed method can produce a result with 77.16 accuracies.
Advancing Agricultural Diagnostics with a Hybrid Deep Learning Model for Sugarcane Leaf Disease Classification
(Scopus, 2024) Nirob, Md. Asraful Sharker; Deeya, Ikteshad Binte Kalam; Rukhsara, Lamia; Rabeya, Tapasy; Jahan, Israt
Sugarcane, an essential crop for sugar production and biofuel, is susceptible to diseases that greatly affect its yield and quality. To address this, we propose a hybrid deep learning model combining ResNet152 and MobileNetV3Large to accurately classify sugarcane leaf diseases. Our method involves careful dataset preparation and data augmentation to enhance model robustness. The hybrid model leverages the strengths of fine-tuned pre-trained networks with custom layers, achieving an impressive accuracy of 98.19%, outperforming other models such as DenseNet121 (76.09%) and NASNetLarge (82.65%). The dataset includes images from ten sugarcane leaf disease classes, including healthy leaves. Thorough metrics, such as recall, precision, and F1-score, along with visualizations of training and validation results, emphasize the model's effectiveness. This hybrid model shows significant potential for agricultural diagnostics. Future work will focus on fine-tuning and exploring ensemble methods to improve accuracy and generalization across various agricultural environments
Bengali Review Analysis for Predicting Popular Cosmetic Brand Using Machine Learning Classifiers
(Springer Nature Limited, 2022-11-14) Rabeya, Tapasy; Khatun, Eshita; Noori, Sheak Rashed Haider; Akter, Sharmin; Jahan, Israt
Nowadays, online platform has become one of the most popular media to express people’s thought of all ages. That made the online platform a precious source for getting almost every kinds of information. As online shopping is rising in no time in recent years, as a result millions of comments are generating every single day. These users generated opinions on social media and different websites has made it easier for the people choosing the right product for them. Hence, sentimental analysis is a sought-after research topic nowadays. Our research paper has portrayed an experimental study on different cosmetics products review. To do so, we have selected ten popular cosmetic brands for analyzing their product review and chosen to analyze Bengali comments or sentences. The main focus of our work was to get out the most popular cosmetic brands among ten chosen brands. We have applied four classification algorithm such as naive Bayes, random forest, decision tree, and support vector machine for analyzing the final outcome and found vaseline and clear are the most popular brands.
Sentiment Analysis of Bangla Song Review
(Proceedings of 2019 3rd IEEE International Conference on Electrical, Computer and Communication Technologies, ICECCT 2021, IEEE, 2019-10-17) Rabeya, Tapasy; Chakraborty, Narayan Ranjan; Ferdous, Sanjida; Dash, Manoranjan; Marouf, Ahmed Al
In the last decade, Sentiment analysis is a prospering experimental topic of research as because of a lot of opinionated data accessible on Blogs & social networking sites. It is the reference to the assignment of Natural Language Processing to decide whether text or content contains any subjective information like positive, negative or not. These social media and other online platforms are giving an immense stage to uncover human's gifts in a fast speed, and average citizens can likewise put their feeling through the remarks which emphatically show how they are accepting another potential. In this paper, we have presented a sentimental analysis of Bengali song reviews from a specific YouTube channel to analyse people acceptance rate of a new young star. For detecting the sentiments, we have used a backtracking algorithm, where the heart of this approach is a sentiment lexicon. And the research showed the backtracking algorithm performed more than 70% accuracy to detect actual public sentiment.
Sentiment Analysis of Products Review
(Daffodil International University, 2022-02-17) Rabeya, Tapasy
E-commerce has become one of the most commodious methods of shopping because of the technological revolution. Research shows internet shopping is much preferred by people rather than the traditional mode of shopping. Millions of user-generated comments are posted daily on the web, and analysis of these opinions could be more directive towards the customer’s and manufacturers. That makes the Sentiment analysis of online reviews one of the most sought-after research topic. This paper portrays our experimental work on domain-specific feature-based sentiment analysis of product review. In this paper, we worked with some fixed predefined core features of a product for presenting the customer’s acceptance of the principal attributes of a product so that the manufacturer can improve the basic features quality. We have proposed a feature-oriented sentiment prediction scheme. That analyses the generated expressions from the textual reviews of a product for predicting sentiment and assigns scores for our predefined features to present a net sentiment profile of a product of all parameters. With 92% accuracy our sentiment detection scheme is proved to be an effective ways to highlight the core attributes that are seems to be the most to the purpose to the customer and manufacturer.