Generative AI (GAI) technologies like ChatGPT are transforming the educational landscape. Their integration presents significant opportunities for personalized learning and enhanced student engagement but also raises challenges related to academic integrity and the role of human educators. This study aims to provide a comprehensive review on using multiple GAI tools and their influence on academic outcomes, bridging the gap in the current literature. A systematic literature review has been conducted, adhering to PRISMA guidelines, to discuss findings on the opportunities and challenges of GAI in education.This review includes theoretical and empirical studies employing qualitative, quantitative, and mixed-methods approaches. We have also examined conceptual frameworks and innovative AI applications, focusing on originality and ease of implementation. To gather insights into students’ experiences and perceptions regarding the use of Generative AI (GAI) in education, a survey has been conducted with 200 undergraduate university students. GAI offers opportunities for personalized learning, task automation, educator support, efficiency enhancement, innovative approaches, and increased student engagement. However, challenges include assessment integrity concerns, potential overshadowing of educational values, accuracy of AI-generated content, disruption of critical thinking skills, and ethical and privacy considerations. The student perception survey reveals that most students find AI systems helpful in assisting with academic work. Besides, they are aware of the challenges and limitations associated with the technology.

Background: As a direct result of the Internet's expansion, the quantity of information shared by Internet users across its numerous platforms has increased. Sentiment analysis functions at a higher level when there are more available perspectives and opinions. However, the lack of labeled data significantly complicates sentiment analysis utilizing Bangla natural language processing (NLP). In recent years, nevertheless, due to the development of more effective deep learning models, Bangla sentiment analysis has improved significantly. Objective: This article presents a curated dataset for Bangla e-commerce sentiment analysis obtained solely from the "Daraz" platform. We aim to conduct sentiment analysis in Bangla for binary and understudied multiclass classification tasks. Methods: Transfer learning (LSTM, GRU) and Transformers (Bangla-BERT) approaches are compared for their effectiveness on our dataset. To enhance the overall performance of the models, we fine-tuned them. Results: The accuracy of Bangla-BERT was highest for both binary and multiclass sentiment classification tasks, with 94.5% accuracy for binary classification and 88.78% accuracy for multiclass sentiment classification. Conclusion: Our proposed method performs noticeably better classifying multiclass sentiments in Bangla than previous deep learning techniques. Keywords: Bangla-BERT, Deep Learning, E-commerce, NLP, Sentiment Analysis

Aims: A smart greenhouse system that utilizes the Blynk IoT app is an innovative technology designed to automate greenhouse operations, thereby reducing labor and resource costs and optimizing plant growth. Study Design: In this paper we propose a real-time mobile app-based monitoring system to automate greenhouse agriculture. The system involves the usage of Blynk mobile application that allows users to monitor, control both manually and remotely as well as automation of various aspects of the greenhouse environment in real-time. Place and Duration of Study: Department of Computer Science and Engineering, American International University-Bangladesh (AIUB), Dhaka, Bangladesh between February 2023 to April 2023. Methodology: Multiple environmental sensors, temperature, humidity, soil moisture, light intensity, and Light Dependent Resistor (LDR) sensors continuously collect data. Farmers can remotely monitor and control the greenhouse using the Blynk IoT application. A fire sensor alarms it. Farmers can tailor alerts for environmental changes like irrigation or heating. They can manually and remotely control fans, lights, and humidifiers. If temperature, light intensity, or humidity drop below a specified threshold, the system turns on fans, lights, and humidifiers. A water pump that kicks on automatically based on soil moisture and can be regulated remotely ensures the plants get enough water. Results: The proposed system allows for real-time updates of sensor data and enables remote control of connected devices via the mobile app from any location, with notable fast processing subject to the quality of the internet connection. Conclusion: Upon successful implementation of this research, we anticipate the emergence of an advanced automated smart greenhouse monitoring system that will greatly benefit farmers in Bangladesh. This new method could change greenhouse and farm management, improving efficiency, cost, and crop yields.

Sentiment analysis, the process of determining the emotional tone of a text, is essential for comprehending user opinions and preferences. Unfortunately, the majority of research on sentiment analysis has focused on reviews written in English, leaving a void in the study of reviews written in other languages. This research focuses on the understudied topic of sentiment analysis of Bangla-language product reviews. The objective of this study is to compare the performance of machine learning models for binary and multiclass sentiment classification in the Bangla language in order to gain a deeper understanding of user sentiments regarding e-commerce product reviews. Creating a dataset of approximately one thousand Bangla product reviews from the e-commerce website 'Daraz', we classified sentiments using a variety of machine learning algorithms and natural language processing (NLP) feature extraction techniques such as TF-IDF, Count Vectorizer with N-gram methods. The overall performance of machine learning models for multiclass sentiment classification was lower than binary class sentiment classification. In multiclass sentiment classification, Logistic Regression with bigram count vectorizer achieved the maximum accuracy of 82.64%, while Random Forest with unigram TF-IDF vectorizer achieved the highest accuracy of 94.44%. Our proposed system outperforms previous multiclass sentiment classification techniques by a fine margin.

Generative artificial intelligence (GAI) systems like ChatGPT have gained popularity due to their ability to generate human-like text. Educators are exploring how to leverage these systems to facilitate and promote learning and develop skills and abilities. This study proposes a conceptual framework aimed at facilitating outcome-based learning through the utilization of GAI tools. The overall aim of this study is to provide a way for integrating GAI to support outcome-based education paradigms focused on learning objectives by aiding cognitive ability development according to Bloom's taxonomy. This paper introduces a framework called the ACE Framework (AI-Enhanced Cognition for Outcome-Based Learning), which organizes the integration of emerging large language models to facilitate advanced analysis, synthesis, and evaluation, as defined by Bloom's taxonomy, from basic knowledge recall to complex conceptualization. To empirically assess the effectiveness of unaided and GAI-assisted approaches, an analysis of real-world scenarios was conducted, where 20 college students created open-ended written solutions. For every response set, human raters classified shown cognitive abilities into Bloom's levels. In structured GAI integration exercises, participants learnt about problems using models such as GPT-4 and framed analytical answers. Comparative benchmarking reveals significant enhancements in average ratings from predominant comprehension (3.35) to top-tier synthesis (4.85) after AI scaffolding based on the methodology. With six students reaching the highest evaluation tier, guided AI interactions showcase excellent ability to promote outcome-based learning. Despite limitations in sample size and assessment techniques requiring further investigation, results align with priorities of outcome-driven education models prioritizing higher-order cognition -- substantiating structured AI incorporation potential.

The utilization of sentiment analysis has gained significant importance as a valuable method for obtaining meaningful insights from textual data. The research progress in languages such as English and Chinese has been notable. However, there is a noticeable dearth of attention towards creating tools for sentiment analysis in the Bangla language. Currently, datasets are limited for Bangla sentiment analysis, especially balanced datasets capturing both binary and multiclass sentiment for e-commerce applications. This paper introduces a new sentiment analysis dataset from the popular Bangladeshi e-commerce site “Daraz”. The dataset contains 1000 reviews across 5 product categories, with both binary (positive/negative) and multiclass (very positive, positive, negative, very negative) sentiment labels manually annotated by native Bangla speakers. Reviews were collected using an organized process, and labels were assigned based on standardized criteria to ensure accuracy. In addition, a benchmark evaluation of the performance achieved by Machine Learning and Deep Learning algorithms on this dataset is also provided. The new dataset can aid research on multiclass and binary Bangla sentiment analysis utilizing both machine learning, deep learning, and Large Language Models. It can aid e-commerce platforms in analysing nuanced user opinions and emotions from online reviews. The utilization of categorized product reviews also facilitates research in the field of text categorization.