10
Md Monir Ahammod Bin Atique, Hyeon-Ah Moon, Isao Sasano, and Kwanghoon Choi
Accepted in Journal of Computer Languages, Elsevier (2025)
Abstract: Code completion is a crucial feature in modern IDEs, improving programming efficiency. Traditional systems rely on prefix filtering and static ranking but often overwhelm users with lengthy, alphabetically sorted lists. Recent research has introduced LR-parsing-based approaches that derive completion candidates from language syntax and compute their ranks using open- source programs; however, these methods only suggest structural candidates, requiring manual refinement into complete code. To address this, we propose a hybrid method that integrates LR parsing with LLMs to enhance accuracy and usability. Our approach refines structural candidates using LR parsing into textual code suggestions via an LLM, referencing a database of ranked candidates from open-source programs. This combines the syntactic precision of LR parsing with the generative capabilities of LLMs. This study examines whether LLMs benefit from LR structural candidates in code completion. By comparing completions with and without these candidates, we assess their impact. Building on prior research, we also explore how leveraging top-ranked structural candidates can effectively enhance LLM-based code completion precision. We also demonstrate our method through VSCode extensions for Microsoft Small Basic and C. As a language-agnostic solution, our system applies to any language with a defined LR grammar. Our findings suggest that integrating LR parsing with LLM-based completion improves both accuracy and usability, paving the way for more effective code completion in modern IDEs.
9
Md. Monir Ahammod Bin Atique and Kwanghoon Choi
Under Review in Annual Symposium of KIPS 2025 (ASK 2025), Kyungpook National University Daegu Campus, Daegu, Korea
Abstract: Modern integrated development environments (IDEs) rely on code completion as a key feature to enhance coding efficiency and streamline the developer workflow. Traditional approaches to code completion have often relied on rule-based techniques, static ranking, and prefix-based filtering, which pose challenges in terms of usability and efficiency. Recent research has introduced LR-parsing-based approaches that generate structural candidate suggestions by leveraging language syntax and open-source programs, but they often require manual refinement. Meanwhile, recent advancements in large language models (LLMs) have significantly amplified predictive performance in code completion tasks. However, despite these progress, the impact of LLM selection on LR-parsing based syntax-aware code generation remains underexplored. In this study, we conduct a comparative analysis of the performance and impact of two prominent LLMs, ChatGPT 3.5 and Llama 3, within an LR parsing based code completion framework. Our experiments, evaluated using SacreBLEU and SequenceMatcher accuracy metrics, reveal that ChatGPT 3.5 achieves higher accuracy than Llama 3, underscoring the importance of selecting an appropriate LLM for enhanced code completion. These findings highlight the role of model selection in LLM based code completion using LR parsing. Future research could extend this comparative analysis to a broader range of LLMs.
8
Md. Monir Ahammod Bin Atique and Kwanghoon Choi
Accepted in 2025 Korea Smart Media Society & Korea Electronic Transactions Society Spring Conference, Chung-Ang University, Seoul, Korea
Abstract: Code completion is a key feature in modern integrated development environments (IDEs), aiding developers in improving coding efficiency. While traditional methods rely on prefix-based filtering and static ranking, LR-parsing-based approaches offer structural candidates from language syntax as code suggestions to the user, but are limited by manual refinement. Recently, large language models (LLMs) have enhanced code completion without explicit knowledge of language grammar in the given prompt. This study investigates whether incorporating LR grammar-based structural candidates into LLM-driven code completion improves performance. Experiments using SacreBLEU and SequenceMatcher on Microsoft Small Basic and C show no significant (accuracy) gain, suggesting LLMs may already internalize grammar or need alternative evaluation.
7
MD Ilias Bappi, Md. Monir Ahammod Bin Atique , Kyungbaek Kim
Published in 2025 International Conference on Artificial Intelligence in Information and Communication (ICAIIC), Fukuoka, Japan (Publisher: IEEE)
Abstract: Diabetic Retinopathy (DR) is a major cause of vision loss and blindness, particularly among diabetic patients. Effective and timely treatment of DR relies on precise and automated detection systems that can assess disease severity from retinal fundus images. Traditional clinical approaches are often time-consuming, and earlier texture attention models may struggle to accurately detect subtle features, such as microaneurysms or abnormal blood vessel patterns, which are crucial for early diagnosis. To address this, we proposed the STMFNet model, a hierarchical framework designed for classifying 11 stages of DR severity. The model combines two primary mechanisms: The Texture Spatial Attention Network, which focuses on identifying critical texture features related to DR while minimizing irrelevant background information through attention gating, and the EfficientNet backbone with Multi-Scale Feature Fusion, which captures a wide range of image patterns. By extracting and fusing features from different layers of EfficientNet-V1 BO, the model effectively learns both low (e.g., edges, blobs) and high (e.g., objects, patterns) level representations. These features are further refined through spatial multi-scale attention, and the final classification into 11 DR stages is achieved through a Fully Connected Network (FCN) and SoftMax. Our experimental results show that STMFNet significantly outperforms existing state-of-the-art models on the publicly available Kaggle fundus dataset, demonstrating its potential for reliable DR diagnosis in clinical settings.
6
Md. Monir Ahammod Bin Atique , Kwanghoon Choi, Isao Sasano, Hyeon-Ah Moon
Published on SCSS 2024: 10th International Symposium on Symbolic Computation in Software Science, August 28–30, 2024, Tokyo, Japan (CEUR Workshop Proceedings)
Abstract: Programmers often use syntax completion and code suggestion features. Our methodology enhances code completion by combining structural candidate information from LR parsing with LLMs. These structural candidates are utilized to compose prompts so that ChatGPT can predict actual code under the specified structure. Tested on Small Basic and C benchmarks, this approach offers textual suggestions rather than just structural ones, showing nearly 50% prediction accuracy for Small Basic programs. While effective for Small Basic, we report that challenges remain with C11 programs.
5
Md. Monir Ahammod Bin Atique , Md Ilias Bappi
Accepted in Journal of Computer Systems Science and Engineering, 2024
Abstract: An industry or firm cannot imagine without employees. It suffers considerably when an employee leaves the firm for personal or professional reasons. During this time, the company wastes additional time and resources on new recruitment and training processes. In addition, the company's ongoing tasks become increasingly difficult to complete on schedule. Therefore, when the voluntary attrition rate is high, the organization will experience significant financial and other difficulties. Under these situations, the Human Resource (HR) department's first priority is to reduce the turnover rate. Artificial intelligence (AI) is the most popular innovative technology that is used nowadays in business strategies, organizational aspects, and people management. From this perspective, further study has been conducted through the use of statistical analysis and various Machine Learning (ML) and Data Mining approaches, such as Extreme Gradient Boosting (XGBoost), Random Forest (RF), Naive Bayes (NB), Decision Trees (DT), etc. This paper has applied a cutting-edge boosting technique, Categorical Boosting (CatBoost), with an enhanced feature engineering process to detect and analyze employee turnover on the IBM dataset. To analyze the effectiveness of our detection system, we have also used our extended dataset to show our proposed method's efficiency. Experimental results show that our detection technology outperforms on both datasets than all other technologies. It reveals a highest accuracy of 90.10% and an F1-score of 0.88 on the IBM dataset. Additionally, this technique shows an optimal accuracy of 98.10% and an F1-score of 0.98 on the extended dataset, strongly indicating the effectiveness of our proposed methodology. Moreover, this work identifies the key causes of attrition.
4
Md. Monir Ahammod Bin Atique , Md Ilias Bappi, Kyungbeak Kim, Kwanghoon Choi, Md Martuza Ahamad, Khondaker Masfiq Reza
Published on medRxiv (Cold Spring Harbor Laboratory Press)
Abstract: The Covid-19 outbreak has adversely influenced university students across the world both physically and psychologically. The psychological struggle faced by students, is effected by various factors, including disruptions to daily routines and academic activities, increased reliance on smartphones and the internet, limited social interaction, and confinement to their homes. These impediments reflect a broader issue of imbalance in cognitive health status among them during the pandemic. In Bangladesh, despite having the necessary population to study, understanding the impact of Covid-19 on the mental health status of university students has been limited. Hence, it is imperative to diagnose mental health issues and deal with the underlying reasons in order to enhance students’ psychological well-being, which leads to academic achievement. Nowadays, Artificial Intelligence (AI) based prediction models have the potential to play a crucial role in predicting mental state early. The purpose of the study is to explore the following effects of the pandemic on the mental health of Bangladeshi university students using Machine Learning (ML) and Deep Learning (DL) techniques. A reliable AI prediction system requires real-world data, that was collected by a survey through a Google form (online questionnaires) among 400 students of 16 universities, and the respondents were 253. In this paper, after data preprocessing, ten widely known ML and four DL models were developed to automatically and accurately predict mental well-being during or after the Covid-19 circumstance. According to our findings, the Random Forest (RF) algorithm and Siamese Neural Networks (SNNs) outperformed other models in terms of accuracy (86% and 75%). Additionally, Chi-Square test was conducted, which revealed the five most common and significant predictors (“Stable family income”, “Disruption of daily life”, “Own income”, “Sleep status”, and “Fear of getting infected with Covid-19”) of psychological health conditions. Overall, this work could assist university administrations, government agencies, and health specialists in taking appropriate measures to understand and maintain students’ mental health. This research also suggests proper monitoring, government support, and social awareness during and after the worldwide epidemic to keep an excellent mental health state of university students.
3
Md. Monir Ahammod Bin Atique , Md. Nesarul Hoque, and Md. Jamal
Uddin
Published in the Machine Intelligence and Emerging Technologies (Conference Proceedings-SpringerLink)
[Scopus Indexed] (Certificate)
Abstract: Almost everywhere, organizations or individuals can adopt technologies that can be supportive to make decisions and get insight from data: artificial intelligence (AI) is an advanced new technology which is used to aid organizations in their business procedures, organizational factors, and human resources management. Employees are the nucleus of the organization. When employees leave an institution of their own volition, the company suffers greatly from various dimensions. Nowadays, we have seen a huge change in companies due to COVID-19; employees are getting fired or resigning voluntarily. It is a big issue to keep the productivity constant of a company or individual as human resources (HR) has to spend a lot of time, from the selection process to the training process. In these circumstances, minimizing the attrition rate is one of the primary concerns of the Human Resource department, which deals with staffing, development, and compensation. From this point of view, more research projects have been done through statistical analysis and applying various types of machine learning (ML) and data mining techniques such as Extreme Gradient Boosting, Random Forest, Naive Bayes, decision trees, etc. In this paper, a state-of-the-art boosting method, CatBoost, and a feature engineering process have been applied for detecting and analyzing employee attrition. Our detection system shows the utmost performance compared to the other existing systems and sorts out the significant reasons behind the attrition. It reveals the best recall rate of 0.89, with an accuracy of .8945.
Short View Paper Link2
Pranta Kumar Sarkar, Pallab Kumar Sarkar, and Md. Monir Ahammod Bin Atique
Published in the ICRPSET 2022 : 7th International IEEE Conference on Recent Progresses in Science, Engineering and Technology
Abstract: The smart grid is a technology that was created today to address the problem of maintaining significant quantities of energy consumption with the help of emerging nations. Smart cities, like the smart grid, have an energy infrastructure that is arguably the single most important feature in any city. Smart cities depend on a smart grid to ensure resilient delivery of energy to supply their many functions, including those responsible for public safety and the public. It consists of smart distribution boards and circuit breakers integrated with home control and demand response, where load control switches and smart appliances are located. The smart grid vision is supported by the use of information and communication technology. In this article, the capabilities and technologies of the smart grid are compared across several machine learning and deep learning technologies. The deep learning (ANN) methodology was used to find the best approaches for forecasting grid stability. Various supervised machine learning classifiers are utilized to estimate accuracy, AUC-ROC, and other metrics. According to the investigation, deep learning-based estimations were actually more accurate than supervised learning-based ones. As a result, accurate power consumption estimation makes sure the entire chain runs smoothly
Link Short View Paper1
Md. Monir Ahammod, Md. Martuza Ahamad, and Md. Jamal Uddin
Under Review in Education
and Information Technologies, Journal of Springer Nature
Abstract: Bangladeshi university students have been affected both physically and mentally by the outbreak of COVID-19. Our university students' mental health, on the other hand, is not given the same level of attention as their physical health. The purpose of the study is to explore the impacts of this pandemic on the mental health state of Bangladeshi university students. In this paper, after data preprocessing, several machine learning models are developed to automatically and accurately predict mental health status. We observed that the Random Forest algorithm performed the best accuracy (>85%) to predict mental health conditions during or after the COVID-19 circumstance. We also conducted statistical analyses, which revealed the most common and significant predictors (Stable family income, Disruption of daily life, Own income, Sleep status) of mental health conditions during Covid-19. Overall, this work could assist university administrations, government agencies, and health specialists in taking appropriate steps to enhance our students' mental health. This research also suggests proper monitoring, government support, and social awareness during the pandemic and after the pandemic to keep an excellent mental health state of university students.
Paper Short View Link
I am developing an on-premise AI-powered product recommendation system tailored for independent online store operators. While large marketplaces benefit from advanced AI models, smaller retailers often lack such capabilities. This project addresses that gap by building a system that analyzes customer input sentences using Natural Language Processing (NLP) and vector search to recommend the top 5 semantically relevant products. Designed for local deployment, the solution ensures data privacy and full control over the recommendation workflow.
This project introduces two new Visual Studio Code (VSCode) extensions designed to enhance code completion capabilities for Microsoft Small Basic and C, addressing the limitations of traditional and AI-based tools like Copilot. Unlike standard VSCode extensions that rely on user-typed prefixes or language-specific grammar assumptions, the proposed extensions provide syntax-structure-based code suggestions without requiring initial input, thereby aiding users unfamiliar with programming grammar. Built using the YAPB parser builder tool and LR grammars, the system uses WithinTop3Guide to present the top three context-aware suggestions. It also features an interactive preview of candidates—displaying both identifier names and decomposable expressions—to improve comprehension and usability for both novice and advanced users. The extensions aim to provide grammar-compliant, generative AI-driven code completions and are fully open-source on GitHub.
The Cardio Vas. Disease Detection project aims to develop a machine learning model in Python to predict the presence or absence of cardiovascular diseases (CVD) based on a set of input features. The project compares the performance of two popular tree-based algorithms, Random Forest and Decision Trees, for CVD prediction. These algorithms are widely used in the field of machine learning and offer different advantages and trade-offs. The project utilizes a dataset containing various clinical and demographic features of individuals, such as age, gender, blood pressure, cholesterol levels, and smoking habits. Each instance in the dataset is labeled as either having a cardiovascular disease or being disease-free.
The Cleveland Heart Disease dataset contains a wide range of patient attributes, including clinical, demographic, and physiological features such as age, gender, cholesterol levels, blood pressure, and electrocardiogram measurements. Each instance in the dataset is labeled as either having heart disease or being disease-free. In this project, various machine learning algorithms will be applied to the dataset, including but not limited to decision trees, random forest, support vector machines (SVM), logistic regression, and neural networks. These algorithms offer different strengths and weaknesses, and by evaluating their performance, we can identify the most effective approach for heart disease prediction.
The project utilizes NLP techniques to analyze the text content of social media posts and classify them as either cyberbullying or non-cyberbullying. NLP encompasses a range of methods and algorithms that enable computers to understand and process human language. By applying techniques such as text preprocessing, sentiment analysis, part-of-speech tagging, and machine learning, the project aims to extract meaningful features from text data and develop a robust model for cyberbullying detection. The development of the model involves several steps. Firstly, a comprehensive dataset of social media posts, labeled as either cyberbullying or non-cyberbullying, is collected and prepared for analysis. The dataset may include various forms of text data, such as tweets, comments, or forum posts, from different social media platforms. Next, the collected data is preprocessed to remove noise, handle punctuation, and transform the text into a format suitable for analysis. NLP techniques, such as tokenization, stemming, and
The Iris flower dataset consists of measurements of sepal length, sepal width, petal length, and petal width for three different species of Iris flowers: Setosa, Versicolor, and Virginica. The goal of the project is to develop a deep learning model that can analyze these feature measurements and accurately predict the corresponding Iris flower species. In this project, a deep learning model will be designed and trained using the Iris flower dataset. The model will be built using popular deep learning frameworks such as TensorFlow or PyTorch, which provide a wide range of tools and functionalities for constructing and training neural networks.
created with
Website Builder Software .