Browsing by Author "Shoumo, Syed Zamil Hasan"

Now showing 1 - 4 of 4

A machine learning approach to predict young voter enthusiasm based on non-political factors
(BRAC University, 2020-04) Rahman, Md. Nowroz Junaed; Pantho, Md. Humaun Kabir; Fuad, Nafis; Majumdar, Mahbubul Alam; Shoumo, Syed Zamil Hasan
The right to vote is considered to be the backbone of democracy. As we are entering the third decade of 21st century more and more countries around the world are adopting the democratic government system. One of most important element of democratic country is the power vested in the common people and one of the way the people are expected to exercise this power is to elect a qualified candidate to lead their country. The only way to make this election process effective is to make sure everybody participates in the process. The people who are eligible to participate in this election process to elect a candidate are called ”voters”. A substantial amount of these voters are young voter or voters who have newly been registered. It has been noticed that young voters in most of the countries are reluctant to participate in the voting process. There are many social, psychological and other non-political factors behind this reluctance. This research seeks to find those factors that motivates or repels a young voter to participate in the voting process. Besides finding the factors this research will also try to determine whether a young voter is likely to vote in an election or not based on those factors mentioned before. The data set of this research was prepared by surveying via Google Forms. Later the data set was analyzed to find out the reasons behind their participation. RFECV was used to select the optimum features and later Support Vector Machine, Random Forest, Extreme Gradient Boosting and Naive Bayes were used to predict voter participation based on the set of optimum features. In such an experimental setup Extreme Gradient Boosting and Support Vector Machine with a Gaussian kernel has shown more promising results than the other aforementioned models. The aim of this research is to predict whether a young voter will participate in the voting process or not and find the reasons behind it so, that maximum voter turnout can be ensured and perfect democracy can be achieved.
Application of machine learning in credit risk assessment: a prelude to smart banking
(BRAC University, 2018-12) Dhruba, Mir Ishrak Maheer; Ghani, Nawab Haider; Hossain, Sazzad; Shoumo, Syed Zamil Hasan; Arif, Hossain
A precise credit risk assessment system is vital to a financial institution for its proper and impeccable functioning. Accurate estimations of credit risk will allow them to continue their operation in a gainful and transparent way. As the rate of loan defaults are gradually increasing, bank authorities are finding it more and more difficult to correctly assess loan requests. Thus the subject of credit risk has become a highly conferred and examined topic throughout the world. Numerous solutions have been given, one being more efficient than the other and several studies are still being made for solving this difficult predicament. Thus keeping the implications of such a problematic matter in mind this paper proposes to build a machine learning model which can precisely assess credit risk and predict possible loan defaulters for any credit lending institution. Taking into account a borrower’s financial and social history this paper proposes a way to accurately define whether a customer’s loan request should be accepted or not which in turn can steadily save the creditor from incurring further loss. Evaluating data from previous successful borrowers and loan defaulters, a comparative analysis have been made using our supervised learning model and the results obtained can be used to predict the behavior of future borrowers. This model can assist a financial institution in assessing whether it should accept a loan request or not. Different combinations of feature selection algorithm and classifiers have been made and based upon metrics such as accuracy, AUC score, F1 score etc. the best model has been selected. Recursive feature elimination with cross validation (RFECV) and Principal Component Analysis (PCA) have been used to find the optimum number of features needed to make an accurate prediction. This allows us to make more efficient and optimal use of the limited available resources. The assessment will be performed in a supervised environment and so Support Vector Machines (SVM), Random Forest, Extreme Gradient Boosting and Logistic Regression have been used as the classifiers. In order to ensure all possible combinations have been properly tested k folds cross validation has been used to bring out a more balanced result. Furthermore, GridSearchCV has been used to tune the selected hyperparameters for each model in order to obtain the best result possible. And based upon this a comparison in a tabular form has been shown which showcases the most and the least accurate model for precisely assessing loan requests.
Causal inference in depression: understanding beyond correlation
(BRAC University, 2025-10) Shoumo, Syed Zamil Hasan; Alam, Md. Golam Rabiul
Depression remains one of the most pressing mental health concerns worldwide, intensified further by the socioeconomic and psychological impacts of the COVID-19 pandemic. Understanding the underlying mechanisms that contribute to depressive symptoms has therefore become a major research priority. While traditional statistical and machine learning models have been effective in identifying associations between risk factors and depression, they often fail to distinguish correlation from causation. Explainable Artificial Intelligence (XAI) methods, such as SHAP and LIME, have improved transparency by revealing which features influence model predictions; however, they remain fundamentally correlational and do not provide insight into the true causal pathways that drive depressive outcomes. To address this limitation, this research integrates machine learning, explainable AI, and causal inference to explore the causal factors behind depression among the Bangladeshi population during the COVID-19 pandemic. Using XGBoost for predictive modeling, the study first evaluates the relative importance of features through gain-based measures and SHAP value interpretation. Subsequently, a causal inference framework is constructed following Judea Pearl’s principles to identify and estimate direct causal effects using the backdoor adjustment method with a generalized linear model estimator. Finally, a combined feature-selection pipeline is developed that retains causally significant variables and iteratively removes weakly correlated ones to test their joint predictive strength. The results reveal that while several factors exhibit high correlation and feature importance in black-box models, only a subset demonstrates genuine causal influence on depressive outcomes. This distinction underscores the importance of causal reasoning in mental health analytics. Overall, the study establishes that integrating causal inference within predictive frameworks not only enhances interpretability and trustworthiness but also provides a clearer understanding of which factors can truly influence and potentially mitigate depression.
Intracranial hemorrhage detection using CNN-LSTM fusion model
(BRAC University, 2022-05) Ahmed, Kazi sabab; Shariar, Khandaker Sadab; Naim, Naimul Hasan; Hazari, MD. Nayimur Rahman; Alam, Md. Golam Rabiul; Shoumo, Syed Zamil Hasan
Intracranial Hemorrhage is a term used to describe bleeding between the brain tissue and the skull or within the brain tissue itself. It is life-threatening and needs immediate medical attention. As the first response, it is indispensable to detect the type of intracranial hemorrhage as soon as possible. Now, the manual detection methods require the help of an imaging expert and are certainly very time-consuming. Although there are several techniques for identifying them such as utilizing CT-scan images, magnetic resonance imaging (MRI), magnetic resonance angiogram (MRA), and ultrasound-based images, the results are still not adequate and have much room for improvement. In addition to these methods, researchers have also used imaging strategies based on Convolutional Neural Network (CNN) or Recurrent Neural Network (RNN) for this purpose. Therefore, this research aims to combine both these two fields and propose a model based on Deep Learning(DL) to detect intracranial hemorrhage. The goal of this paper is to automate the detection of intracranial hemorrhage and make the process more efficient and accurate. The model is expected to provide us with satisfactory results and can be used as an effective alternative to the existing methods.