http://dspace.ewubd.edu:8080/handle/123456789/1023 Sat, 04 Apr 2026 11:34:34 GMT 2026-04-04T11:34:34Z http://dspace.ewubd.edu:8080/handle/123456789/4342 Suicide & Depression Detection Using Machine Learning Ahmed, Towsif; Poly, Irean Hossain; Mim, Mohsina Zaman; Faruk, Md. Omar Nowadays, depression is the most common mental-health issue and the leading cause of suicide and self-injurious behavior. Due to social stigma and lack of knowledge, clinical diagnosis of various mental health conditions is costly and often disregarded. These days, people choose to communicate using online social media platforms to convey their ideas, sentiments, and emotions. As a result, non-clinical mental health screening and surveillance can be facilitated by using user-generated information from online social media platforms. For quite some time, social media data has been utilized to automatically identify mental health problems through the use of traditional machine learning and natural language processing approaches. And the goal of our research is to examine how Machine Learning methods may be applied to the early detection and non-clinical, predictive diagnosis. To the best of our knowledge, we did not find any systematic literature review that studies the applications of machine learning techniques in this domain. In order to address this research gap, we conducted a systematic literature review relevant research studies published until date that have applied machine learning techniques for detecting depression and suicide or self-harm behavior from social media content. Our work comprehensively covers state-of-the-art WRT. techniques, features, datasets, and performance metrics, which can be of great value to researchers. We enumerate all the available datasets in this domain and discuss their characteristics. We also discuss the research gaps, challenges, and future research directions for advancing & catalyzing research in this domain. To the best of our knowledge, our study is the only and the most recent survey for this domain covering the latest research published until date. Based on our learnings from this review, we have also proposed a framework for mental health surveillance. We believe the findings of our work will be beneficial for researchers working in this domain. This thesis submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Information and Communication Engineering of East West University, Dhaka, Bangladesh Thu, 18 Apr 2024 00:00:00 GMT http://dspace.ewubd.edu:8080/handle/123456789/4342 2024-04-18T00:00:00Z http://dspace.ewubd.edu:8080/handle/123456789/4248 Unmasking Deception: Analyzing Fake Product Reviews through Machine and Deep Learning Islam, Eva; Moon, Marzana Rahman; Vasha, Tasnim Karim; Mahdi, Md. Tasean Online product evaluations have become a vital resource for shoppers looking for knowledge and direction for their purchases in the age of digital commerce. The prevalence of fraudulent product evaluations has jeopardized this priceless source of knowledge, seriously compromising the trustworthiness and integrity of online review systems. By deeply examining the identification and analysis of phony product evaluations and utilizing the capabilities of machine and deep learning algorithms, this thesis aims to address this problem. This research project starts by gathering a wide range of product reviews from different online sources. We make sure the data is consistent and of good quality by using rigorous preprocessing. We then extract important features from the reviews, like sentiment, language, and metadata, to help us analyze them. We use regular machine learning models to figure out if the reviews are real or fake, which helps us measure performance. Then, we use deep learning techniques like CNNs to get into the details of the text, which helps us detect fake reviews more accurately. This study also emphasizes the interpretability and explainability of model predictions, offering insight into the variables influencing the detection of false reviews. Our algorithms are applied to real-world datasets and settings, proving their efficacy in identifying fraudulent product evaluations across a variety of sectors, allowing us to evaluate the practical value of our research. In addition to algorithmic skill, ethical issues relating to privacy and fairness in fake review analysis are thoughtfully addressed, ensuring that the creation and application of these models are in line with responsible AI practices. To sum up, this thesis helps with continuous efforts to protect the integrity of online review systems, allowing customers to make wise decisions, and upholding the reliability of e-commerce platforms. By exploring the complex world of bogus products, This thesis submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Information and Communication Engineering of East West University, Dhaka, Bangladesh Tue, 28 Nov 2023 00:00:00 GMT http://dspace.ewubd.edu:8080/handle/123456789/4248 2023-11-28T00:00:00Z http://dspace.ewubd.edu:8080/handle/123456789/3942 Analyzing the Performance of DV-Hop Based Localization Algorithms in Range-free Wireless Sensor Networks Afrin, Tanjina; Shahan, Shakibul Hasnat; Rupom, Md. Ashraful Haque Wireless sensor networks are used in a wide range of applications, including environmental monitoring, industrial automation, and healthcare. In order to effectively gather and analyze data from these networks, it is crucial to accurately determine the position of the sensors. Range-free localization algorithms and techniques are often used for this purpose, as they are able to adapt to changing conditions and do not require the use of fixed ranges or predetermined reference points. The DV-Hop algorithm is a popular choice for range-free localization, as it is able to adapt to changing conditions and can be used in a variety of environments. However, the DV-Hop algorithm has its limitations, as it can be prone to error and may not be as accurate as other localization methods. In this study, we evaluated the performance of the DV-Hop algorithm and its improved versions in order to determine their accuracy and effectiveness in range-free localization. Our results showed that the DV-Hop algorithm and its improved versions are effective at determining the position of the sensors in a range-free wireless sensor network, with the improved versions performing significantly better than the original algorithm. However, there is still a significant level of error present, indicating that further improvements could be made to increase the accuracy of the algorithm. Overall, our study highlights the importance of range-free localization in wireless sensor networks and the need for further research and development in this area in order to improve the accuracy, efficiency, and adaptability of range-free localization methods. This thesis submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Information and Communication Engineering of East West University, Dhaka, Bangladesh Wed, 18 Jan 2023 00:00:00 GMT http://dspace.ewubd.edu:8080/handle/123456789/3942 2023-01-18T00:00:00Z http://dspace.ewubd.edu:8080/handle/123456789/3938 Analysis on Covid-19 Detection System Using Machine Learning & Deep Learning Models Rithin, Nuzhat Tabassum; Das, Joy; Das, Bijoy COVID-19, considered the deadliest virus of the twenty-first century, has claimed the lives of millions of people worldwide in less than two years. The respiratory disease (COVID- 19) is caused by the novel coronavirus SARS-CoV-2, which originated in Wuhan, [14] China in late December of 2019. By October 2020, the virus already infected almost 40,000,000 people dead over one million (Hopkins (2020)). This infection has rapidly expanded across China and into other nations since then, creating a global pandemic in 2020 due to its ease of transmission from person to person via respiratory droplets. Pneumonia is another infectious condition that is frequently caused by a bacterial infection in the alveoli of the lungs. When an infected lung tissue becomes inflamed, pus forms in it. Because the virus first affects the lungs of patients, X-ray imaging of the chest is useful for accurate diagnosis. To determine whether a patient has these conditions, experts conduct physical examinations and diagnose them with a chest X-ray, ultrasound, or a lung biopsy. In this analysis, we recommend using a chest X-ray to prioritize people for subsequent RT-PCR testing. It would also aid in the identification of patients with a high chance of COVID and a false-negative RT-PCR who require additional testing. It is urgent to create auto- mated technologies that could diagnose this disease in its early stages, in a non-invasive manner, and in a shorter amount of time. However, selecting the most accurate models to characterize COVID-19 patients is challenging due to the inability to compare the outputs of diverse data types and gathering methods. This is the only way to remedy the issue. As a result, much research has been conducted to establish an appropriate method for diagnosing and classifying people as COVID-19-positive, healthy, or affected by other pulmonary lung illnesses. In a few earlier scholarly works, semiautomatic machine learning techniques with limited precision were proposed. In this study, we wanted to develop reliable deep learning approaches, which are a subset of machine learning and AI that model the way humans acquire knowledge. Data science encompasses fields like statistics and predictive modeling, two of which benefit greatly from deep learning. One component of this is what are known as convolutional neural networks (CNN). Any automatic, reliable, and accurate screening strategy for COVID- 19 detection would be helpful for rapid diagnosis and reducing exposure to the virus for medical or healthcare personnel. The work takes advantage of a versatile and successful deep learning approach by employing the CNN model to predict and identify a patient as being unaffected or impacted by the disease using an image from a chest X-ray. In order to prove how well the CNN model was trained, the researchers employed a dataset consisting of 10,000 images with a resolution of 224x224 and 29 batches. Convolutional neural networks (CNNs) were demonstrated to be very effective for medical picture classification. The authors of this piece propose using convolutional neural networks (CNNs) to automatically classify chest X-ray images for signs of COVID-19. Using the dataset, eleven current CNN models— max poling operation, and SoftMax activation function—that can distinguish between COVID-19 and other lung diseases—were first used to identify the symptoms of COVID-19. A stratified 5machine learning technique was utilized with a ratio of 80 percent for training and for testing (unseen folds), and 20 percent of the training data was used as a validation set to prevent overfitting problems. During the performance training, the trained model produced an accuracy rate of 98 percent. The research study can use chest X-ray pictures to identify and de- test COVID-19, normal, and pneumonia infections, according to the results of the tests. This thesis submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Information and Communication Engineering of East West University, Dhaka, Bangladesh Wed, 18 Jan 2023 00:00:00 GMT http://dspace.ewubd.edu:8080/handle/123456789/3938 2023-01-18T00:00:00Z