Thesis (Bachelor of Science in Computer Science)

Browse

Author: search.filters.author.Tasnim, Zarin

Search Results

Now showing 1 - 4 of 4
  • Thumbnail Image
    Item
    MOODBOOST
    (BRAC University, 2025-06) Khan, Sami; Das, Arnab; Tasnim, Zarin; Shahnewaz, MD Golam; Debu, Soumitro Sharkar; Karim, Dewan Ziaul; Rahman, Rafeed
    An AI-Powered App for the Future. An App For Mental Health — Revolutionizing How we approach Mental Wellness. What is MoodBoost? MoodBoost is an AI-Powered application designed to revolutionize the mental health space by providing personalized, real-time interventions based on user emotional needs. In a time of mental health crises, which affect over 970 million people across the globe (WHO, 2022) with problems like stress, anxiety, and depression, current solutions provide static content and generalized recommendations that are often not enough. MoodBoost fills this gap by weaving in Gemini’s state-of-the-art Natural Language Processing (NLP)—which analyzes user inputs in real time—to allow the app to create contextually relevant affirmations, actionable recommendations, and therapeutic content. The app’s central innovation is its ability to dynamically adjust to users’ emotional states. For instance, a user who logs a ”stressed” mood may receive a curated meditation guide, while one who feels ”motivated” could be offered suggestions for goal-setting activities. This level of personalization is enabled by a sophisticated technical architecture like Frontend Built with Flutter for crossplatform performance on Android and iOS, For Backend, Utilizes Django REST Framework to manage API logic, handle user authentication, and validate data. For Database, The Firebase is used for structured and scalable storage of user profiles, mood logs, and AI generated content. For AI Integration, We are using Gemini, it uses NLP models to process mood data for insights and affirmations — replies are relevant and empathetic. The app’s approach integrates the latest advancements in AI with an emphasis on user-centered design, positioning it as a vital component in the landscape of contemporary mental health care, providing a forward-looking, data-guided experience with the potential to supplant traditional offerings.
  • Thumbnail Image
    Item
    A domain and noise adversarial bird tune classification pipeline using deep neural network
    (BRAC University, 9/29/2022) Riya, Aparna Sarker; Roy, Arpita; Fahim, Md. Abrar; Tasnim, Zarin; Islam, Rakibul; Mostakim, Moin; Reza, Md Tanzim
    Birds are an important category of animals that ecologists keep track of utilizing autonomous recording units as a key indication of environmental health. Because of the consequences of climate change and the rising number of endangered species, many experts suggested developing an animal species recognition system to help them in specialized research. Researchers can improve their ability to assess the state of biodiversity and its patterns in crucial ecosystems by precise sound detection and categorization, which is supported by machine learning, allowing them to better support global conservation efforts. However, producing analysis outputs with high precision and recall remains a difficulty. Due to a lack of appropriate methods for efficient and accurate extraction of interest signals, the vast bulk of data remains unexplored (e.g., bird calls). Moreover, due to strong source-domain specific features and artificial/natural noises, these acquired raw data create different distributions in datasets. So, to ensure a generalized feature learning, domain adaptation [1] techniques will be implemented in this work to make the networks familiar towards both acquisition sensor noises and background noises without having to do intensive dataset specific augmentations. We used 3 popular and powerful DNN models, including CNN, VGG19 and ResNet50. Out of them, for the bird species classification task VGG19 achieved the best accuracy of 96.02% in testing and 94.01% in training. To the best of our knowledge, this will guide towards convenient and deployable in real life models which will allow future works into the pipeline to ensure better coverage.
  • Thumbnail Image
    Item
    Detection of prodromal parkinson’s disease with fMRI data and deep neural network approaches
    (BRAC University, 2021-06) Shahriar, Farhan; Dey, Amarttya Prasad; Rahman, Naimur; Tasnim, Zarin; Tanvir, Mohammad Zubayer; Parvez, Zavid
    Parkinson’s Disease is the second most common neurological disease after Alzheimer’s Disease. The disease is incurable. However, if the disease can be detected earlier, then the consequences of it’s effect can be relieved. The early phase of PD is called by Prodromal Parkinson’s Disease. The symptoms of the Prodromal phase includes hyposmia, constipation, mood disorders, REM sleep behavior disorder, olfaction dis orders etc. RBD or REM sleep behavior disorder is the most common symptom of Prodromal PD. In this study, we used various deep convolutional neural network architectures and trained them to detect Prodromal PD patients. We collected 20 Prodromal patients and 20 healthy control subject data from the PPMI website and applied CNN architecture mobilenet v1, incception v3, vgg19 and inception resnet v2 to achieve our goal. We ensembled inception resnet v2 and mobilenet v1 with the hope of getting a better result as well. However, we successfully carried out our training and with mobilenet v1 we gained the highest classification accuracy of 81.22%. Inception resnet V2, inception v3, vgg19 and ensemble model achieved respectively 75.30%, 62.55% and 63.32% accuracy.
  • Thumbnail Image
    Item
    Automate crime news segmentation, knowledge-based generation and public opinion mining for social awareness
    (BRAC University, 2021-06) Sabit, Ahmed Bin; Ahsan, Asif; Tasnim, Zarin; Rasel, Annajiat Alim; Alam, Md. Golam Rabiul
    In the context of the crime scenario in Bangladesh, people face enormous crime situations which can be eradicated by applying some technological solution. For this a research work needs to be conducted by news segmentation and knowledge base generation. This study aims to determine how authority can target crime scenarios based on public opinion for increasing social awareness. Based on this context there will be an algorithm which will sort out the data sheet as well as the mined data will figure out solutions for respective crime scenarios. These data will be collected from news articles from which a data library will be generated by using natural language processing. After analyzing all data, it will automatically notify the authority with a precise report. Further research is needed to identify other factors that could strengthen the effectiveness of this report.