Browsing by Author "Nasrin, Sonia"

Filter results by typing the first few letters
Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Bangla movie success prediction using machine learning
    (Daffodil International University, 2019-05-26) Nasrin, Sonia
    In present world, machine learning techniques and models are used to predict future. Prediction the movie success before its release has been an immense point of concern for movie industry related all people, especially producer, shake holders and director. Since Bangladeshi movie industry is in threat, they need some kind of assurance that the movie will be successful or not and which factor can improve the profit. The purpose of the work is to make a model which predict Bangla movie success depending on some pre-release factors like actor, actress, director, producer, genre, budget, release date, duration, playback singer, music director that helps Bangladeshi movie industry to determinate specific reason for success so that they can resolve before release. In this model, data mining process and algorithm are applied for movie classification. Decision tree, Random forest, Logistic Regression, Support Vector machine are used and evaluated on this dataset and also focus on to find out some interesting relationship between features using feature engineering techniques and tools. Here, it is supervised classification which classify movie based on IMdb movie rating, where rating are divided into five classes (flop, bad, watchable, super hit and block buster). I visited all Bangladeshi movie related sites to collect data. As Bangladeshi movie data collection is highly unorganized and unavailable, our dataset is not much more longer. This prototype model has been provided much better performance in this challenging scenario.
  • Thumbnail Image
    Item
    Deepretina: Deep Learning Approach To Detect Retinal Abnormality In Computer Vision
    (DAFFODIL INTERNATIONAL UNIVERSITY, 2024-01-30) Nasrin, Sonia
    Currently, almost 1.2 million people in our country are blind, while around 3.51 lakh people have low vision. The pattern of eye abnormalities is changing along with an increasing rate of dry eye, cornea-related problems and eye problems related to diabetes. Early identification of eye diseases especially retinal abnormality plays a vital role to prevent the blurry vision in patients. In my research, a hybrid deep learning model is proposed to detect retinal abnormality by scanning a single retinal image of a patient. First, a new multi-label retinal disease dataset, Retinal Fundus Multi-Disease Image Dataset (RFMiD) version 02 is collected from a renewed journal website “Multidisciplinary Digital Publishing Institute” (mdpi), where 46 retinal diseases labels are available with high resolution. Next, dataset is going through analysis and preprocessing techniques to deals with data imbalance and large size (8gb) problem. Numerous analysis and experiments are performed to evaluate the models for better results. In the model, Convolutional neural models – EfficientNet, VGG16, NesNetMobile are used to analysis comparative result as well. EffectiveNet gives the highest accuracy among them and that is 85%. Voting Ensemble method is used to increase model accuracy (88%) for better prediction than could be gained from any of the constituent learning algorithms. This model is used to detect normal or abnormal retinal conditions for early treatment.