Browsing by Author "Kazi, Md. Aslam"

Now showing 1 - 3 of 3

Analysing Most Efficient Deep Learning Model To Detect COVID-19 From Computer Tomography Images
(Daffodil International University, 2022-09-16) Shamrat, F. M. Javed Mehedi; Chakraborty, Sovon; Ahammad, Rasel; Shitab, Tanzil Mahbub; Kazi, Md. Aslam; Hossain, Alamin; Mahmud, Imran
COVID-19 illness has a detrimental impact on the respiratory system, and the severity of the infection may be determined utilizing a selected imaging technique. Chest computer tomography (CT) imaging is a reliable diagnostic technique for finding COVID-19 early and slowing its progression. Recent research shows that deep learning algorithms, particularly convolutional neural network (CNN), may accurately diagnose COVID-19 using lung CT scan images. But in an emergency, detection accuracy simply is not enough. Determinants of data loss and classification completion time play a critical element. This study addresses the issue by finding the most efficient CNN model with the least data loss and classification time. Eight deep learning models, including Max Pooling 2D, Average Pooling 2D, VGG19, VGG16, MobileNetV2, InceptionV3, AlexNet, NFNet using a dataset of 16000 CT scans image data of COVID-19 and non-COVID-19 are compared in the study. Using the confusion matrix, the performance of the models is compared and together with the data loss and completion time. It is observed from the research that MobileNetV2 provides the highest accurate result of 99.12% with the least data loss of 0.0504% in the lowest classification completion time of 16.5secs per epoch. Thus, employing MobileNetV2 gives the best and the quickest result in an emergency.
Crime Data Mining : A General Framework and Approach to Identify Crime Pattern Using Machine Learning
(Daffodil International University, 2018-11) Sharmin, Mahmuda; Kazi, Md. Aslam
Crime data mining has been a hot topic in the law-enforcement sector as it can solve crime related problems more effectively. In this thesis, we have focused in determining crime patterns using effective data mining algorithms. To classify our raw data we used Naïve Bayes classifier along with NER (Named Entity Recognition) and Co-reference Resolution concept. In order to get higher accuracy, we have trained numerous train data with required keywords. To get the related keywords and train data, both online and offline data are used. For the validation of the process, filed work like talking to “Adabor Thana Police Station” is also performed. Test results are generated based on sample input and results. Our thesis results generates the report which represents crimes stativity of different divisions of Bangladesh. The output data also shows how much crimes are occurring in certain place which can be used for crime pattern analysis for a specific location like an individual district or even for a police station. Upon successful implementation, it our thought that law-enforcement authority will find our findings useful to detect crime sensitive zones of Bangladesh and take precautions so that number of crimes can be reduced.
Implementation of Machine Learning Algorithms to Detect the Prognosis Rate of Kidney Disease
(2020 IEEE International Conference for Innovation in Technology (INOCON) , IEEE, 2020-11) Shamrat, F.M. Javed Mehedi; Ghosh, Pronab; Sadek, Mahbubul Hasan; Kazi, Md. Aslam; Shultana, Shahana
The chronic kidney disease is the loss of kidney function. Often time, the symptoms of the disease is not noticeable and a significant amount of lives are lost annually due to the disease. Using machine learning algorithm for medical studies, the disease can be predicted with a high accuracy rate and a very short time. Using four of the supervised classification learning algorithms, i.e., logistic regression, Decision tree, Random Forest and KNN algorithms, the prediction of the disease can be done. In the paper, the performance of the predictions of the algorithms are analyzed using a pre-processed dataset. The performance analysis is done base on the accuracy of the results, prediction time, ROC and AUC Curve and error rate. The comparison of the algorithms will suggest which algorithm is best fit for predicting the chronic kidney disease.