Sheikh Abujar

Abu Kaisar Mohammad Masum, Nusrat Badhon, S. M. Saiful Islam Badhon et al.

Abnormal growth of cells in the brain and its surrounding tissues is known as a brain tumor. There are two types, one is benign (non-cancerous) and another is malignant (cancerous) which may cause death. The radiologists' ability to diagnose malignancies is greatly aided by magnetic resonance imaging (MRI). Brain cancer diagnosis has been considerably expedited by the field of computer-assisted diagnostics, especially in machine learning and deep learning. In our study, we categorize three different kinds of brain tumors using four transfer learning techniques. Our models were tested on a benchmark dataset of $3064$ MRI pictures representing three different forms of brain cancer. Notably, ResNet-50 outperformed other models with a remarkable accuracy of $99.06\%$. We stress the significance of a balanced dataset for improving accuracy without the use of augmentation methods. Additionally, we experimentally demonstrate our method and compare with other classification algorithms on the CE-MRI dataset using evaluations like F1-score, AUC, precision and recall.

Md. Zahin Hossain George, Naimul Hossain, Md. Rafiuzzaman Bhuiyan et al.

There have recently been many cases of unverified or misleading information circulating quickly over bogus web networks and news portals. This false news creates big damage to society and misleads people. For Example, in 2019, there was a rumor that the Padma Bridge of Bangladesh needed 100,000 human heads for sacrifice. This rumor turns into a deadly position and this misleading information takes the lives of innocent people. There is a lot of work in English but a few works in Bangla. In this study, we are going to identify the fake news from the unconsidered news source to provide the newsreader with natural news or real news. The paper is based on the combination of convolutional neural network (CNN) and long short-term memory (LSTM), where CNN is used for deep feature extraction and LSTM is used for detection using the extracted feature. The first thing we did to deploy this piece of work was data collection. We compiled a data set from websites and attempted to deploy it using the methodology of deep learning which contains about 50k of news. With the proposed model of Multichannel combined CNN-LSTM architecture, our model gained an accuracy of 75.05%, which is a good sign for detecting fake news in Bangla.

Hasmot Ali, Md. Fahad Hossain, Md. Mehedi Hasan et al.

Voice based applications are ruling over the era of automation because speech has a lot of factors that determine a speakers information as well as speech. Modern Automatic Speech Recognition (ASR) is a blessing in the field of Human-Computer Interaction (HCI) for efficient communication among humans and devices using Artificial Intelligence technology. Speech is one of the easiest mediums of communication because it has a lot of identical features for different speakers. Nowadays it is possible to determine speakers and their identity using their speech in terms of speaker recognition. In this paper, we presented a method that will provide a speakers geographical identity in a certain region using continuous Bengali speech. We consider eight different divisions of Bangladesh as the geographical region. We applied the Mel Frequency Cepstral Coefficient (MFCC) and Delta features on an Artificial Neural Network to classify speakers division. We performed some preprocessing tasks like noise reduction and 8-10 second segmentation of raw audio before feature extraction. We used our dataset of more than 45 hours of audio data from 633 individual male and female speakers. We recorded the highest accuracy of 85.44%.

AKM Shahariar Azad Rabby, Hasmot Ali, Md. Majedul Islam et al.

This research paper presents a unique Bengali OCR system with some capabilities. The system excels in reconstructing document layouts while preserving structure, alignment, and images. It incorporates advanced image and signature detection for accurate extraction. Specialized models for word segmentation cater to diverse document types, including computer-composed, letterpress, typewriter, and handwritten documents. The system handles static and dynamic handwritten inputs, recognizing various writing styles. Furthermore, it has the ability to recognize compound characters in Bengali. Extensive data collection efforts provide a diverse corpus, while advanced technical components optimize character and word recognition. Additional contributions include image, logo, signature and table recognition, perspective correction, layout reconstruction, and a queuing module for efficient and scalable processing. The system demonstrates outstanding performance in efficient and accurate text extraction and analysis.

Jabir Al Nahian, Abu Kaisar Mohammad Masum, Sheikh Abujar et al.

In this era, the moment has arrived to move away from disease as the primary emphasis of medical treatment. Although impressive, the multiple techniques that have been developed to detect the diseases. In this time, there are some types of diseases COVID-19, normal flue, migraine, lung disease, heart disease, kidney disease, diabetics, stomach disease, gastric, bone disease, autism are the very common diseases. In this analysis, we analyze disease symptoms and have done disease predictions based on their symptoms. We studied a range of symptoms and took a survey from people in order to complete the task. Several classification algorithms have been employed to train the model. Furthermore, performance evaluation matrices are used to measure the model's performance. Finally, we discovered that the part classifier surpasses the others.

Mirajul Islam, Nushrat Jahan Ria, Jannatul Ferdous Ani et al.

A deep learning model gives an incredible result for image processing by studying from the trained dataset. Spinach is a leaf vegetable that contains vitamins and nutrients. In our research, a Deep learning method has been used that can automatically identify spinach and this method has a dataset of a total of five species of spinach that contains 3785 images. Four Convolutional Neural Network (CNN) models were used to classify our spinach. These models give more accurate results for image classification. Before applying these models there is some preprocessing of the image data. For the preprocessing of data, some methods need to happen. Those are RGB conversion, filtering, resize & rescaling, and categorization. After applying these methods image data are pre-processed and ready to be used in the classifier algorithms. The accuracy of these classifiers is in between 98.68% - 99.79%. Among those models, VGG16 achieved the highest accuracy of 99.79%.

Abu Kaisar Mohammad Masum, Sheikh Abujar, Sharmin Akter et al.

An AI chatbot provides an impressive response after learning from the trained dataset. In this decade, most of the research work demonstrates that deep neural models superior to any other model. RNN model regularly used for determining the sequence-related problem like a question and it answers. This approach acquainted with everyone as seq2seq learning. In a seq2seq model mechanism, it has encoder and decoder. The encoder embedded any input sequence, and the decoder embedded output sequence. For reinforcing the seq2seq model performance, attention mechanism added into the encoder and decoder. After that, the transformer model has introduced itself as a high-performance model with multiple attention mechanism for solving the sequence-related dilemma. This model reduces training time compared with RNN based model and also achieved state-of-the-art performance for sequence transduction. In this research, we applied the transformer model for Bengali general knowledge chatbot based on the Bengali general knowledge Question Answer (QA) dataset. It scores 85.0 BLEU on the applied QA data. To check the comparison of the transformer model performance, we trained the seq2seq model with attention on our dataset that scores 23.5 BLEU.