Browsing by Author "Ahamed, Shafin"

Filter results by typing the first few letters
Now showing 1 - 3 of 3
  • No Thumbnail Available
    Item
    A Comprehensive Study of Sentiment Analysis on Covid19 Vaccine using ML and DL Based on Social Media Reaction of Bangladeshi People
    (Scopus, 2023-03-01) Ahamed, Shafin; Hadi, Hanzala Ahamed; Ali, Md. Ahad; Pervej, Md.; Mamun, Md. Shariar
    Recently world is passing through a deadly pandemic called COVID-19. A global pandemic that was introduced or some say produced first in Wuhan, CHINA, and within some months spread around the world creating chaos in people's minds because mankind had not seen such kind of virus in decades. Spending life in hard lockdown and keeping social distance affects people's mental health and behavior. Within a year by the blessing of science, vaccines were introduced by different profitable and non-profitable organizations of different countries around the globe to fight back against the deadly virus. Due to mental health and behavior issue some peoples refuse to take the vaccine and also discoursed other people. Most of the activity happened on social media platforms like Facebook, and Twitter because of less monitoring. This effected the decision of people to take the vaccine as the vaccine is not taken as expected. On the other hand, some people express the benefit of talking about vaccines. In this paper, The Facebook comment that was posted related to covid-19 vaccine and vaccination campaign and tried to analyze the sentiment on covid-19 vaccine of people of Bangladesh has been examined. Due to the fact that Facebook comments allow users to voice their opinions, a dataset has been prepared from these comments. Dataset contained 1700 entries of Bangla comments and was labeled each entry with the corresponding sentiment from positive, negative, or neutral. Bangla comment was converted into numerical value by using the tokenizer method and performed various machine learning and deep learning algorithm on 1300 dataset which were where the LSTM algorithm appears as the best fit with an accuracy of 92.27%. LSTM works perfectly in the dataset containing Bangla comments.
  • No Thumbnail Available
    Item
    Performance Evaluation of Several Transfer Learning Models for Classification of Road Surface State
    (Springer Nature, 2023-10-22) Rahman, Fahim Ur; Ahmed, Md. Tanvir; Khan, Emran; Rahman, Md Mahfuzur; Ahamed, Shafin; Mamun, Shahriar; Hasan, Md Mehedi
    Using a customized approach, automatic classification of road surface condition and categorized data storing are proposed using DenseNet201. Road surface distress is one of the main issues affecting transportation safety. The first indication of a catastrophic asphalt pavement collapsing is a surface crack, that can later develop into a pothole and result in high repair costs. By replacing the surveillance system with the automated software program that we are recommending in this analysis, the traditional methods for identifying cracks or degradation in a road's surface, which involved manual examination by people, can be eliminated. DenseNet201 has outperformed other compared models with an accuracy of 98.75% and the most minimal model loss while testing while classifying damaged and smooth road surfaces. Later, certain governing bodies responsible for preserving the quality of road infrastructure can use the model's classified images.
  • No Thumbnail Available
    Item
    Predicting Sugarcane Yields using Supervised Learning: A Comparative Study
    (IEEE, 2023-12-23) Papon, Parvez Ahmed; Rahman, Md Mahfuzur; Ahamed, Shafin; Mamun, Shahriar; Mehadi, Md Zahirul Islam; Polin, Johora Akter
    Bangladesh's sole source of white sugar is sugarcane, an agricultural commodity used to produce biofuels and other products. In recent years, ML techniques have been used to forecast yield, with encouraging outcomes. In this academic study, a supervised ML technique is suggested to forecast sugarcane yield from the perspective of Bangladesh. Weather patterns, sugarcane yield, and other relevant information are gathered from a variety of sources, including the BBS and the BMD. Many ML models, including Linear Regression, GBR, DTR, RFR, and XGBR, are trained. Different evaluation measures are used to compare the effectiveness of various models, such as MAE, MSE, and RMSE. The GBR model fared better than other models, according to the results. The results of this study can be used by farmers, policymakers, and the sugar industry to improve sugarcane yield in Bangladesh and make educated decisions. Future studies can investigate how to estimate sugarcane productivity in other areas and for other crops using ML approaches.