img
img
img
img
img
link
Home / all-journals/ /Article

Impact of Weather on Crops in Few Northern Parts of Bangladesh: HCI and Machine Learning Based Approach


Md. Toukir Ahmed1*, Md. Niaz Imtiaz1, and Nurun Sakiba Mitu1


1Department of Computer Science and Engineering, Pabna University of Science and Technology, Pabna, Bangladesh

*Correspondence: toukirahmedreal@gmail.com

Powered by Froala Editor


ABSTRACT

As Bangladesh is an agricultural country, the economy, as well as the food security of this country, mostly depends on the production level of different crops over the year. Therefore, there exists immense pressure on exaggerated crop production due to the fast growth of the population. But, the average production level is being hampered by the bad nature of the weather. We have conducted a survey on near about 100 farmers of two northern districts of Bangladesh: Pabna and Rajshahi and assessed the impact of rough nature on production. According to farmers and agriculturalists, it is noticed that rough weather causes about 30% to 70% production shortage than expectation with all other factors remaining constant. In this study, we have adopted Human-computer interaction (HCI) based approach (Soft System Methodology-SSM) to this aspect for efficacious collaboration with root-level farmers and agricultural trainers providing ease for understanding weather-related issues on the production of crops. Finally, some machine learning algorithms were also implemented on the obtained dataset to accurately classify the range of production level of rice and a comparison is made among the algorithms based on performance metrics. Moreover, an android based application is created to depict the summary of the study. 


Keywords: HCI, Machine learning, Performance metrics, SSM, Weather, Impact, and Northern parts.


Citation: Ahmed MT, Imtiaz MN, and Mitu NS. (2020). Impact of weather on crops in few northern parts of Bangladesh: HCI and machine learning based approach, Aust. J. Eng. Innov. Technol., 2(1), 7-15. https://doi.org/10.34104/ajeit.020.07015


Powered by Froala Editor