Publication: Auto Training an AI for Detecting Plant Disease Using Twitter Data Annexed With a Plant Anthology
Type:
Article
Date
2021-12-09
Journal Title
Journal ISSN
Volume Title
Publisher
2021 3rd International Conference on Advancements in Computing (ICAC), SLIIT
Abstract
Agricultural productivity plays a vital role in
contributing to a nation’s economy. Farmers nowadays are
concerned due to disease persistence in crops and plants, and it
also affects the economy indirectly, so it is important to come up
with a solution to detect plant diseases and educate the farmers
about the solutions to retaliate against the diseases. Proper care
is mandatory to safeguard the quality of plants. The existing
traditional methods consume a massive amount of time and
resources hence, it’s costly. Due to the importance of continuous
monitoring, it seems impractical for a farmer to implement the
traditional methods on large scale. The Traditional systems
which are used lack the ability to identify diseases out of their
predefined scope. As a solution, we came up with an autolearning
system that identifies new plant diseases and provides
remedies. This paper showcases the image processing
techniques to detect plant diseases, Auto ML techniques to
create new models for plants and corresponding diseases,
Diseases are identified using image processing, Remedies are
extracted for the given plant diseases using unstructured data
from web data crawling. The business intelligence model uses
NLP to provide ideas about the trending plants and plantrelated
diseases are also discussed in this paper.
Description
Keywords
Continuous Learning, Twitter, CNN, Auto-Keras, AutoML, Image Classification, Web Data Extraction, Business Intelligence
