Multilayer Perceptron and Logistic Regression Algorithms for Rice Dataset Prediction: A Comparative Analysis
Abstract
Rice being perhaps of the most broadly delivered and consumed cereal harvest on the planet, is likewise the one of the fundamental food in our country on account of its efficient and nutritious nature. The prediction of crop yield plays a crucial role in agricultural planning and resource allocation. In this study, we compare the performance of two popular machine learning algorithms, Multilayer Perceptron (MLP) and Logistic Regression (LR), for predicting rice crop yield using the Rice Dataset. The dataset contains various features related to environmental conditions, soil properties, and agricultural practices, along with corresponding yield measurements. The algorithms were trained and evaluated based on accuracy, precision, and recall metrics to assess their predictive capabilities.
Keywords
Download Options
Introduction
At the point when we take a gander at the development of grain items all through the world, rice is the main item following wheat and corn. Rice is very wealthy in sugars and starch. Rice has extraordinary significance in human nourishment in our country as well as on the planet as far as being nutritious and prudent [6]. It is likewise broadly utilized in industry. Different quality models for rice creation in our nation is made accessible. These are actual appearance, cooking qualities, fragrance, taste and smell are issues, for example, proficiency close to the properties. According to the point of view of the end buyer's, the main element actual appearance strikes a chord from the rules that hang out in the rice assortments that are sold bundled on market racks [9]. After creation, it is seen that the requirement for mechanical techniques increments on the grounds that the alignment of rice, assurance of its sorts, and detachment of different quality components are wasteful and tedious, particularly regarding those with high creation volume. Subsequently, when we take a gander at the new examinations on cereal items utilizing machine vision frameworks and picture handling strategies, it is seen that the items are analyzed as far as numerous actual properties like tone, surface, quality and size.
Conclusion
In conclusion, this study demonstrates the successful application of both Multilayer Perceptron and Logistic Regression algorithms for predicting rice crop yield. The results obtained underscore the potential of machine learning techniques in agricultural planning and decision-making, providing valuable insights for optimizing resource allocation and improving agricultural productivity.