Authors

1. Ms. Aditi Tulchhia
2. Dr. Monika Rathore

Affiliation

• Research Scholar, RTU, Kota
• Associate Professor, International School of Informatics & Management, Jaipur

Abstract

The Random Forest model is utilised to differentiate Raisins into Kecimen and Besni. Chorionic Villus
Sampling (CVS) was used to collect photos of the Turkish raisin cultivars Kecimen and Besni. In all,
900 raisin kernels were used, with 450 pieces of each type. Following various stages of preprocessing,
seven morphological features were extracted from these pictures. These features were
classified using three artificial intelligence systems. On the basis of the features, the dispersion of
both raisin kinds were analysed and graphed. The subsequent model is built through using Random
Forest Machine Learning approach. As a result, the categorization accuracy achieved is 90.44
percent, classification error: 9.56 percent, weighted mean recall: 90.44 percent, weights: 1, 1,
weighted mean precision: 90.91 percent, weights: 1, 1, absolute error: 0.194 +/- 0.196, correlation:
0.814.

Keywords

Random Forest, Classification, Raisin Classification, Machine Learning, Prediction

References

1. Hossam M. Zawbaa, Hazman M, Abbass M, Hassanien A., (2014). Automatic fruit classification using random forest algorithm IEEE, 978-1-4799-7633-1/14.
2. Matthew M. Hayes, Scott N. Miller, and Melanie A. Murphy., (2014). High-resolution land cover classification using Random Forest, Remote Sensing Letters, Vol. 5, No. 1, 112–121.
3. Escobar, C. A., & Morales-Menendez, R. (2018). Machine learning techniques for quality control in high conformance manufacturing environment. Advances in Mechanical Engineering, 10(2), 1–16.
4. Cinar I., Koklu M. and Tasdemir S., (2020). Classification of Raisin Grains Using Machine Vision and Artificial Intelligence Method, Gazi Journal of Engineering Sciences,vol.6, no.3, pp. 200-209.