Predicting the Presence of Newt-Amphibian Using Genetic Programming

Created by W.Langdon from gp-bibliography.bib Revision:1.6946

@InProceedings{kumar:2022:ADIS,

• author = "Arvind Kumar and Nishant Sinha and Arpit Bhardwaj",
• title = "Predicting the Presence of Newt-Amphibian Using Genetic Programming",
• booktitle = "Advances in Data and Information Sciences",
• year = "2022",
• editor = "Shailesh Tiwari and Munesh C. Trivedi and Mohan Lal Kolhe and K. K. Mishra and Brajesh Kumar Singh",
• volume = "318",
• series = "Lecture Notes in Networks and Systems",
• pages = "215--223",
• address = "Agra",
• month = may # " 14-15",
• publisher = "Springer",
• keywords = "genetic algorithms, genetic programming, Amphibian classification, EDWB fitness function, Imbalanced data classification, Fitness function",
• isbn13 = "978-981-16-5689-7",
• URL = "http://link.springer.com/chapter/10.1007/978-981-16-5689-7_19",
• DOI = "doi:10.1007/978-981-16-5689-7_19",
• abstract = "In nature, aquatic ecosystems play a very important aspect. River valleys, wetlands, and water reservoirs are territories for various species of vegetation and wildlife. The prediction of these species is very important for natural resource planning. In this work, a publicly available UCI dataset containing extracted features from satellite imagery is used to classify the presence of newt-amphibians. We convert this multi-class classification problem to the binary classification problem. The transformation leads to being unbalanced classification problem. For the unbalanced classification, in the original form, most machine learning techniques give biased classification results, and their results are inclined in favor of the majority class. We use genetic programming with a newly proposed Euclidean distance and weight-based (EDWB) fitness function to resolve this problem. The result outcomes are compared with original work, support vector machine (SVM), and GP with the standard fitness function. The proposed approach achieves better results than the original work, SVM, and compared GP methods.",
• notes = "Also know as \cite{kumar2022predicting}",

}

Genetic Programming entries for Arvind Kumar Nishant Sinha Arpit Bhardwaj

Citations