dc.creator | Jocković, Milan | |
dc.creator | Cvejić, Sandra | |
dc.creator | Jocić, Siniša | |
dc.creator | Radeka, Ilija | |
dc.creator | Jocković, Jelena | |
dc.creator | Radanović, Aleksandra | |
dc.creator | Terzić, Sreten | |
dc.creator | Dedić, Boško | |
dc.date.accessioned | 2022-07-20T10:05:57Z | |
dc.date.available | 2022-07-20T10:05:57Z | |
dc.date.issued | 2022 | |
dc.identifier.isbn | 978-86-80417-89-9 | |
dc.identifier.uri | http://fiver.ifvcns.rs/handle/123456789/2900 | |
dc.description.abstract | Use of multivariate modelling in order to improve prediction accuracy has been widely applied in plant breeding programs. In these models phenotype prediction is based on large number of independent variables which is at the same time strength and weakness. Lately, intensive research in order to improve prediction accuracy resulted in extensive use of different machine learning techniques. The aim of this study is to present new approaches in phenotype prediction based on complex relationships between genotypes and phenotypes. Widely used, one of the main tools in machine learning is artificial neural network (ANN). Although it has a long history, this powerful class of algorithms has been recently used as a state-of-the-art solution for non-linear relationship between the genotype and the trait of interest. Another important advance, capable of identifying extremely complex patterns of prediction and classification of information is called deep learning (DL). Main difference between DL and ANN is in the numbers of layers of neurons. Being based on how humans learn and process information, machine learning is powerful tool for processing complex data in order to make accurate predictions. | sr |
dc.language.iso | en | sr |
dc.publisher | Paris : International Sunflower Association | sr |
dc.publisher | Novi Sad : Institute of Field and Vegetable Crops | sr |
dc.relation | info:eu-repo/grantAgreement/ScienceFundRS/Ideje/7732457/RS// | sr |
dc.relation | info:eu-repo/grantAgreement/MESTD/inst-2020/200032/RS// | sr |
dc.relation | COST Action CA18111: Genome editing in plants - a technology with transformative potential | sr |
dc.relation | COST Action CA16212: Impact of Nuclear Domains on Gene Expression and Plant Traits | sr |
dc.relation | Climate Crops - Centre of Excellence for Innovations in Breeding of Climate-Resilient Crops, Institute of Field and Vegetable Crops | |
dc.rights | openAccess | sr |
dc.rights.uri | https://creativecommons.org/licenses/by/4.0/ | |
dc.source | Proceedings, 20th International Sunflower Conference, 20-23 June 2022, Novi Sad, Serbia | sr |
dc.subject | multivariate modelling | sr |
dc.subject | phenotype prediction | sr |
dc.subject | artificial neural networks | sr |
dc.subject | deep learning | sr |
dc.subject | machine learning | sr |
dc.title | New approaches in phenotype prediction – machine learning techniques | sr |
dc.type | conferenceObject | sr |
dc.rights.license | BY | sr |
dc.citation.epage | 140 | |
dc.citation.spage | 140 | |
dc.identifier.fulltext | http://fiver.ifvcns.rs/bitstream/id/8098/bitstream_8098.pdf | |
dc.identifier.rcub | https://hdl.handle.net/21.15107/rcub_fiver_2900 | |
dc.type.version | publishedVersion | sr |