20.500.12556/RUL-110204 Napovedovanje težavnosti problemov v balvanskem plezanju Predicting problem difficulty in bouldering Ocenjevanje problemov v balvanskem plezanju je zelo pomemben del tega športa, prav tako pa je to tudi zelo zahtevna naloga, uspeh katere je odvisen predvsem od ocenjevalčevih izkušenj in doslednosti pri ocenjevanju. Diplomska naloga primerja različne predstavitve lastnosti balvanskih problemov ter uspešnost metod strojnega učenja pri ocenjevanju na podlagi izpeljanih atributov v upanju, da so modeli, kot so model naključnih gozdov, linearne regresije in najbližjih sosedov, dovolj natančni, da bi lahko podali smiselno oceno za neocenjen balvanski problem. V nalogi rezultate razdelimo na več faz, kjer v vsaki fazi dodamo prej neuporabljene ali na novo izpeljane atribute ter primerjamo, kakšno uspešnost dosežejo modeli z novim naborom atributov. Na koncu atribute posplošimo, saj tako dobimo rezultate za nabor atributov, ki ga lahko prenesemo na katero koli plezalno steno, s čimer dobi diplomska naloga uporabno vrednost. Rezultat dela so izdelani modeli strojnega učenja, ki se pri napovedovanju ocene neocenjenega balvanskega problema za poljubno plezalno steno zmotijo za 1,69 ocene. Grading bouldering problems is a very important part of bouldering. It’s also a very demanding task, the success of which is determined by the grader’s experience and consistency when grading. This thesis compares different approaches to encode bouldering problem properties and the accuracy of machine learning models, such as random forest, linear regression and k nearest neighbours, when grading bouldering problems, in hope that the machine learning models prove to be sufficiently accurate to appropriately grade an ungraded bouldering problem. Results are presented in different phases. A new set of unused or newly created features is added in each subsequent phase and the accuracy of learning is compared across different models for the selected subset of features. Finally a subset of generic features is used, such that it can be transferred to any climbing wall, which adds additional value to the work done in the thesis. The end result is a set of machine learning models, which, when grading an ungraded bouldering problem for an arbitrary climbing wall, miss the correct prediction for 1.69th of the grade. strojno učenje ocenjevanje balvanskih problemov balvansko plezanje atributi machine learning grading bouldering problems bouldering features true false false Slovenski jezik Angleški jezik Diplomsko delo/naloga 2019-09-12 21:45:19 2019-09-12 21:45:23 2022-08-22 03:42:20 0000-00-00 00:00:00 2019 0 0 0000-00-00 NiDoloceno NiDoloceno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 1970-01-01 23594 1538355651 Repe_Jure_-_Napovedovanje_tezavnosti_problemov_v_balvanskem_plezanju.pdf Repe_Jure_-_Napovedovanje_tezavnosti_problemov_v_balvanskem_plezanju.pdf 1 DC9BBFE8581D9D18F15642C4CE63DA02 5a0d2201d413f3af9f78183b7ecb4277eadcca2a6f72e95cb51ed0613e797ed7 108fc9e6-a1b7-11eb-a523-00155dcfd717 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=121849 Fakulteta za računalništvo in informatiko 0 0 0