Predviđanje količine prodaje proizvoda zasnovano na modelima strojnog učenja

master's thesis

• Title (english): Prediction of product sales using machine learning models
• Author: Šimun Smolčec
• Mentor: Goran Delač (mentor)
• Committee member: Goran Delač (predsjednik povjerenstva)
• Committee member: Marin Šilić (član povjerenstva)
• Committee member: Klemo Vladimir (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2023-07-04, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

U ovom radu objašnjen je postupak predviđanja količine prodaje proizvoda pomoću modela zasnovanih na strojnom učenju. Objašnjen je cijeli postupak od pripreme podataka do odabira i učenja modela. Modeli su isprobani na univarijantnom i multivarijantnom skupu podataka. Isprobani su sljedeći modeli: LGBM, XGBoost, Catboost, Hrbatna regresija, Hist Gradient Boosting, Huberova regresija, Kneighbors, Holt winters, Povijesni prosjek, Optimized Theta i MSTL. Modeli koji su zasnovani na stablima odluke i gradient boostingu su se pokazali najbolji za isprobani multivarijantni skup, a na isprobanom univarijantnom skupu podataka HistGradientBoosting model se pokazao kao najprecizniji.

Abstract (english)

This paper explains the process of forecasting the amount of product sales using models based on machine learning. The entire process from data preparation to model selection and training is explained. The models were tested on univariate and multivariate datasets. The following models were tested: LGBM, XGBoost, Catboost, Ridge regression, Hist Gradient Boosting, Huber regression, Neighbors, Holt winters, Historic average, Optimized Theta and MSTL. Models based on decision trees and gradient boosting proved to be the best for the tested multivariate dataset, and on the tested univariate dataset the HistGradientBoosting model proved to be the most accurate.

• Keywords: strojno učenje; model; predviđanje prodaje
• Keywords (english): machine learning; models; sales forecasting
• Language: croatian
• URN:NBN: urn:nbn:hr:168:724704
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-10-11 10:28:19