Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE credits
Context: The context of this research is to forecast the sales of truck componentsusing machine learning algorithms which can help the organization in activity oftrade and business and it also plays a major role for firms in decision-making operationsin the areas corresponding to sales, production, purchasing, finance, and accounting.
Objectives: This study first investigates to find the suitable machine learning algorithmsthat can be used to forecast the sales of truck components and then theexperiment is performed with the chosen algorithms to forecast the sales and to evaluatethe performances of the chosen machine learning algorithms.
Methods: Firstly, a Literature review is used to find suitable machine learningalgorithms and then based on the results obtained, an experiment is performed toevaluate the performances of machine learning algorithms.
Results: Results from the literature review shown that regression algorithms namely Supports Vector Machine Regression, Ridge Regression, Gradient Boosting Regression, and Random Forest Regression are suitable algorithms and results from theexperiment showed that Ridge Regression has performed well than the other machine learning algorithms for the chosen dataset.
Conclusion: After the experimentation and the analysis, the Ridge regression algorithmhas been performed well when compared with the performances of the otheralgorithms and therefore, Ridge Regression is chosen as the optimal algorithm forperforming the sales forecasting of truck components for the chosen data.
2019. , p. 49