MACHINE LEARNING FOR MECHANICAL ANALYSIS
2019 (English)Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE credits
Student thesis
Abstract [en]
It is not reliable to depend on a persons inference on dense data of high dimensionality on a daily basis. A person will grow tired or become distracted and make mistakes over time. Therefore it is desirable to study the feasibility of replacing a persons inference with that of Machine Learning in order to improve reliability. One-Class Support Vector Machines (SVM) with three different kernels (linear, Gaussian and polynomial) are implemented and tested for Anomaly Detection. Principal Component Analysis is used for dimensionality reduction and autoencoders are used with the intention to increase performance. Standard soft-margin SVMs were used for multi-class classification by utilizing the 1vsAll and 1vs1 approaches with the same kernels as for the one-class SVMs. The results for the one-class SVMs and the multi-class SVM methods are compared against each other within their respective applications but also against the performance of Back-Propagation Neural Networks of varying sizes. One-Class SVMs proved very effective in detecting anomalous samples once both Principal Component Analysis and autoencoders had been applied. Standard SVMs with Principal Component Analysis produced promising classification results. Twin SVMs were researched as an alternative to standard SVMs.
Place, publisher, year, edition, pages
2019. , p. 39
Keywords [en]
AI, Machine Learning, Mechanical, SVM, Neural Networks, Support Vector Machines, Autoencoders, PCA, Principal Component Analysis, Classification, Anomaly Detection
National Category
Robotics
Identifiers
URN: urn:nbn:se:mdh:diva-44325OAI: oai:DiVA.org:mdh-44325DiVA, id: diva2:1327842
External cooperation
Prevas
Subject / course
Computer Science
Presentation
2019-06-10, Kappa, Högskoleplan 1, Västerås, 09:15 (English)
Supervisors
Examiners
2019-09-182019-06-202019-09-18Bibliographically approved