Institutional Repository of Vilnius University
Anomalijų aptikimo algoritmų analizė Ramano spektroskopijos duomenims /
Abstract
dc:descriptionThere is a problem in the analysis of spectroscopic data - sometimes, there may be out-of-context anomalies in the data, which can mean some unknown material, a corrupted spectra or some other atypical situation. Searching for such atypical data is called anomaly detection. The aim of the work is to analyze and compare the results of 4 traditional anomaly detection algorithms and to try to apply the Kernel Density Estimation technique to a specific set of data obtained by Raman spectroscopy. A set of spectra of honey from various plants was selected for analysis. After applying the selected algorithms to this set, the results obtained showed that when examining this set, the best results among traditional algorithms were obtained when using the One-Class Support Vector Machine algorithm. The conducted study also shows that it is possible to obtain some fairly good results when applying Kernel Density Estimation to an anomaly detection problem.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2021
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Zinkevičius, Mindaugas,
- Contributors dc:contributor
-
- Raila, Tomas
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD88378519&prefLang=en_US