Hierarchical Clustering of Time Series using Gaussian Mixture Models and Variational Autoencoders
(2019) In Master's Theses in Mathematical Sciences FMSM01 20191Mathematical Statistics
- Abstract
- This thesis proposes a hierarchical clustering algorithm for time series, comprised of a variational autoencoder to compress the series and a Gaussian mixture model to merge them into an appropriate cluster hierarchy. This approach is motivated by the autoencoders good results in dimensionality reduction tasks and by the likelihood framework given by the Gaussian mixture model. In contrast to similar clustering algorithms, this algorithm tries to answer the question of the true number of clusters in the data set and gives superior visualisation possibilities. Furthermore, the thesis shows how cluster analysis in general can be applied to several interesting problems in finance, and specifically how this algorithm is engineered to... (More)
- This thesis proposes a hierarchical clustering algorithm for time series, comprised of a variational autoencoder to compress the series and a Gaussian mixture model to merge them into an appropriate cluster hierarchy. This approach is motivated by the autoencoders good results in dimensionality reduction tasks and by the likelihood framework given by the Gaussian mixture model. In contrast to similar clustering algorithms, this algorithm tries to answer the question of the true number of clusters in the data set and gives superior visualisation possibilities. Furthermore, the thesis shows how cluster analysis in general can be applied to several interesting problems in finance, and specifically how this algorithm is engineered to outperform standard clustering algorithms on these problems. (Less)
- Popular Abstract
- Machine learning is rapidly changing the financial industry. This thesis contributes by clustering financial assets using artificial neural networks. The goal is to enhance analysis and predict the future.
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/8982213
- author
- Wilhelmsson, Per LU
- supervisor
- organization
- alternative title
- Hierarkisk klustring av tidsserier med Gaussiska blandmodeller och variationsautoencoder
- course
- FMSM01 20191
- year
- 2019
- type
- H2 - Master's Degree (Two Years)
- subject
- keywords
- Clustering, Deep Learning, Machine Learning, Time Series, Variational Autoencoders, Gaussian Mixture Models
- publication/series
- Master's Theses in Mathematical Sciences
- report number
- LUTFMS-3367-2019
- ISSN
- 1404-6342
- other publication id
- 2019:E27
- language
- English
- id
- 8982213
- date added to LUP
- 2019-10-08 13:58:50
- date last changed
- 2019-10-08 13:58:50
@misc{8982213, abstract = {{This thesis proposes a hierarchical clustering algorithm for time series, comprised of a variational autoencoder to compress the series and a Gaussian mixture model to merge them into an appropriate cluster hierarchy. This approach is motivated by the autoencoders good results in dimensionality reduction tasks and by the likelihood framework given by the Gaussian mixture model. In contrast to similar clustering algorithms, this algorithm tries to answer the question of the true number of clusters in the data set and gives superior visualisation possibilities. Furthermore, the thesis shows how cluster analysis in general can be applied to several interesting problems in finance, and specifically how this algorithm is engineered to outperform standard clustering algorithms on these problems.}}, author = {{Wilhelmsson, Per}}, issn = {{1404-6342}}, language = {{eng}}, note = {{Student Paper}}, series = {{Master's Theses in Mathematical Sciences}}, title = {{Hierarchical Clustering of Time Series using Gaussian Mixture Models and Variational Autoencoders}}, year = {{2019}}, }