Improving the Applicability of Variational Deep Embedding in Unsupervised Large-Scale Data Clustering
The purpose of the thesis is to apply deep clustering (DC) on King'splayer segmentation. To that end we propose six crucial properties a DCneeds to meet in the context of big data applicability. We implement our method based on VaDE (Variational Deep Embedding) together with four improvements t...
Main Author: | |
---|---|
Format: | Others |
Language: | English |
Published: |
Uppsala universitet, Institutionen för informationsteknologi
2020
|
Subjects: | |
Online Access: | http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-439300 |
id |
ndltd-UPSALLA1-oai-DiVA.org-uu-439300 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-UPSALLA1-oai-DiVA.org-uu-4393002021-04-01T05:27:57ZImproving the Applicability of Variational Deep Embedding in Unsupervised Large-Scale Data ClusteringengZhu, WenfeiUppsala universitet, Institutionen för informationsteknologi2020Engineering and TechnologyTeknik och teknologierThe purpose of the thesis is to apply deep clustering (DC) on King'splayer segmentation. To that end we propose six crucial properties a DCneeds to meet in the context of big data applicability. We implement our method based on VaDE (Variational Deep Embedding) together with four improvements to meet the six criteria, the method is called S3VaDE, a simple, stable and scalable VaDE. The experiments investigate the accuracy, stability and scalability between S3VaDE and VaDE on three benchmark datasets. The results show that S3VaDE out performed state-ofthe-art. In the thesis, we also demonstrate how to do model selection by visualizing latent space. We then apply S3VaDE on King's dataset and interpret the clusters with three KPIs, player engagement, skill leveland monetization. The analysis shows that the clusters are balanced and interpretable. The investigation further shows that the model is stableduring fine-tune. Student thesisinfo:eu-repo/semantics/bachelorThesistexthttp://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-439300IT ; 20086application/pdfinfo:eu-repo/semantics/openAccess |
collection |
NDLTD |
language |
English |
format |
Others
|
sources |
NDLTD |
topic |
Engineering and Technology Teknik och teknologier |
spellingShingle |
Engineering and Technology Teknik och teknologier Zhu, Wenfei Improving the Applicability of Variational Deep Embedding in Unsupervised Large-Scale Data Clustering |
description |
The purpose of the thesis is to apply deep clustering (DC) on King'splayer segmentation. To that end we propose six crucial properties a DCneeds to meet in the context of big data applicability. We implement our method based on VaDE (Variational Deep Embedding) together with four improvements to meet the six criteria, the method is called S3VaDE, a simple, stable and scalable VaDE. The experiments investigate the accuracy, stability and scalability between S3VaDE and VaDE on three benchmark datasets. The results show that S3VaDE out performed state-ofthe-art. In the thesis, we also demonstrate how to do model selection by visualizing latent space. We then apply S3VaDE on King's dataset and interpret the clusters with three KPIs, player engagement, skill leveland monetization. The analysis shows that the clusters are balanced and interpretable. The investigation further shows that the model is stableduring fine-tune. |
author |
Zhu, Wenfei |
author_facet |
Zhu, Wenfei |
author_sort |
Zhu, Wenfei |
title |
Improving the Applicability of Variational Deep Embedding in Unsupervised Large-Scale Data Clustering |
title_short |
Improving the Applicability of Variational Deep Embedding in Unsupervised Large-Scale Data Clustering |
title_full |
Improving the Applicability of Variational Deep Embedding in Unsupervised Large-Scale Data Clustering |
title_fullStr |
Improving the Applicability of Variational Deep Embedding in Unsupervised Large-Scale Data Clustering |
title_full_unstemmed |
Improving the Applicability of Variational Deep Embedding in Unsupervised Large-Scale Data Clustering |
title_sort |
improving the applicability of variational deep embedding in unsupervised large-scale data clustering |
publisher |
Uppsala universitet, Institutionen för informationsteknologi |
publishDate |
2020 |
url |
http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-439300 |
work_keys_str_mv |
AT zhuwenfei improvingtheapplicabilityofvariationaldeepembeddinginunsupervisedlargescaledataclustering |
_version_ |
1719395037279682560 |