學術產出-Theses
Article View/Open
Publication Export
-
題名 使用正規隨機漫步及相似度進行異常偵測
Anomaly Detection Using Regulated Random Walk and Similarity Degree作者 陳柏龍
Chen, Po-Lung貢獻者 周珮婷
Chou, Pei-Ting
陳柏龍
Chen, Po-Lung關鍵詞 異常偵測
相似度
正規隨機漫步
多尺度
自我調整
Anomaly detection
Similarity
Regulated random walk
Multi-scale
Self-tuning日期 2019 上傳時間 5-Sep-2019 15:42:18 (UTC+8) 摘要 資料雲幾何樹是一個透過正規隨機漫步捕捉資料結構,再進行分群的一個演算法。本論文從資料雲幾何樹的概念中延伸出了兩種異常偵測的方法,第一種是使用樣本間的相似度加總來進行異常偵測,第二種則是透過正規隨機漫步探索數據,以探索到的時間點做為異常值。而在使用多尺度的模擬資料時,發現演算法表現不穩定,因此使用了self-tuning的策略來改良演算法,能克服在資料多尺度時進行異常偵測的問題,最後在實際資料上和經典方法LOF比較。
Data cloud geometry tree is a clustering algorithm that explores data structures by regulated random walk. Based on the concept of data cloud geometry tree, the current study proposes two anomaly detection methods. The first method uses sum of similarities between samples for anomaly detection. The second method explores data through regulated random walk to detect unusual pattern. Samples that were later explored are treated as abnormal. However, the performance of the proposed algorithms are unstable when dealing with multi-scaled simulated data. Therefore, self-tuning strategy is applied to improve the performance of algorithms and to overcome the anomaly detection problem for multi-scaled data. Finally, the performance of proposed methods are compared to the performance resulting from the classical method, LOF, with many real examples.參考文獻 Breunig, M. M., Kriegel, H.-P., Ng, R. T., & Sander, J. (2000). LOF: identifying density-based local outliers. Paper presented at the ACM sigmod record.Chandola, V., Banerjee, A., & Kumar, V. (2009). Anomaly detection: A survey. ACM computing surveys (CSUR), 41(3), 15.Ester, M., Kriegel, H.-P., Sander, J., & Xu, X. (1996). A density-based algorithm for discovering clusters in large spatial databases with noise. Paper presented at the Kdd.Fushing, H., & McAssey, M. P. (2010). Time, temperature, and data cloud geometry. Phys Rev E Stat Nonlin Soft Matter Phys, 82(6 Pt 1), 061110. doi:10.1103/PhysRevE.82.061110Goldstein, M. (2012). FastLOF: An expectation-maximization based local outlier detection algorithm. Paper presented at the Proceedings of the 21st International Conference on Pattern Recognition (ICPR2012).Grubbs, F. E. (1950). Sample criteria for testing outlying observations. The Annals of Mathematical Statistics, 21(1), 27-58.He, Z., Xu, X., & Deng, S. (2003). Discovering cluster-based local outliers. Pattern Recognition Letters, 24(9-10), 1641-1650.Kriegel, H.-P., & Zimek, A. (2008). Angle-based outlier detection in high-dimensional data. Paper presented at the Proceedings of the 14th ACM SIGKDD international conference on Knowledge discovery and data mining.Lazarevic, A., & Kumar, V. (2005). Feature bagging for outlier detection. Paper presented at the Proceedings of the eleventh ACM SIGKDD international conference on Knowledge discovery in data mining.Lee, Y.-J., Yeh, Y.-R., & Wang, Y.-C. F. (2012). Anomaly detection via online oversampling principal component analysis. IEEE Transactions on Knowledge and Data Engineering, 25(7), 1460-1470.Liu, F. T., Ting, K. M., & Zhou, Z.-H. (2008). Isolation forest. Paper presented at the 2008 Eighth IEEE International Conference on Data Mining.Maaten, L. v. d., & Hinton, G. (2008). Visualizing data using t-SNE. Journal of machine learning research, 9(Nov), 2579-2605.Ng, A. Y., Jordan, M. I., & Weiss, Y. (2002). On spectral clustering: Analysis and an algorithm. Paper presented at the Advances in neural information processing systems.Pokrajac, D., Lazarevic, A., & Latecki, L. J. (2007). Incremental local outlier detection for data streams. Paper presented at the 2007 IEEE symposium on computational intelligence and data mining.Sakurada, M., & Yairi, T. (2014). Anomaly detection using autoencoders with nonlinear dimensionality reduction. Paper presented at the Proceedings of the MLSDA 2014 2nd Workshop on Machine Learning for Sensory Data Analysis.Zelnik-Manor, L., & Perona, P. (2005). Self-tuning spectral clustering. Paper presented at the Advances in neural information processing systems.Zenati, H., Foo, C. S., Lecouat, B., Manek, G., & Chandrasekhar, V. R. (2018). Efficient gan-based anomaly detection. arXiv preprint arXiv:1802.06222. 描述 碩士
國立政治大學
統計學系
106354026資料來源 http://thesis.lib.nccu.edu.tw/record/#G0106354026 資料類型 thesis dc.contributor.advisor 周珮婷 zh_TW dc.contributor.advisor Chou, Pei-Ting en_US dc.contributor.author (Authors) 陳柏龍 zh_TW dc.contributor.author (Authors) Chen, Po-Lung en_US dc.creator (作者) 陳柏龍 zh_TW dc.creator (作者) Chen, Po-Lung en_US dc.date (日期) 2019 en_US dc.date.accessioned 5-Sep-2019 15:42:18 (UTC+8) - dc.date.available 5-Sep-2019 15:42:18 (UTC+8) - dc.date.issued (上傳時間) 5-Sep-2019 15:42:18 (UTC+8) - dc.identifier (Other Identifiers) G0106354026 en_US dc.identifier.uri (URI) http://nccur.lib.nccu.edu.tw/handle/140.119/125518 - dc.description (描述) 碩士 zh_TW dc.description (描述) 國立政治大學 zh_TW dc.description (描述) 統計學系 zh_TW dc.description (描述) 106354026 zh_TW dc.description.abstract (摘要) 資料雲幾何樹是一個透過正規隨機漫步捕捉資料結構,再進行分群的一個演算法。本論文從資料雲幾何樹的概念中延伸出了兩種異常偵測的方法,第一種是使用樣本間的相似度加總來進行異常偵測,第二種則是透過正規隨機漫步探索數據,以探索到的時間點做為異常值。而在使用多尺度的模擬資料時,發現演算法表現不穩定,因此使用了self-tuning的策略來改良演算法,能克服在資料多尺度時進行異常偵測的問題,最後在實際資料上和經典方法LOF比較。 zh_TW dc.description.abstract (摘要) Data cloud geometry tree is a clustering algorithm that explores data structures by regulated random walk. Based on the concept of data cloud geometry tree, the current study proposes two anomaly detection methods. The first method uses sum of similarities between samples for anomaly detection. The second method explores data through regulated random walk to detect unusual pattern. Samples that were later explored are treated as abnormal. However, the performance of the proposed algorithms are unstable when dealing with multi-scaled simulated data. Therefore, self-tuning strategy is applied to improve the performance of algorithms and to overcome the anomaly detection problem for multi-scaled data. Finally, the performance of proposed methods are compared to the performance resulting from the classical method, LOF, with many real examples. en_US dc.description.tableofcontents 摘要 iAbstract ii表次 v圖次 vi第一章 緒論 1第二章 文獻探討 2第一節 基於統計 3第二節 基於與鄰近點的距離 3第三節 基於密度 4第四節 基於分群 5第六節 異常偵測的難點 7第七節 總結 7第三章 研究方法 8第一節 資料雲幾何樹(Data Cloud Geometry Tree,DCGT) 8一、定義樣本間的相似度 10二、隨機漫步過程 10三、建立同群機率矩陣 12四、決定分群數量 12五、使用階層式分群進行分群 14第二節 Regulated Random Walk Outlier Factor(RRWOF) 15第三節 Similarity Degree Outlier Factor(SDOF) 15第五節 模擬資料實驗 18一、溫度尺度(T)=1 21二、溫度尺度(T)=10 22三、溫度尺度(T) =100 23四、小結 24第五節 溫度自我調整(self-tuning) 25一、k = 20 26二、k = 100 27三、k = 500 28四、小結 28第四章 研究過程 29第一節 評估準則 29第二節 資料集介紹 32一、APS Failure at S cania Trucks Data Set(APS Failure) 33二、Credit Card Fraud Detection data set(Credit Card) 35三、Epileptic Seizure Recognition Data Set(Epileptic) 37第三節 實驗流程 40第五章 實驗結果及結論 41第一節APS Failure 41第二節Credit Card 43第三節Epileptic 45第四節 小結 47第六章 結論與未來展望 48參考文獻 49 zh_TW dc.format.extent 3882655 bytes - dc.format.mimetype application/pdf - dc.source.uri (資料來源) http://thesis.lib.nccu.edu.tw/record/#G0106354026 en_US dc.subject (關鍵詞) 異常偵測 zh_TW dc.subject (關鍵詞) 相似度 zh_TW dc.subject (關鍵詞) 正規隨機漫步 zh_TW dc.subject (關鍵詞) 多尺度 zh_TW dc.subject (關鍵詞) 自我調整 zh_TW dc.subject (關鍵詞) Anomaly detection en_US dc.subject (關鍵詞) Similarity en_US dc.subject (關鍵詞) Regulated random walk en_US dc.subject (關鍵詞) Multi-scale en_US dc.subject (關鍵詞) Self-tuning en_US dc.title (題名) 使用正規隨機漫步及相似度進行異常偵測 zh_TW dc.title (題名) Anomaly Detection Using Regulated Random Walk and Similarity Degree en_US dc.type (資料類型) thesis en_US dc.relation.reference (參考文獻) Breunig, M. M., Kriegel, H.-P., Ng, R. T., & Sander, J. (2000). LOF: identifying density-based local outliers. Paper presented at the ACM sigmod record.Chandola, V., Banerjee, A., & Kumar, V. (2009). Anomaly detection: A survey. ACM computing surveys (CSUR), 41(3), 15.Ester, M., Kriegel, H.-P., Sander, J., & Xu, X. (1996). A density-based algorithm for discovering clusters in large spatial databases with noise. Paper presented at the Kdd.Fushing, H., & McAssey, M. P. (2010). Time, temperature, and data cloud geometry. Phys Rev E Stat Nonlin Soft Matter Phys, 82(6 Pt 1), 061110. doi:10.1103/PhysRevE.82.061110Goldstein, M. (2012). FastLOF: An expectation-maximization based local outlier detection algorithm. Paper presented at the Proceedings of the 21st International Conference on Pattern Recognition (ICPR2012).Grubbs, F. E. (1950). Sample criteria for testing outlying observations. The Annals of Mathematical Statistics, 21(1), 27-58.He, Z., Xu, X., & Deng, S. (2003). Discovering cluster-based local outliers. Pattern Recognition Letters, 24(9-10), 1641-1650.Kriegel, H.-P., & Zimek, A. (2008). Angle-based outlier detection in high-dimensional data. Paper presented at the Proceedings of the 14th ACM SIGKDD international conference on Knowledge discovery and data mining.Lazarevic, A., & Kumar, V. (2005). Feature bagging for outlier detection. Paper presented at the Proceedings of the eleventh ACM SIGKDD international conference on Knowledge discovery in data mining.Lee, Y.-J., Yeh, Y.-R., & Wang, Y.-C. F. (2012). Anomaly detection via online oversampling principal component analysis. IEEE Transactions on Knowledge and Data Engineering, 25(7), 1460-1470.Liu, F. T., Ting, K. M., & Zhou, Z.-H. (2008). Isolation forest. Paper presented at the 2008 Eighth IEEE International Conference on Data Mining.Maaten, L. v. d., & Hinton, G. (2008). Visualizing data using t-SNE. Journal of machine learning research, 9(Nov), 2579-2605.Ng, A. Y., Jordan, M. I., & Weiss, Y. (2002). On spectral clustering: Analysis and an algorithm. Paper presented at the Advances in neural information processing systems.Pokrajac, D., Lazarevic, A., & Latecki, L. J. (2007). Incremental local outlier detection for data streams. Paper presented at the 2007 IEEE symposium on computational intelligence and data mining.Sakurada, M., & Yairi, T. (2014). Anomaly detection using autoencoders with nonlinear dimensionality reduction. Paper presented at the Proceedings of the MLSDA 2014 2nd Workshop on Machine Learning for Sensory Data Analysis.Zelnik-Manor, L., & Perona, P. (2005). Self-tuning spectral clustering. Paper presented at the Advances in neural information processing systems.Zenati, H., Foo, C. S., Lecouat, B., Manek, G., & Chandrasekhar, V. R. (2018). Efficient gan-based anomaly detection. arXiv preprint arXiv:1802.06222. zh_TW dc.identifier.doi (DOI) 10.6814/NCCU201900895 en_US