Detection of Cross-Channel Anomalies
The data deluge has created a great challenge for data mining applications wherein the rare topics of interest are often buried in the flood of major headlines. We identify and formulate a novel problem: cross-channel anomaly detection from multiple data channels. Cross-channel anomalies are common...
| Main Authors: | , , , |
|---|---|
| Format: | Journal Article |
| Published: |
Springer
2012
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/20.500.11937/26770 |
| _version_ | 1848752080886431744 |
|---|---|
| author | Pham, DucSon Saha, Budhaditya Phung, Dinh Venkatesh, Svetha |
| author_facet | Pham, DucSon Saha, Budhaditya Phung, Dinh Venkatesh, Svetha |
| author_sort | Pham, DucSon |
| building | Curtin Institutional Repository |
| collection | Online Access |
| description | The data deluge has created a great challenge for data mining applications wherein the rare topics of interest are often buried in the flood of major headlines. We identify and formulate a novel problem: cross-channel anomaly detection from multiple data channels. Cross-channel anomalies are common amongst the individual channel anomalies, and are often portent of significant events. Central to this new problem is a development of theoretical foundation and methodology. Using the spectral approach, we propose a two-stage detection method: anomaly detection at a single-channel level, followed by the detection of cross-channel anomalies from the amalgamation of single channel anomalies. We also derive the extension of the proposed detection method to an online settings, which automatically adapts to changes in the data over time at low computational complexity using incremental algorithms. Our mathematical analysis shows that our method is likely to reduce the false alarm rate by establishing theoretical results on the reduction of an impurity index. We demonstrate our method in two applications: document understanding with multiple text corpora, and detection of repeated anomalies in large-scale video surveillance. The experimental results consistently demonstrate the superior performance of our method compared with related state-of-art methods, including the one-class SVM and principal component pursuit. In addition, our framework can be deployed in a decentralized manner, lending itself for large scale data stream analysis. |
| first_indexed | 2025-11-14T08:02:56Z |
| format | Journal Article |
| id | curtin-20.500.11937-26770 |
| institution | Curtin University Malaysia |
| institution_category | Local University |
| last_indexed | 2025-11-14T08:02:56Z |
| publishDate | 2012 |
| publisher | Springer |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | curtin-20.500.11937-267702017-09-13T16:08:58Z Detection of Cross-Channel Anomalies Pham, DucSon Saha, Budhaditya Phung, Dinh Venkatesh, Svetha Anomaly detection · Multiple channels · Topic modeling · Residual subspace analysis · Text data analysis · Video surveillance · Data mining · Collaborative subspace learning The data deluge has created a great challenge for data mining applications wherein the rare topics of interest are often buried in the flood of major headlines. We identify and formulate a novel problem: cross-channel anomaly detection from multiple data channels. Cross-channel anomalies are common amongst the individual channel anomalies, and are often portent of significant events. Central to this new problem is a development of theoretical foundation and methodology. Using the spectral approach, we propose a two-stage detection method: anomaly detection at a single-channel level, followed by the detection of cross-channel anomalies from the amalgamation of single channel anomalies. We also derive the extension of the proposed detection method to an online settings, which automatically adapts to changes in the data over time at low computational complexity using incremental algorithms. Our mathematical analysis shows that our method is likely to reduce the false alarm rate by establishing theoretical results on the reduction of an impurity index. We demonstrate our method in two applications: document understanding with multiple text corpora, and detection of repeated anomalies in large-scale video surveillance. The experimental results consistently demonstrate the superior performance of our method compared with related state-of-art methods, including the one-class SVM and principal component pursuit. In addition, our framework can be deployed in a decentralized manner, lending itself for large scale data stream analysis. 2012 Journal Article http://hdl.handle.net/20.500.11937/26770 10.1007/s10115-012-0509-6 Springer restricted |
| spellingShingle | Anomaly detection · Multiple channels · Topic modeling · Residual subspace analysis · Text data analysis · Video surveillance · Data mining · Collaborative subspace learning Pham, DucSon Saha, Budhaditya Phung, Dinh Venkatesh, Svetha Detection of Cross-Channel Anomalies |
| title | Detection of Cross-Channel Anomalies |
| title_full | Detection of Cross-Channel Anomalies |
| title_fullStr | Detection of Cross-Channel Anomalies |
| title_full_unstemmed | Detection of Cross-Channel Anomalies |
| title_short | Detection of Cross-Channel Anomalies |
| title_sort | detection of cross-channel anomalies |
| topic | Anomaly detection · Multiple channels · Topic modeling · Residual subspace analysis · Text data analysis · Video surveillance · Data mining · Collaborative subspace learning |
| url | http://hdl.handle.net/20.500.11937/26770 |