Detecting Traffic Incidents Using Persistence Diagrams
We introduce a novel methodology for anomaly detection in time-series data. The method uses persistence diagrams and bottleneck distances to identify anomalies. Specifically, we generate multiple predictors by randomly bagging the data (reference bags), then for each data point replacing the data po...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-09-01
|
| Series: | Algorithms |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1999-4893/13/9/222 |
| id |
doaj-4737c01adab8431cbed53356a4482c2d |
|---|---|
| record_format |
Article |
| spelling |
doaj-4737c01adab8431cbed53356a4482c2d2020-11-25T02:53:10ZengMDPI AGAlgorithms1999-48932020-09-011322222210.3390/a13090222Detecting Traffic Incidents Using Persistence DiagramsEric S. Weber0Steven N. Harding1Lee Przybylski2Department of Mathematics, Iowa State University, Ames, IA 50011, USADepartment of Mathematics, Iowa State University, Ames, IA 50011, USADepartment of Mathematics, Iowa State University, Ames, IA 50011, USAWe introduce a novel methodology for anomaly detection in time-series data. The method uses persistence diagrams and bottleneck distances to identify anomalies. Specifically, we generate multiple predictors by randomly bagging the data (reference bags), then for each data point replacing the data point for a randomly chosen point in each bag (modified bags). The predictors then are the set of bottleneck distances for the reference/modified bag pairs. We prove the stability of the predictors as the number of bags increases. We apply our methodology to traffic data and measure the performance for identifying known incidents.https://www.mdpi.com/1999-4893/13/9/222persistence diagrambottleneck distanceanomaly detectionbaggingincident detection |
| collection |
DOAJ |
| language |
English |
| format |
Article |
| sources |
DOAJ |
| author |
Eric S. Weber Steven N. Harding Lee Przybylski |
| spellingShingle |
Eric S. Weber Steven N. Harding Lee Przybylski Detecting Traffic Incidents Using Persistence Diagrams Algorithms persistence diagram bottleneck distance anomaly detection bagging incident detection |
| author_facet |
Eric S. Weber Steven N. Harding Lee Przybylski |
| author_sort |
Eric S. Weber |
| title |
Detecting Traffic Incidents Using Persistence Diagrams |
| title_short |
Detecting Traffic Incidents Using Persistence Diagrams |
| title_full |
Detecting Traffic Incidents Using Persistence Diagrams |
| title_fullStr |
Detecting Traffic Incidents Using Persistence Diagrams |
| title_full_unstemmed |
Detecting Traffic Incidents Using Persistence Diagrams |
| title_sort |
detecting traffic incidents using persistence diagrams |
| publisher |
MDPI AG |
| series |
Algorithms |
| issn |
1999-4893 |
| publishDate |
2020-09-01 |
| description |
We introduce a novel methodology for anomaly detection in time-series data. The method uses persistence diagrams and bottleneck distances to identify anomalies. Specifically, we generate multiple predictors by randomly bagging the data (reference bags), then for each data point replacing the data point for a randomly chosen point in each bag (modified bags). The predictors then are the set of bottleneck distances for the reference/modified bag pairs. We prove the stability of the predictors as the number of bags increases. We apply our methodology to traffic data and measure the performance for identifying known incidents. |
| topic |
persistence diagram bottleneck distance anomaly detection bagging incident detection |
| url |
https://www.mdpi.com/1999-4893/13/9/222 |
| work_keys_str_mv |
AT ericsweber detectingtrafficincidentsusingpersistencediagrams AT stevennharding detectingtrafficincidentsusingpersistencediagrams AT leeprzybylski detectingtrafficincidentsusingpersistencediagrams |
| _version_ |
1724726298599751680 |