Online anomaly detection in case of limited feedback with accurate distribution learning

Simple item page
dc.contributor.authorMarivani, Imanen_US
dc.contributor.authorKari, Dariushen_US
dc.contributor.authorKurt, Ali Emirhanen_US
dc.contributor.authorManış, Erenen_US
dc.coverage.spatialAntalya, Turkeyen_US
dc.date.accessioned2018-04-12T11:44:37Z
dc.date.available2018-04-12T11:44:37Z
dc.date.issued2017en_US
dc.departmentDepartment of Electrical and Electronics Engineeringen_US
dc.descriptionDate of Conference: 15-18 May 2017en_US
dc.descriptionConference Name: IEEE 25th Signal Processing and Communications Applications Conference, SIU 2017en_US
dc.description.abstractWe propose a high-performance algorithm for sequential anomaly detection. The proposed algorithm sequentially runs over data streams, accurately estimates the nominal distribution using exponential family and then declares an anomaly when the assigned likelihood of the current observation is less than a threshold. We use the estimated nominal distribution to assign a likelihood to the current observation and employ limited feedback from the end user to adjust the threshold. The high performance of our algorithm is due to accurate estimation of the nominal distribution, where we achieve this by preventing anomalous data to corrupt the update process. Our method is generic in the sense that it can operate successfully over a wide range of data distributions. We demonstrate the performance of our algorithm with respect to the state-of-the-art over time varying distributions.en_US
dc.identifier.doi10.1109/SIU.2017.7960595en_US
dc.identifier.urihttp://hdl.handle.net/11693/37583
dc.language.isoTurkishen_US
dc.publisherIEEEen_US
dc.relation.isversionofhttp://dx.doi.org/10.1109/SIU.2017.7960595en_US
dc.source.titleProceedings of the IEEE 25th Signal Processing and Communications Applications Conference, SIU 2017en_US
dc.subjectAnomaly detectionen_US
dc.subjectExponential familyen_US
dc.subjectLikelihood assignmenten_US
dc.subjectLimited feedbacken_US
dc.subjectOnline learningen_US
dc.subjectE-learningen_US
dc.subjectSignal detectionen_US
dc.subjectSignal processingen_US
dc.titleOnline anomaly detection in case of limited feedback with accurate distribution learningen_US
dc.title.alternativeHatasız dağılım öğrenme ile sınırlı geri besleme durumunda çevrimiçi anomali sezimien_US
dc.typeConference Paperen_US
Files
Collections
Department of Electrical and Electronics Engineering