A review of novelty detection
Novelty detection is the task of classifying test data that differ in some respect from the data that are available during training. This may be seen as "one-class classification", in which a model is constructed to describe "normal" training data. The novelty detection approach...
| Главные авторы: | , , , |
|---|---|
| Формат: | Journal article |
| Опубликовано: |
Elsevier
2014
|
| _version_ | 1826285774649163776 |
|---|---|
| author | Pimentel, M Clifton, D Clifton, L Tarassenko, L |
| author_facet | Pimentel, M Clifton, D Clifton, L Tarassenko, L |
| author_sort | Pimentel, M |
| collection | OXFORD |
| description | Novelty detection is the task of classifying test data that differ in some respect from the data that are available during training. This may be seen as "one-class classification", in which a model is constructed to describe "normal" training data. The novelty detection approach is typically used when the quantity of available "abnormal" data is insufficient to construct explicit models for non-normal classes. Application includes inference in datasets from critical systems, where the quantity of available normal data is very large, such that "normality" may be accurately modelled. In this review we aim to provide an updated and structured investigation of novelty detection research papers that have appeared in the machine learning literature during the last decade. © 2014 Published by Elsevier B.V. |
| first_indexed | 2024-03-07T01:33:52Z |
| format | Journal article |
| id | oxford-uuid:947eb5ef-ea04-4156-840a-ae957d35d4f6 |
| institution | University of Oxford |
| last_indexed | 2024-03-07T01:33:52Z |
| publishDate | 2014 |
| publisher | Elsevier |
| record_format | dspace |
| spelling | oxford-uuid:947eb5ef-ea04-4156-840a-ae957d35d4f62022-03-26T23:39:46ZA review of novelty detectionJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:947eb5ef-ea04-4156-840a-ae957d35d4f6Symplectic Elements at OxfordElsevier2014Pimentel, MClifton, DClifton, LTarassenko, LNovelty detection is the task of classifying test data that differ in some respect from the data that are available during training. This may be seen as "one-class classification", in which a model is constructed to describe "normal" training data. The novelty detection approach is typically used when the quantity of available "abnormal" data is insufficient to construct explicit models for non-normal classes. Application includes inference in datasets from critical systems, where the quantity of available normal data is very large, such that "normality" may be accurately modelled. In this review we aim to provide an updated and structured investigation of novelty detection research papers that have appeared in the machine learning literature during the last decade. © 2014 Published by Elsevier B.V. |
| spellingShingle | Pimentel, M Clifton, D Clifton, L Tarassenko, L A review of novelty detection |
| title | A review of novelty detection |
| title_full | A review of novelty detection |
| title_fullStr | A review of novelty detection |
| title_full_unstemmed | A review of novelty detection |
| title_short | A review of novelty detection |
| title_sort | review of novelty detection |
| work_keys_str_mv | AT pimentelm areviewofnoveltydetection AT cliftond areviewofnoveltydetection AT cliftonl areviewofnoveltydetection AT tarassenkol areviewofnoveltydetection AT pimentelm reviewofnoveltydetection AT cliftond reviewofnoveltydetection AT cliftonl reviewofnoveltydetection AT tarassenkol reviewofnoveltydetection |