A brute force tuning of training length for concept drift
We present a brute-force approach to analyze the concept drift behind time sequence data. This approach, named SELECT, searches for the optimal length of training data to minimize error metrics. In other words, SELECT searches for the start point of a new concept from the input sequence. Unlike many...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2023-01-01
|
| Series: | Frontiers in Physics |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fphy.2022.1016564/full |
| _version_ | 1797951766123249664 |
|---|---|
| author | Takumi Uchida Takumi Uchida Kenichi Yoshida |
| author_facet | Takumi Uchida Takumi Uchida Kenichi Yoshida |
| author_sort | Takumi Uchida |
| collection | DOAJ |
| description | We present a brute-force approach to analyze the concept drift behind time sequence data. This approach, named SELECT, searches for the optimal length of training data to minimize error metrics. In other words, SELECT searches for the start point of a new concept from the input sequence. Unlike many related methods, SELECT does not require a pre-specified error threshold to detect drift. In addition, the visual analysis obtained from SELECT enables us to understand how significant a drift has occurred. We test SELECT on two real-world datasets, stock price and COVID-19 infection data. The experimental results show that SELECT can improve the model performance of both datasets. In addition, the visual analysis shows the points of significant drifts, e.g., Lehman’s collapse in stock price data and the spread of variants in COVID-19 data. These results show the effectiveness of the brute-force approach in analyzing concept drift. |
| first_indexed | 2024-04-10T22:35:48Z |
| format | Article |
| id | doaj.art-08405b3615b74b16bdf066725bc1b232 |
| institution | Directory Open Access Journal |
| issn | 2296-424X |
| language | English |
| last_indexed | 2024-04-10T22:35:48Z |
| publishDate | 2023-01-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Physics |
| spelling | doaj.art-08405b3615b74b16bdf066725bc1b2322023-01-16T13:47:54ZengFrontiers Media S.A.Frontiers in Physics2296-424X2023-01-011010.3389/fphy.2022.10165641016564A brute force tuning of training length for concept driftTakumi Uchida0Takumi Uchida1Kenichi Yoshida2Tsukuba AI Application Support Center Co., Tokyo, JapanBusiness Informatics on Social Science, University of Tsukuba, Tokyo, JapanGraduate School of Business Sciences, University of Tsukuba, Tokyo, JapanWe present a brute-force approach to analyze the concept drift behind time sequence data. This approach, named SELECT, searches for the optimal length of training data to minimize error metrics. In other words, SELECT searches for the start point of a new concept from the input sequence. Unlike many related methods, SELECT does not require a pre-specified error threshold to detect drift. In addition, the visual analysis obtained from SELECT enables us to understand how significant a drift has occurred. We test SELECT on two real-world datasets, stock price and COVID-19 infection data. The experimental results show that SELECT can improve the model performance of both datasets. In addition, the visual analysis shows the points of significant drifts, e.g., Lehman’s collapse in stock price data and the spread of variants in COVID-19 data. These results show the effectiveness of the brute-force approach in analyzing concept drift.https://www.frontiersin.org/articles/10.3389/fphy.2022.1016564/fullconcept driftdrift detectionwindow strategystock price predictionCOVID-19 |
| spellingShingle | Takumi Uchida Takumi Uchida Kenichi Yoshida A brute force tuning of training length for concept drift Frontiers in Physics concept drift drift detection window strategy stock price prediction COVID-19 |
| title | A brute force tuning of training length for concept drift |
| title_full | A brute force tuning of training length for concept drift |
| title_fullStr | A brute force tuning of training length for concept drift |
| title_full_unstemmed | A brute force tuning of training length for concept drift |
| title_short | A brute force tuning of training length for concept drift |
| title_sort | brute force tuning of training length for concept drift |
| topic | concept drift drift detection window strategy stock price prediction COVID-19 |
| url | https://www.frontiersin.org/articles/10.3389/fphy.2022.1016564/full |
| work_keys_str_mv | AT takumiuchida abruteforcetuningoftraininglengthforconceptdrift AT takumiuchida abruteforcetuningoftraininglengthforconceptdrift AT kenichiyoshida abruteforcetuningoftraininglengthforconceptdrift AT takumiuchida bruteforcetuningoftraininglengthforconceptdrift AT takumiuchida bruteforcetuningoftraininglengthforconceptdrift AT kenichiyoshida bruteforcetuningoftraininglengthforconceptdrift |