Streaming Feature Selection for Multi-Label Data with Dynamic Sliding Windows and Feature Repulsion Loss
| العنوان: | Streaming Feature Selection for Multi-Label Data with Dynamic Sliding Windows and Feature Repulsion Loss |
|---|---|
| المؤلفون: | Yu Li, Yusheng Cheng |
| المصدر: | Entropy Volume 21 Issue 12 |
| بيانات النشر: | Multidisciplinary Digital Publishing Institute, 2019. |
| سنة النشر: | 2019 |
| مصطلحات موضوعية: | 0209 industrial biotechnology, feature repulsion loss, Feature vector, Stability (learning theory), General Physics and Astronomy, Feature selection, 02 engineering and technology, Function (mathematics), sliding window, Article, Term (time), 020901 industrial engineering & automation, Feature (computer vision), Sliding window protocol, 0202 electrical engineering, electronic engineering, information engineering, streaming feature selection, 020201 artificial intelligence & image processing, multi-label learning, Algorithm, Statistical hypothesis testing |
| الوصف: | In recent years, there has been a growing interest in the problem of multi-label streaming feature selection with no prior knowledge of the feature space. However, the algorithms proposed to handle this problem seldom consider the group structure of streaming features. Another shortcoming arises from the fact that few studies have addressed atomic feature models, and particularly, few have measured the attraction and repulsion between features. To remedy these shortcomings, we develop the streaming feature selection algorithm with dynamic sliding windows and feature repulsion loss (SF-DSW-FRL). This algorithm is essentially carried out in three consecutive steps. Firstly, within dynamic sliding windows, candidate streaming features that are strongly related to the labels in different feature groups are selected and stored in a fixed sliding window. Then, the interaction between features is measured by a loss function inspired by the mutual repulsion and attraction between atoms in physics. Specifically, one feature attraction term and two feature repulsion terms are constructed and combined to create the feature repulsion loss function. Finally, for the fixed sliding window, the best feature subset is selected according to this loss function. The effectiveness of the proposed algorithm is demonstrated through experiments on several multi-label datasets, statistical hypothesis testing, and stability analysis. |
| وصف الملف: | application/pdf |
| اللغة: | English |
| تدمد: | 1099-4300 |
| DOI: | 10.3390/e21121151 |
| URL الوصول: | https://explore.openaire.eu/search/publication?articleId=doi_dedup___::7a25f24049a5d61391028b46fe41a353 |
| حقوق: | OPEN |
| رقم الأكسشن: | edsair.doi.dedup.....7a25f24049a5d61391028b46fe41a353 |
| قاعدة البيانات: | OpenAIRE |
| تدمد: | 10994300 |
|---|---|
| DOI: | 10.3390/e21121151 |