Causal discovery and fault diagnosis based on mixed data types for system reliability modeling
Abstract Causal relationships play an irreplaceable role in revealing the mechanisms of phenomena and guiding intervention actions. However, due to limitations in existing frameworks regarding model representations and learning algorithms, only a few studies have explored causal discovery on non-Euc...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2025-01-01
|
| Series: | Complex & Intelligent Systems |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s40747-024-01740-5 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832571208180170752 |
|---|---|
| author | Xiaokang Wang Siqi Jiang Xinghan Li Mozhu Wang |
| author_facet | Xiaokang Wang Siqi Jiang Xinghan Li Mozhu Wang |
| author_sort | Xiaokang Wang |
| collection | DOAJ |
| description | Abstract Causal relationships play an irreplaceable role in revealing the mechanisms of phenomena and guiding intervention actions. However, due to limitations in existing frameworks regarding model representations and learning algorithms, only a few studies have explored causal discovery on non-Euclidean data. In this paper, we address the issue by proposing a causal mapping process based on coordinate representations for heterogeneous non-Euclidean data. We propose a data generation mechanism between the parent nodes and the child nodes and create a causal mechanism based on multi-dimensional tensor regression. Furthermore, within the aforementioned theoretical framework, we propose a two-stage causal discovery approach based on regularized generalized canonical correlation analysis. Using the discrete representation in the shared projection direction, causal relationships between heterogeneous non-Euclidean variables can be discovered more accurately. Finally, empirical research is conducted on real-world industrial sensor data, which demonstrates the effectiveness of the proposed method for discovering causal relationships in heterogeneous non-Euclidean data. |
| format | Article |
| id | doaj-art-f494cb1cf97a4855b71ccf4432e142a2 |
| institution | Kabale University |
| issn | 2199-4536 2198-6053 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Springer |
| record_format | Article |
| series | Complex & Intelligent Systems |
| spelling | doaj-art-f494cb1cf97a4855b71ccf4432e142a22025-02-02T12:50:07ZengSpringerComplex & Intelligent Systems2199-45362198-60532025-01-0111111610.1007/s40747-024-01740-5Causal discovery and fault diagnosis based on mixed data types for system reliability modelingXiaokang Wang0Siqi Jiang1Xinghan Li2Mozhu Wang3School of Economics and Management, Beijing University of Posts and TelecommunicationsSchool of Economics and Management, Beijing University of Posts and TelecommunicationsInternational School, Beijing University of Posts and TelecommunicationsSchool of Economics and Management, Beijing University of Posts and TelecommunicationsAbstract Causal relationships play an irreplaceable role in revealing the mechanisms of phenomena and guiding intervention actions. However, due to limitations in existing frameworks regarding model representations and learning algorithms, only a few studies have explored causal discovery on non-Euclidean data. In this paper, we address the issue by proposing a causal mapping process based on coordinate representations for heterogeneous non-Euclidean data. We propose a data generation mechanism between the parent nodes and the child nodes and create a causal mechanism based on multi-dimensional tensor regression. Furthermore, within the aforementioned theoretical framework, we propose a two-stage causal discovery approach based on regularized generalized canonical correlation analysis. Using the discrete representation in the shared projection direction, causal relationships between heterogeneous non-Euclidean variables can be discovered more accurately. Finally, empirical research is conducted on real-world industrial sensor data, which demonstrates the effectiveness of the proposed method for discovering causal relationships in heterogeneous non-Euclidean data.https://doi.org/10.1007/s40747-024-01740-5Causal discoveryNon-Euclidean dataCanonical correlation analysisIndustrial fault diagnosis |
| spellingShingle | Xiaokang Wang Siqi Jiang Xinghan Li Mozhu Wang Causal discovery and fault diagnosis based on mixed data types for system reliability modeling Complex & Intelligent Systems Causal discovery Non-Euclidean data Canonical correlation analysis Industrial fault diagnosis |
| title | Causal discovery and fault diagnosis based on mixed data types for system reliability modeling |
| title_full | Causal discovery and fault diagnosis based on mixed data types for system reliability modeling |
| title_fullStr | Causal discovery and fault diagnosis based on mixed data types for system reliability modeling |
| title_full_unstemmed | Causal discovery and fault diagnosis based on mixed data types for system reliability modeling |
| title_short | Causal discovery and fault diagnosis based on mixed data types for system reliability modeling |
| title_sort | causal discovery and fault diagnosis based on mixed data types for system reliability modeling |
| topic | Causal discovery Non-Euclidean data Canonical correlation analysis Industrial fault diagnosis |
| url | https://doi.org/10.1007/s40747-024-01740-5 |
| work_keys_str_mv | AT xiaokangwang causaldiscoveryandfaultdiagnosisbasedonmixeddatatypesforsystemreliabilitymodeling AT siqijiang causaldiscoveryandfaultdiagnosisbasedonmixeddatatypesforsystemreliabilitymodeling AT xinghanli causaldiscoveryandfaultdiagnosisbasedonmixeddatatypesforsystemreliabilitymodeling AT mozhuwang causaldiscoveryandfaultdiagnosisbasedonmixeddatatypesforsystemreliabilitymodeling |