Flight Delay Prediction Model Based on Lightweight Network ECA-MobileNetV3
In exploring the flight delay problem, traditional deep learning algorithms suffer from low accuracy and extreme computational complexity; therefore, the deep flight delay prediction algorithm is difficult to directly deploy to the mobile terminal. In this paper, a flight delay prediction model base...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-03-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/12/6/1434 |
| _version_ | 1797612207202107392 |
|---|---|
| author | Jingyi Qu Bo Chen Chang Liu Jinfeng Wang |
| author_facet | Jingyi Qu Bo Chen Chang Liu Jinfeng Wang |
| author_sort | Jingyi Qu |
| collection | DOAJ |
| description | In exploring the flight delay problem, traditional deep learning algorithms suffer from low accuracy and extreme computational complexity; therefore, the deep flight delay prediction algorithm is difficult to directly deploy to the mobile terminal. In this paper, a flight delay prediction model based on the lightweight network ECA-MobileNetV3 algorithm is proposed. The algorithm first preprocesses the data with real flight information and weather information. Then, in order to increase the accuracy of the model without increasing the computational complexity too much, feature extraction is performed using the lightweight ECA-MobileNetV3 algorithm with the addition of the Efficient Channel Attention mechanism. Finally, the flight delay classification prediction level is output via a Softmax classifier. In the experiments of single airport and airport cluster datasets, the optimal accuracy of the ECA-MobileNetV3 algorithm is 98.97% and 96.81%, the number of parameters is 0.33 million and 0.55 million, and the computational volume is 32.80 million and 60.44 million, respectively, which are better than the performance of the MobileNetV3 algorithm under the same conditions. The improved model can achieve a better balance between accuracy and computational complexity, which is more conducive mobility. |
| first_indexed | 2024-03-11T06:38:00Z |
| format | Article |
| id | doaj.art-a2196a4289b643d9835bee236258f497 |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-11T06:38:00Z |
| publishDate | 2023-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-a2196a4289b643d9835bee236258f4972023-11-17T10:45:29ZengMDPI AGElectronics2079-92922023-03-01126143410.3390/electronics12061434Flight Delay Prediction Model Based on Lightweight Network ECA-MobileNetV3Jingyi Qu0Bo Chen1Chang Liu2Jinfeng Wang3Tianjin Key Laboratory of Advanced Signal Processing, Civil Aviation University of China, Tianjin 300300, ChinaTianjin Key Laboratory of Advanced Signal Processing, Civil Aviation University of China, Tianjin 300300, ChinaTianjin Key Laboratory of Advanced Signal Processing, Civil Aviation University of China, Tianjin 300300, ChinaTianjin Key Laboratory of Advanced Signal Processing, Civil Aviation University of China, Tianjin 300300, ChinaIn exploring the flight delay problem, traditional deep learning algorithms suffer from low accuracy and extreme computational complexity; therefore, the deep flight delay prediction algorithm is difficult to directly deploy to the mobile terminal. In this paper, a flight delay prediction model based on the lightweight network ECA-MobileNetV3 algorithm is proposed. The algorithm first preprocesses the data with real flight information and weather information. Then, in order to increase the accuracy of the model without increasing the computational complexity too much, feature extraction is performed using the lightweight ECA-MobileNetV3 algorithm with the addition of the Efficient Channel Attention mechanism. Finally, the flight delay classification prediction level is output via a Softmax classifier. In the experiments of single airport and airport cluster datasets, the optimal accuracy of the ECA-MobileNetV3 algorithm is 98.97% and 96.81%, the number of parameters is 0.33 million and 0.55 million, and the computational volume is 32.80 million and 60.44 million, respectively, which are better than the performance of the MobileNetV3 algorithm under the same conditions. The improved model can achieve a better balance between accuracy and computational complexity, which is more conducive mobility.https://www.mdpi.com/2079-9292/12/6/1434delay prediction modellightweight neural networklightweight attention mechanism |
| spellingShingle | Jingyi Qu Bo Chen Chang Liu Jinfeng Wang Flight Delay Prediction Model Based on Lightweight Network ECA-MobileNetV3 Electronics delay prediction model lightweight neural network lightweight attention mechanism |
| title | Flight Delay Prediction Model Based on Lightweight Network ECA-MobileNetV3 |
| title_full | Flight Delay Prediction Model Based on Lightweight Network ECA-MobileNetV3 |
| title_fullStr | Flight Delay Prediction Model Based on Lightweight Network ECA-MobileNetV3 |
| title_full_unstemmed | Flight Delay Prediction Model Based on Lightweight Network ECA-MobileNetV3 |
| title_short | Flight Delay Prediction Model Based on Lightweight Network ECA-MobileNetV3 |
| title_sort | flight delay prediction model based on lightweight network eca mobilenetv3 |
| topic | delay prediction model lightweight neural network lightweight attention mechanism |
| url | https://www.mdpi.com/2079-9292/12/6/1434 |
| work_keys_str_mv | AT jingyiqu flightdelaypredictionmodelbasedonlightweightnetworkecamobilenetv3 AT bochen flightdelaypredictionmodelbasedonlightweightnetworkecamobilenetv3 AT changliu flightdelaypredictionmodelbasedonlightweightnetworkecamobilenetv3 AT jinfengwang flightdelaypredictionmodelbasedonlightweightnetworkecamobilenetv3 |