19199

Kendini tekrarlayan derin sinir ağlarının öznitelik seçim yöntemleri ile iyileştirilmesi ve zaman serisi olarak ele alınan otomatik tanımlama sistemi verilerinde kullanımı

Otomatik Tanımlama Sistemi (AIS), deniz taşımacılığının, çarpışma, yangın ve tehlikeli veya kirletici maddelerin dökülmesi gibi risklere sahip olması nedeniyle günümüzde zorunlu hale gelmiş gözlem ve analiz sistemidir. Literatürde, bu tehlikeli durumların önceden tespitinin yapılıp, gemilerin kontrollü ve güvenli seyahatlerini gerçekleştirmeleri için AIS verilerinin kullanıldığı temel matematiksel modellerin, istatistiksel modellerin ve makine öğrenmesi algoritmaların uygulamalarını görebilmekteyiz. Bu çalışmada AIS verileri zaman serileri bakış açısıyla ele alınmış ve geleneksel rota tahminleme modeli yanında; Bütünleşik Otoregresif Hareketli Ortalama, Çok Katmanlı Algılayıcı (ÇKA) ve Kendini Tekrarlayan Derin Sinir Ağları (KT-DSA) ile farklı modeller oluşturularak doğruluk karşılaştırmaları yapılmıştır. Ayrıca ÇKA ve KT-DSA modellerinde, öznitelik seçim tekniklerinden yararlanılarak nitelikler ağırlıklandırılmış ve bu iyileştirilmelerle yeni algoritmalar önerilmiştir. Öznitelik seçimlerinden Relief, Pearson’nun Korelasyonu, Kazanım Oranı ve Bilgi Kazanımı (BK) metotları kullanılmış ve verdikleri rota ve çarpışma tahminlemelerinin doğrulukları karşılaştırılmıştır. Bu doğruluk testlerinde kullanılmak üzere veri seti olarak belirli zamanlara ait Çanakkale Boğazı ve Marmara Denizi AIS verilerinden faydalanılmıştır. Sonuçlara bakıldığında Çanakkale Boğazı’ndaki gemilerin doğrusal bir hareket yapısına sahip olmasından dolayı tüm yaklaşımların birbirine yakın ve yüksek doğruluklara sahip olduğu gözlemlenirken, düzensiz yapısından dolayı Marmara Denizi’nde en iyi sonucu veren yaklaşımın BK ile iyileştirilmiş KT-DSA olduğu sonucuna varılmıştır.

Anahtar Kelimeler:

Büyük veri, derin öğrenme, AIS verisi, öznitelik seçimi, kümeleme

Improvement of recurrent deep neural networks algorithm by feature selection methods and its usage of automatic identification system data evaluated as time series.

Automatic Identification System (AIS) is an observation and analysis system that has become compulsory nowadays due to the risks of maritime transportation such as collision, fire, and spillage of hazardous or polluting substances. In the literature, we can see the applications of basic mathematical models, statistical models and machine learning algorithms using AIS data in order to detect these dangers in advance and to make controlled and safe travel of ships. In this study, AIS data have been evaluated as time series, and accuracy comparisons have been made by being developed different models with Autoregressive Integrated Moving Average, Multilayer Perceptron (MLP) and Deep Recurrent Neural Networks (DRNN) beside traditional route estimation model. In addition, feature selection techniques have been weighted in MLP and RDNN models, and new algorithms have been proposed with these improving. Relief, Pearson's Correlation, Gain Ratio and Information Gain (IG) methods were used to compare the accuracy of the route and collision estimations. In order to be used in these accuracy tests, AIS data related into certain times of Çanakkale Strait and Marmara Sea were used. The results showed that all the approaches were close and high accuracy due to the linear movement of the ships in the Dardanelles. On the other hand, it has been observed that the best approach in the Marmara Sea was the improved DRNN with IG due to its irregular structure.

Keywords:

Big data, deep learning, AIS data, feature selection, clustering,

PDF

___

1. UNCTAD, Review of Maritime Transport, 2016.
2. SOLAS, Safety of Life At Sea Consolidated Edition, 2014.
3. IMO, Revised Guidelines for the Onboard Operational Use of Shipborne Automatic Identification Systems (AIS), 2015.
4. Mustaffa, M., Ahmat, N. H., Ahmad, S., Mapping vessel path of marine traffic density of Port Klang, Malaysia using Automatic Identification System data, International Journal of Science and Research (IJSR), 4 (11), 245-248, 2015.
5. Cimino, G., Ancieri, G., Horn, S., Bryan, K., Sensor data management to achieve information superiority in maritime situational awareness, CMRE Formal Report, NATO Unclassified, 2014.
6. ITU, Technical characteristics for an automatic identification system using time division multiple access in the VHF maritime mobile frequency band, Recommendation ITU-R M.1371-5, 2014.
7. Aarsæther, K. G., Moan, T., Estimating navigation patterns from AIS, Journal of Navigation, 62 (4), 587-607, 2009.
8. Sang, L. Z., Yan, X. P., Wall, A., Wang, J., Mao, Z., CPA calculation method based on AIS position prediction, Journal of Navigation, 69 (6), 1409-1426, 2016.
9. Tang, Q., Gu, D., Day-ahead electricity prices forecasting using artificial neural networks, IEEE International Conference on Artificial Intelligence and Computational Intelligence, 2, 511-514, 2009.
10. Vahidinasab, V., Jadid, S., Kazemi, A., Day-ahead price forecasting in restructured power systems using artificial neural networks, Electric Power Systems Research, 78 (8), 1332-1342, 2008.
11. Zhang, Z. G., Yin, J. C., Wang, N. N., Hui, Z. G., Vessel traffic flow analysis and prediction by an improved PSO-BP mechanism based on AIS data, Evolving Systems, 10 (3), 397-407, 2009.
12. Århus, G. H., Salen, S. R., Predicting shipping freight rate movements using recurrent neural networks and aıs data-on the tanker route between the Arabian Gulf and Singapore, Master's thesis, Norwegian University of Science and Technology, Trondheim, Norway, 2018.
13. Nguyen, Q. V., Extreme weather disaster resilient port and waterway infrastructure for sustainable global supply chain, University of Mississippi, 2017.
14. Xiao, Z., Ponnambalam, L., Fu, X., Zhang, W., Maritime traffic probabilistic forecasting based on vessels’ waterway patterns and motion behaviors, IEEE Transactions on Intelligent Transportation Systems, 18 (11), 3122-3134, 2017.
15. Pallotta, G., Vespe, M., Bryan, K., Vessel pattern knowledge discovery from AIS data: A framework for anomaly detection and route prediction, Entropy, 15 (6), 2218-2245, 2013.
16. Lei, B., A DBSCAN based algorithm for ship spot area detection in AIS trajectory data. MATEC Web of Conferences, EDP Sciences, 291, 2019.
17. Liang, M., Liu, R. W., Zhong, Q., Liu, J., Zhang, J., Neural network-based automatic reconstruction of missing vessel trajectory data, IEEE 4th International Conference on Big Data Analytics, 426-430, 2019.
18. Westerdijk, L., Classifying vessel types based on AIS data, MSc thesis, Vrije University, Amsterdam, Holland, 2019.
19. Zhou, Y., Daamen, W., Vellinga, T., Hoogendoorn, S. P., Ship classification based on ship behavior clustering from AIS data, Ocean Engineering, 175, 176-187, 2019.
20. Lei, P. R., Mining maritime traffic conflict trajectories from a massive AIS data, Knowledge and Information Systems. 1-27, 2019.
21. Hanyang, Z., Xin, S., Zhenguo, Y., Vessel sailing patterns analysis from s-aıs data dased on k-means clustering algorithm, IEEE 4th International Conference on Big Data Analyticsi, 10-13, 2019.
22. Mustaffa, M., Ahmad, S., Ali, A. M. M., Ahmad, N., Jais, M., Hamidi, M., Data mining analysis on Ships collision risk and marine traffic characteristic of Port Klang Malaysia waterways from automatic identification system (AIS), International MultiConference of Engineers and Computer Scientists, 242-246, 2019.
23. Yang, D., Wu, L., Wang, S., Jia, H., Li, K. X., How big data enriches maritime research–a critical review of Automatic Identification System (AIS) data applications, Transport Reviews, 39 (6), 755-773, 2019.
24. Bye, R. J., Almklov, P. G., Normalization of maritime accident data using AIS, Marine Policy, 109, 103675, 2019.
25. Goerlandt, F., Kujala, P., Traffic simulation based ship collision probability modeling, Reliability Engineering & System Safety, 96 (1), 91-107, 2011.
26. Wang, K., Liang, M., Li, Y., Liu, J., Liu, R. W., Maritime traffic data visualization: a brief review, IEEE 4th International Conference on Big Data Analytics, 67-72, 2019.
27. Fujii, M., Hashimoto, H., Taniguchi, Y., Kobayashi, E., Statistical validation of a voyage simulation model for ocean-going ships using satellite AIS data, Journal of Marine Science and Technology, 1-11, 2019.
28. Liu, Y., Song, R., Bucknall, R., Intelligent tracking of moving ships in constrained maritime environments using aıs, Cybernetics and Systems, 50 (6), 539-555, 2019.
29. Liu, Z., Wu, Z., Zheng, Z., A novel framework for regional collision risk identification based on AIS data, Applied Ocean Research, 89, 261-272, 2019.
30. Vadaine, R., Hajduch, G., Garello, R., Fablet, R., A Multi-task Deep Learning Architecture for Maritime Surveillance using AIS Data Streams, Cornell University, New York, USA, 2018.
31. Lechtenberg, S., Siqueira Braga, D. D., Hellingrath, B., Automatic identification system (AIS) data based ship-supply forecasting. In Proceedings of the Hamburg International Conference of Logistics (HICL), 3-24, epubli GmbH, 2019.
32. Hoque, X., & Sharma, S. K.., Ensembled deep learning approach for maritime anomaly detection system, Springer ICETIT 2019, 862-869, 2019.
33. García, S., Luengo, J., Herrera, F., Data preprocessing in data mining, New York: Springer, 59-139, 2015.
34. Packiam, R. M., Prakash, V. S. J., A novel integrated framework based NN modular optimization for efficient analytics on Twitter big data, Springer Information and Communication Technology for Intelligent Systems, 213-224, 2019.
35. Krouska, A., Troussas, C., Virvou, M., The effect of preprocessing techniques on Twitter sentiment analysis, IEEE 7th International Conference on Information, Intelligence, Systems & Applications (IISA), 1-5, 2016.
36. Hassler, A. P., Menasalvas, E., García-García, F. J., Rodríguez-Mañas, L., Holzinger, A., Importance of medical data preprocessing in predictive modeling and risk factor discovery for the frailty syndrome, BMC medical informatics and decision making, 19 (1), 33, 2019.
37. Benhar, H., Idri, A., Fernández-Alemán, J. L., Data preprocessing for decision making in medical informatics: potential and analysis, Springer World Conference on Information Systems and Technologies, 1208-1218, 2018.
38. Tian, C., Hao, Y., Hu, J., A novel wind speed forecasting system based on hybrid data preprocessing and multi-objective optimization, Applied Energy, 231, 301-319, 2018.
39. Xiao, L., Wang, J., Yang, X., Xiao, L., A hybrid model based on data preprocessing for electrical power forecasting, International Journal of Electrical Power & Energy Systems, 64, 311-327, 2015.
40. Zhang, X., Peng, Y., Zhang, C., Wang, B., Are hybrid models integrated with data preprocessing techniques suitable for monthly streamflow forecasting? Some experiment evidences, Journal of Hydrology, 530, 137-152, 2015.
41. Al Shalabi, L., Shaaban, Z., Kasasbeh, B., Data mining: A preprocessing engine, Journal of Computer Science, 2 (9), 735-739, 2006.
42. Dash, M., Liu, H., Feature selection for classification. Intelligent Data Analysis, 1 (1-4), 131-156, 1997.
43. Ben-David, A., Comparison of classification accuracy using Cohen’s Weighted Kappa, Expert Systems with Applications, 34 (2), 825-832, 2008.
44. Babu, M. S., Vijayalakshmi, V., An effective approach for sub-acute Ischemic stroke lesion segmentation by adopting meta-heuristics feature selection technique along with hybrid Naive Bayes and sample-weighted random forest classification, Sensing and Imaging, 20 (1), 7, 2019.
45. Alwidian, J., Hammo, B. H., Obeid, N., WCBA: Weighted classification based on association rules algorithm for breast cancer disease, Applied Soft Computing, 62, 536-549, 2018.
46. Jindal, R., Taneja, S., A novel weighted classification approach using linguistic text mining, Int J Comput Appl, 180 (2), 9-15, 2017.
47. Cao, X., Ge, Y., Li, R., Zhao, J., Jiao, L., Hyperspectral imagery classification with deep metric learning, Neurocomputing, 356, 217-227, 2019.
48. Zhang, L., Chen, H., Hu, Y., Compressive tracking via weighted classification boosted by feature selection, Springer Electronics, Communications and Networks V.,137-145, 2016.
49. Fredstam, M., Johansson, G., Comparing Database Management Systems With Sqlalchemy: A Quantitative Study on Database Management Systems, 2019.
50. De Silva, A. M., Leong, P. H. W., Grammar-Based Feature Generation for Time-Series Prediction, Springer, Berlin, Germany, 2015.
51. Dogan, Y., Birant, D., Kut, A., SOM++: integration of self-organizing map and k-means++ algorithms, Springer International Workshop on Machine Learning and Data Mining in Pattern Recognition, 246-259, 2013.
52. Le-Tien, T., Phung-The, V., Routing and tracking system for mobile vehicles in large area. IEEE 5th International Symposium on Electronic Design, Test & Applications, 297-300, 2010.
53. Montgomery, D. C., Jennings, C. L., Kulahci, M., Introduction to time series analysis and forecasting, John Wiley & Sons, 394-419, 2015.
54. Eswaran, C., Logeswaran, R., An enhanced hybrid method for time series prediction using linear and neural network models, Applied Intelligence, 37 (4), 511-519, 2012.
55. Zhang, Q., Yang, L. T., Chen, Z., Li, P., A survey on deep learning for big data, Information Fusion, 42, 146-157, 2018.
56. Chamorro, J. A., Bermudez, J. D., Happ, P. N., Feitosa, R. Q., A many-to-many fully convolutıonal recurrent network for multıtemporal crop recognition, Isprs Annals of Photogrammetry, Remote Sensing & Spatial Information Sciences, 4, 2019.
57. Abdel-Nasser, M., Mahmoud, K., Accurate photovoltaic power forecasting models using deep LSTM-RNN. Neural Computing and Applications, 31 (7), 2727-2740, 2019.
58. Hochreiter, S., Schmidhuber, J., Long short-term memory, Neural computation, 9 (8), 1735-1780, 1997.