AI can be utilized in conserving species of aquatic life forms. AI helps in maintain sustainability in open sea fishery. AI helps significantly in preventing illegal, unreported, and unregulated (IUU) fishing to coach AI methods, it's required to supply historical labeled data. so that we provided a suggestion to utilize neural artificial networks models trained for a special location to a replacement location supported the similarity between the 2 locations in the research area in Iran. Results show that model relocation can significantly reduce the shortcoming generated from data unavailability for a specific location. Data produced by the sensors are sometimes faulty. the choices supported faulty sensor reading will end in a wrong conclusion but we make much added of kit like sensors, limit switch, proximity switch to present a unique ensemble classifier approach for assessing the standard of sensor data. the bottom classifiers are constructed by random under-sampling of the training data where the sampling process is guided by clustering and make it by validation. The inclusion of cluster-based under-sampling and multi–classifier learning has been shown to enhance the accuracy of quality assessment.
Kies, F., Fakhry Kamel Mohamad, A., De Los Ríos-Escalante, P., Zorriehzahra, J. (2020). Benefits of Artificial Intelligence AI: Iran cases study. Intervento presentato a: 3rd International Conference on Applied Zoology (ICAZ), Pakistan [10.13140/RG.2.2.34102.63043].
Benefits of Artificial Intelligence AI: Iran cases study
Kies, F
;
2020
Abstract
AI can be utilized in conserving species of aquatic life forms. AI helps in maintain sustainability in open sea fishery. AI helps significantly in preventing illegal, unreported, and unregulated (IUU) fishing to coach AI methods, it's required to supply historical labeled data. so that we provided a suggestion to utilize neural artificial networks models trained for a special location to a replacement location supported the similarity between the 2 locations in the research area in Iran. Results show that model relocation can significantly reduce the shortcoming generated from data unavailability for a specific location. Data produced by the sensors are sometimes faulty. the choices supported faulty sensor reading will end in a wrong conclusion but we make much added of kit like sensors, limit switch, proximity switch to present a unique ensemble classifier approach for assessing the standard of sensor data. the bottom classifiers are constructed by random under-sampling of the training data where the sampling process is guided by clustering and make it by validation. The inclusion of cluster-based under-sampling and multi–classifier learning has been shown to enhance the accuracy of quality assessment.File | Dimensione | Formato | |
---|---|---|---|
Dr. KIES F_et al_ ICAZ, 2020_V3.pdf
accesso aperto
Descrizione: Scientifico
Tipologia di allegato:
Publisher’s Version (Version of Record, VoR)
Dimensione
1.89 MB
Formato
Adobe PDF
|
1.89 MB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.