نوع مقاله : مقاله مهندسی معدن
نویسندگان
1 گروه مهندسی معدن، دانشگاه بیرجند، دانشگاه بیرجند
2 گروه مهندسی معدن، دانشگاه بیرجند، بیرجند
3 گروه زمین شناسی، دانشکده علوم، دانشگاه آزاد واحد میانه
4 گروه زمین شناسی، دانشکده علوم، دانشگاه صنعتی شاهرود
چکیده
کلیدواژه‌ها
موضوعات
عنوان مقاله [English]
نویسندگان [English]
This study applied machine learning algorithms, namely Support Vector Machine (SVM) and Random Forest (RF), to develop a mineral potential map for the Shadan region, situated within the Lut Block and Flysch-Ophiolite Belt of Eastern Iran. The research integrated multiple exploration datasets, including geological, geochemical, satellite imagery, and geomagnetic data, to identify promising areas for mineral exploration, specifically targeting porphyry copper and gold deposits. The performance of the models was evaluated using metrics like Accuracy, Sensitivity, ROC curves, and P-A plots. The SVM model demonstrated superior accuracy, successfully predicting 13% of the study area as high-potential zones for future drilling, which corresponded closely with existing drilling results. These identified target zones were predominantly located in regions with intense tectonic activity and were associated with rock units such as andesite, granite, and granodiorite. The study underscores the effectiveness of the SVM model in accurately delineating mineral-rich areas, providing a valuable basis for future exploration programs.
کلیدواژه‌ها [English]
)