Seri Machine Learning | ROC vs AUC adalah... - #90 

Sama2... Resume link video 17 algorithma Machine Learning: A. Supervised Learning Terbagi dalam 2 model: 1. Regression * Linear (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-3V_mdvRx5Zc.html) * Logistic (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-MqZvStQrKAc.html) * Polynomial (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-GdXy6qEPmVo.html) 2. Classification * K-Nearest Neighbors (KNN) (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-0WpK5g4EOq4.html) * Decision Tree (DT) (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-WVvLK6RwIoY.html) * Naive Bayes (NB) (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-ih9D9B1vz1c.html) * Support Vector Machine (SVM) (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-nE-2uG7RU28.html) B. Unsupervised Learning Terbagi dalam 3 model & 2 metode (ML & DL): A. Machine Learning 1. Clustering * K-Means (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-vlouPyMa1AA.html) * Hierarchical Clustering (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-t_9WXQVC-q0.html) * T-SNE Clustering (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-2sWGlS8Wnic.html) * DBScan (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-gKQu31ZTB9g.html) 2. Dimension Reduction * Principal Component Analysis (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-LoEtynX4NRk.html) * Anomaly Detection (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-Hgl2wUTjxMI.html) * Auto-Encoder (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-Y1xaiZpvR_c.html) * Hebbian Learning (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-lQ3Tm5P7228.html) B. Deep Learning Generative Models * Generative Adversarial Network (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-vHdoOOCGr0I.html) * Self Organizing Maps (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-iVLixLtNYbk.html) Selamat mencoba & praktek ✌

Silahkan, ini materi pengantar dari 17 algoritma machine learning: A. Supervised Learning Terbagi dalam 2 model: 1. Regression * Linear (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-3V_mdvRx5Zc.html) * Logistic (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-MqZvStQrKAc.html) * Polynomial (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-GdXy6qEPmVo.html) 2. Classification * K-Nearest Neighbors (KNN) (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-0WpK5g4EOq4.html) * Decision Tree (DT) (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-WVvLK6RwIoY.html) * Naive Bayes (NB) (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-ih9D9B1vz1c.html) * Support Vector Machine (SVM) (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-nE-2uG7RU28.html) B. Unsupervised Learning Terbagi dalam 3 model & 2 metode (ML & DL): A. Machine Learning 1. Clustering * K-Means (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-vlouPyMa1AA.html) * Hierarchical Clustering (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-t_9WXQVC-q0.html) * T-SNE Clustering (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-2sWGlS8Wnic.html) * DBScan (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-gKQu31ZTB9g.html) 2. Dimension Reduction * Principal Component Analysis (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-LoEtynX4NRk.html) * Anomaly Detection (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-Hgl2wUTjxMI.html) * Auto-Encoder (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-Y1xaiZpvR_c.html) * Hebbian Learning (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-lQ3Tm5P7228.html) B. Deep Learning Generative Models * Generative Adversarial Network (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-vHdoOOCGr0I.html) * Self Organizing Maps (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-iVLixLtNYbk.html) Selamat mencoba & praktek ✌

Ini tabel dimana kah? Dari hasil eksekusi di SPSS? Pertanyaannya rada teori banget ya, tapi gpp. Menurut ane, isi konten tabel itu menunjukkan nilai Sensitivity & Specificity apakah maksimal atau tidak karena akan terkait dengan tingkat akurasi (Confusing Matrix). Semoga sedikit mencerahkan... 👌

Logikanya sih bs aja ya

Sekedar teori pengantar sebelum masuk ke materi teknis 👉 ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-MqZvStQrKAc.html

Resume link video 17 algorithma Machine Learning: A. Supervised Learning Terbagi dalam 2 model: 1. Regression * Linear (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-3V_mdvRx5Zc.html) * Logistic (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-MqZvStQrKAc.html) * Polynomial (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-GdXy6qEPmVo.html) 2. Classification * K-Nearest Neighbors (KNN) (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-0WpK5g4EOq4.html) * Decision Tree (DT) (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-WVvLK6RwIoY.html) * Naive Bayes (NB) (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-ih9D9B1vz1c.html) * Support Vector Machine (SVM) (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-nE-2uG7RU28.html) B. Unsupervised Learning Terbagi dalam 3 model & 2 metode (ML & DL): A. Machine Learning 1. Clustering * K-Means (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-vlouPyMa1AA.html) * Hierarchical Clustering (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-t_9WXQVC-q0.html) * T-SNE Clustering (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-2sWGlS8Wnic.html) * DBScan (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-gKQu31ZTB9g.html) 2. Dimension Reduction * Principal Component Analysis (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-LoEtynX4NRk.html) * Anomaly Detection (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-Hgl2wUTjxMI.html) * Auto-Encoder (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-Y1xaiZpvR_c.html) * Hebbian Learning (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-lQ3Tm5P7228.html) B. Deep Learning Generative Models * Generative Adversarial Network (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-vHdoOOCGr0I.html) * Self Organizing Maps (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-iVLixLtNYbk.html) Selamat mencoba & praktek ✌

@@EkoWahyudiharto Terimakasih ilmunya 🙏

Sama2,@@mfaruqrm . Semoga sedikit mencerahkan...

Sama2 merupakan perhitungan pengukuran perbedaan performansi dari metode yang digunakan dengan kriteria yang ditentukan, bedanya OPTIMISTIC untuk menghasilkan nilai yang TIDAK SESUAI & PESIMISTIC untuk nilai yang SESUAI. Umumnya optimistik berada di area kiri grafik, berlawanan arah dengan pesimistik. Semoga sedikit mencerahkan 🫰

@@EkoWahyudiharto mohon pencerahannya pak🙏, misal saya punya nilai AUC optimistic 0,957. kemudian nilai AUC netral nilainya 0,500 dan AUC pessimistic 0,280. berarti untuk menentukan apakah model data bagus atau tidak, dilihat dari AUC yang mana?

Keknya gada yg standardize sih,@@heartdynamite1342 , tapi umumnya bisa dianggap EXCELLENT utk AUC dengan range antara 0.9-1, GOOD utk AUC antara 0.8-0.9, FAIR utk AUC antara 0.7-0.8, POOR utk AUC antara 0.6-0.7 &FAILED utk AUC antara 0.5-0.6.