What will we get under Area of ROC curve here? How it evaluates the model here?
", "messages": [ { "id": "1387301", "postDate": "07/14/2021 04:43:40", "content": "What will we get under Area of ROC curve here? How it evaluates the model here?
", "rawMarkdown": "What will we get under Area of ROC curve here? How it evaluates the model here?", "votes": null }, { "id": "1387927", "postDate": "07/14/2021 14:20:20", "content": "It is one of the most important evaluation metrics for checking any classification model’s performance. The below link will help you to better understand the concept.
\n", "rawMarkdown": "It is one of the most important evaluation metrics for checking any classification model’s performance. The below link will help you to better understand the concept.\n\n[Link 1](https://developers.google.com/machine-learning/crash-course/classification/roc-and-auc)\n[Link 2](https://towardsdatascience.com/understanding-auc-roc-curve-68b2303cc9c5)", "votes": null }, { "id": "1388022", "postDate": "07/14/2021 15:45:08", "content": "Thanks.
\nSubmissions are evaluated on the area under the ROC curve between the predicted probability and the observed target.
\nMy question is what are we exactly doing here with ROC!
Link 2 above has a good paragraph:
\n\"AUC - ROC curve is a performance measurement for the classification problems at various threshold settings. ROC is a probability curve and AUC represents the degree or measure of separability. It tells how much the model is capable of distinguishing between classes. Higher the AUC, the better the model is at predicting 0 classes as 0 and 1 classes as 1. By analogy, the Higher the AUC, the better the model is at distinguishing between patients with the disease and no disease.\"
\nThis is probably one of the better ways of explaining the importance/function of the AUC-ROC curve.
", "rawMarkdown": "[Link 2 ](https://towardsdatascience.com/understanding-auc-roc-curve-68b2303cc9c5)above has a good paragraph:\n\n\"AUC - ROC curve is a performance measurement for the classification problems at various threshold settings. ROC is a probability curve and AUC represents the degree or measure of separability. It tells how much the model is capable of distinguishing between classes. Higher the AUC, the better the model is at predicting 0 classes as 0 and 1 classes as 1. By analogy, the Higher the AUC, the better the model is at distinguishing between patients with the disease and no disease.\"\n\nThis is probably one of the better ways of explaining the importance/function of the AUC-ROC curve.", "votes": null }, { "id": "1388391", "postDate": "07/14/2021 22:32:05", "content": "This discussion might also add some context as well - https://www.kaggle.com/c/rsna-miccai-brain-tumor-radiogenomic-classification/discussion/252900
", "rawMarkdown": "This discussion might also add some context as well - https://www.kaggle.com/c/rsna-miccai-brain-tumor-radiogenomic-classification/discussion/252900", "votes": null } ], "comments": [ { "id": 1387927, "author_name": "sanikamal", "author_url": "", "post_date": "07/14/2021 14:20:20", "content": "It is one of the most important evaluation metrics for checking any classification model’s performance. The below link will help you to better understand the concept.
\n", "votes": null, "replies": [ { "id": 1388022, "author_name": "ashvin864", "author_url": "", "post_date": "07/14/2021 15:45:08", "content": "Thanks.
\nSubmissions are evaluated on the area under the ROC curve between the predicted probability and the observed target.
\nMy question is what are we exactly doing here with ROC!
Link 2 above has a good paragraph:
\n\"AUC - ROC curve is a performance measurement for the classification problems at various threshold settings. ROC is a probability curve and AUC represents the degree or measure of separability. It tells how much the model is capable of distinguishing between classes. Higher the AUC, the better the model is at predicting 0 classes as 0 and 1 classes as 1. By analogy, the Higher the AUC, the better the model is at distinguishing between patients with the disease and no disease.\"
\nThis is probably one of the better ways of explaining the importance/function of the AUC-ROC curve.
", "votes": null, "replies": [] }, { "id": 1388391, "author_name": "elenaeb", "author_url": "", "post_date": "07/14/2021 22:32:05", "content": "This discussion might also add some context as well - https://www.kaggle.com/c/rsna-miccai-brain-tumor-radiogenomic-classification/discussion/252900
", "votes": null, "replies": [] } ] } ], "raw_markdown_by_id": { "1387301": "What will we get under Area of ROC curve here? How it evaluates the model here?", "1387927": "It is one of the most important evaluation metrics for checking any classification model’s performance. The below link will help you to better understand the concept.\n\n[Link 1](https://developers.google.com/machine-learning/crash-course/classification/roc-and-auc)\n[Link 2](https://towardsdatascience.com/understanding-auc-roc-curve-68b2303cc9c5)", "1388022": "Thanks.\nSubmissions are evaluated on the area under the ROC curve between the predicted probability and the observed target.\nMy question is what are we exactly doing here with ROC!", "1388384": "[Link 2 ](https://towardsdatascience.com/understanding-auc-roc-curve-68b2303cc9c5)above has a good paragraph:\n\n\"AUC - ROC curve is a performance measurement for the classification problems at various threshold settings. ROC is a probability curve and AUC represents the degree or measure of separability. It tells how much the model is capable of distinguishing between classes. Higher the AUC, the better the model is at predicting 0 classes as 0 and 1 classes as 1. By analogy, the Higher the AUC, the better the model is at distinguishing between patients with the disease and no disease.\"\n\nThis is probably one of the better ways of explaining the importance/function of the AUC-ROC curve.", "1388391": "This discussion might also add some context as well - https://www.kaggle.com/c/rsna-miccai-brain-tumor-radiogenomic-classification/discussion/252900" }, "source": "meta" }