Exam AI-900 topic 1 question 5 discussion

N/Y/N ?

even i feel answer shld be N-Y-N

Third case: No (with some doubt) At first, I thought it was easy because it looked like an Microsoft Learn example, but there is a difference. MS example: "A health clinic might use the CHARACTERISTICS of a patient (such as age, weight, blood pressure, and so on) to predict whether the patient is at risk of diabetes." This case: "Predicting whether a patient will develop diabetes based on the patient's MEDICAL HISTORY..." In the MS example you have data from one moment, in this case you have ongoing changing data in time. I think they use some recent data for the prediction, and not a change in data, but I'm still not 100% sure... https://docs.microsoft.com/en-us/learn/modules/create-classification-model-azure-machine-learning-designer/introduction

I found the video useful in this article https://www.anodot.com/blog/what-is-anomaly-detection/#:~:text=Anomaly%20detection%20(aka%20outlier%20analysis,a%20change%20in%20consumer%20behavior. which already helps me to decide that C has to be YES because it talks about two things: First it says: " a patient WILL develop " which means it's not like classification examples we have from microsoft before "Does he has Diabeties or not?". Second point is " patient's medical HISTORY" which completely shows that it might be regression or Anomaly detection, but I prefered Anomaly detection to regression because regression should refers to an integer results, while Anomaly detection will referes to abnoraml increasing of sugar in blood which will allows DRs to warning the patient in advance

forecasting = recursive algorithm

NYN is the answer. https://learn.microsoft.com/en-us/azure/cognitive-services/anomaly-detector/overview Anomaly Detector is an AI service with a set of APIs, which enables you to monitor and detect anomalies in your time series data with little machine learning (ML) knowledge, either batch validation or real-time inference.

for me given answers are correct. The right algorithm for these 3 scenarios should be regression, anomaly detection and classification

Why shouldn´t be the first one Yes?? You can use it to find anomalyes in the prizing if the predicted value is very diferent from the real one

No Yes No A - Regression B - Anomly C - Cassification

2nd Option is correct. 1st is an example of Regression and the 3rd is an example of Classification.

No Yes No

N/Y/N Anomaly Detection = Keyword is find Regression = Predict Classification = Label, Organize

N/Y/N is correct

No Yes No

NO YES NO

Agreed N/Y/N (Regression/Time series ??, Anomaly, Classification

Answer is NYN