1.False Positive (FP): This occurs when the model predicts a positive outcome (e.g., the presence of a condition) when it’s actually negative (e.g., the condition is not present). In medical terms, this could mean a healthy person being diagnosed with a disease.
False Negative (FN): This happens when the model predicts a negative outcome (e.g., no condition) when it’s actually positive (e.g., the condition is present). In a medical context, this could mean a person with a disease being told they are healthy.
If count of False Positive is greater than count of False Negative, cost / penalty for company is more when False Negative are predicted.
Accuracy: Accuracy is a measure of how many predictions the model got right, both true positives (TP) and true negatives (TN), divided by the total number of predictions. The formula for accuracy is:
Accuracy = (TP + TN) / (TP + TN + FP + FN)
Recall=TP / (TP+FN) measures how well we capture the postives
Precision=TP / (TP+FP) gets a penalty for FP
2. Collaborative Filtering: It’s a technique commonly used in recommendation systems that analyzes user behavior and preferences to make recommendations. It can be user-based or item-based, and it identifies patterns and relationships among users and items to suggest products or content to users based on what similar users have liked or interacted with.
3. k-Nearest-Neighbors (kNN) algorithm: It is a type of instance-based learning where the model makes predictions based on the similarity of new data points to existing data points in the training dataset.
Random Cut Forest (RCF): It is particularly well-suited for high-dimensional data and streaming data scenarios. RCF is based on the concept of a random projection tree and is used for both anomaly detection and data summarization.
Linear Learner algorithm: combines the principles of linear models and gradient-based optimization to make predictions. The
predictor_type parameter accepts one of the following values:
- ‘binary_classifier’: Use this option when you are performing binary classification, where the target…