Identifying Potential Customers for New Products | AFS Banking Group
Question
Allianz Financial Services (AFS) is a banking group offering end-to-end banking and financial solutions in South East Asia through its consumer banking, business banking, Islamic banking, investment finance and stock broking businesses as well as unit trust and asset administration, having served the financial community over the past five decades. AFS being one the largest banks in the region is planning to improve its segment business by launching a campaign to identify potential customers for various new products launched based on their past behavior? AFS is looking for both batch and real-time predictive analytics AFS understands that they are losing customers to competitive banks and a campaign is launched to address customer retention.
AFS has identified a long list of potential customers that have reduced transactions with the bank.
AFS look at ML to advise whether a customer would continue with the bank or not? select 3 options.
Answers
Explanations
Click on the arrows to vote for the correct answer
A. B. C. D. E. F. G.Answer : A,D, E.
Option A is correct - A.
ML models for binary classification problems predict a binary outcome.
https://docs.aws.amazon.com/machine-learning/latest/dg/types-of-ml-models.htmlOption B is incorrect - ML models for multiclass classification problems allow you to generate predictions for multiple classes (predict one of more than two outcomes).
https://docs.aws.amazon.com/machine-learning/latest/dg/types-of-ml-models.htmlOption C is incorrect -ML models for regression problems predict a numeric value.
https://docs.aws.amazon.com/machine-learning/latest/dg/types-of-ml-models.htmlOption D is correct - Amazon ML provides an industry-standard accuracy metric for binary classification models called Area Under the (Receiver Operating Characteristic) Curve (AUC).
https://docs.aws.amazon.com/machine-learning/latest/dg/binary-model-insights.htmlOption E is correct -Cross-validation is a technique for evaluating ML models by training several ML models on subsets of the available input data and evaluating them on the complementary subset of the data.
Use cross- validation to detect overfitting.
https://docs.aws.amazon.com/machine-learning/latest/dg/cross-validation.htmlOption F is incorrect -The macro-average F1 score is used to evaluate the predictive accuracy of a multiclass metric.
https://docs.aws.amazon.com/machine-learning/latest/dg/multiclass-model-insights.htmlOption G is incorrect - For linear regression tasks, Amazon ML uses the industry standard root mean square error (RMSE) metric.
https://docs.aws.amazon.com/machine-learning/latest/dg/regression-model-insights.htmlThe business problem of AFS is customer retention, and they want to identify potential customers for various new products based on their past behavior. AFS is looking for both batch and real-time predictive analytics to solve the business problem. Amazon Machine Learning (ML) can be used to solve the problem.
Let's evaluate each option provided:
A. Amazon ML uses logistic regression algorithm through Binary classification to solve the business problem
Logistic regression is a supervised learning algorithm used for binary classification problems. In this case, the business problem is to predict whether a customer would continue with the bank or not, which is a binary classification problem. Therefore, logistic regression is an appropriate algorithm to use for this problem.
B. Amazon ML uses multi-nominal logistic regression algorithm through multi-class classification to solve the business problem
Multi-nominal logistic regression is a supervised learning algorithm used for multi-class classification problems. In this case, the business problem is to predict whether a customer would continue with the bank or not, which is a binary classification problem. Therefore, multi-nominal logistic regression is not an appropriate algorithm to use for this problem.
C. Amazon ML uses linear regression algorithm through regression model to solve the business problem
Linear regression is a supervised learning algorithm used for regression problems. In this case, the business problem is to predict whether a customer would continue with the bank or not, which is a binary classification problem. Therefore, linear regression is not an appropriate algorithm to use for this problem.
D. Amazon ML uses Area Under the (Receiver Operating Characteristic) Curve (AUC) to provide accuracy of the model
Area Under the Receiver Operating Characteristic Curve (AUC-ROC) is a common metric used to evaluate the performance of binary classification models. It provides a measure of the model's ability to distinguish between positive and negative classes. A higher AUC-ROC score indicates better model performance. Therefore, AUC-ROC is an appropriate metric to use to evaluate the model's performance.
E. Cross-validation is a technique for evaluating ML models by training several ML models on subsets of the available input data to detect overfitting which eventually fails to generalize the pattern
Cross-validation is a technique used to evaluate the performance of ML models by training and testing the model on different subsets of the data. It helps to detect overfitting, which occurs when a model fits the training data too closely and fails to generalize to new data. Therefore, cross-validation is an appropriate technique to use to evaluate the performance of the model.
F. Amazon ML uses macro-average F1 score to provide accuracy of the model
F1 score is a common metric used to evaluate the performance of binary classification models. The macro-average F1 score calculates the average F1 score for each class and then takes the average of those scores. It provides a measure of the model's ability to balance precision and recall for each class. Therefore, macro-average F1 score is an appropriate metric to use to evaluate the model's performance.
G. Amazon ML uses standard root mean square error (RMSE) metric to provide accuracy of the model.
RMSE is a common metric used to evaluate the performance of regression models. In this case, the business problem is a binary classification problem, not a regression problem. Therefore, RMSE is not an appropriate metric to use to evaluate the performance of the model.
In summary, options A, D, and E are appropriate to solve the business problem of AFS. Option A describes the appropriate algorithm to use for the binary classification problem. Option D describes an appropriate metric to use to evaluate the performance of the model. Option E describes an appropriate technique to use to evaluate the performance of the model.
