Identifying Potential Customers and Recommending New Products for PMG Bookstores | SEO Tips for Big Data Specialty Exam

Identifying Potential Customers and Recommending New Products

Question

PMG Group Malaysia is a Chinese group of companies best known for its book retailing and online retailing services as well as being involved in the printing, publishing and supply of books and library services in China and Taiwan.

PMG Bookstores currently has 65 outlets in China and 6 in Taiwan. The management team has been strengthened to improve its customer service and its range of books.

Steps have been taken to upgrade the computer system to improve the efficiency of PMG Bookstores' inventory control and customer service delivery.

PMG Bookstores continues to seek choice locations for new outlets in China.

PMG Group hosts their web application to sell the books and improve web sales.

The application is built on AWS running out EC2 and RDS. PMG Group has lot of existing customers.

They want to launch a campaign to identify potential customers and recommend new products based on the customer's interest that can upscale the business.

Please advise.

select 3 options.

Answers

Explanations

Click on the arrows to vote for the correct answer

A. B. C. D. E. F. G.

Answer : B,E, F.

Option A is incorrect -ML models for binary classification problems predict a binary outcome.

https://docs.aws.amazon.com/machine-learning/latest/dg/types-of-ml-models.html

Option B is correct - 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.html

Option 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.html

Option D is incorrect -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.html

Option 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.html

Option F is correct -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.html

Option 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.html

The business problem that PMG Group is facing is to identify potential customers and recommend new products based on their interests to upscale the business. To solve this problem, PMG Group can use Amazon Machine Learning (ML), which is a cloud-based service that enables businesses to create predictive models using large amounts of data.

There are several algorithms that can be used in Amazon ML, such as logistic regression, multi-nominal logistic regression, and linear regression. Logistic regression is a binary classification algorithm that predicts a binary outcome, while multi-nominal logistic regression is a multi-class classification algorithm that predicts a multi-class outcome. Linear regression is a regression model that predicts a continuous outcome.

In this case, since the business problem is to recommend new products based on the customer's interest, the appropriate algorithm to use is multi-nominal logistic regression. This algorithm will classify the customer's interests into multiple categories and recommend products accordingly.

To evaluate the performance of the model, Amazon ML uses different metrics such as Area Under the Receiver Operating Characteristic Curve (AUC), macro-average F1 score, and standard root mean square error (RMSE). AUC is a commonly used metric for binary classification problems that measures the model's ability to distinguish between positive and negative classes. F1 score is a measure of the model's accuracy that considers both precision and recall. RMSE is a measure of the difference between predicted values and actual values.

In this case, since the problem is a multi-class classification problem, the appropriate metric to use is the macro-average F1 score. This metric will provide an overall measure of the model's accuracy in predicting the customer's interests and recommending appropriate products.

To ensure that the model is not overfitting the data, PMG Group can use cross-validation, which is a technique for evaluating ML models by training several ML models on subsets of the available input data. This technique will help to detect overfitting and ensure that the model can generalize the pattern to new data.

Therefore, the three options that are relevant to this business problem are:

B. Amazon ML uses multi-nominal logistic regression algorithm through multi-class classification to solve the business problem F. Amazon ML uses macro-average F1 score to provide accuracy of the model 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.