Predictive Analytics / Data Mining

These flashcards cover key concepts and definitions related to Predictive Analytics and Data Mining, including applications, errors in prediction models, and specific uses in industries like insurance and customer management.

What is Predictive Analytics?

Predictive Analytics aims to determine what is likely to happen in the future by analyzing past data.

What is Data Mining?

Data mining is the process of finding anomalies, patterns, and correlations within large data sets to predict outcomes.

What are the main applications of Predictive Analytics?

Applications include forecasting future events, customer relationship management, financial predictions, and fraud detection.

What are the types of patterns found in Data Mining?

The types of patterns include association, prediction, cluster (segmentation), and sequential relationships.

What is the difference between Supervised and Unsupervised Learning?

Supervised learning requires labeled data to train models, whereas unsupervised learning identifies patterns in data without pre-labeled outcomes.

What is a Type I error in prediction models?

A Type I error is a false positive, where it falsely infers the existence of something that is not there.

What is a Type II error in prediction models?

A Type II error is a false negative, where it falsely infers the absence of something that is present.

How does predictive modeling assist in Student Retention?

Predictive modeling helps in identifying at-risk students and enabling targeted interventions to improve student retention rates.

What are the benefits of using data in Customer Relationship Management?

Benefits include maximizing return on marketing campaigns, improving customer retention, and identifying valuable customers for targeted marketing.

How can Predictive Analytics be applied in the insurance industry?

It can forecast claim costs, optimize rate plans, and identify fraud, improving business planning and customer service.