Mutual Information vs. Sensitivity Analysis for Bank Telemarketing Feature Selection

A comparative study on arXiv explores two feature selection techniques: mutual information and data-based sensitivity analysis. Applied to a bank telemarketing dataset, both methods identified influential features for predicting contact success. The study found that mutual information performed better with a higher false positive ratio, while sensitivity analysis was superior for lower false positive rates, suggesting mutual information remains a valid, albeit older, method for feature selection. AI

IMPACT This research provides insights into feature selection methods for predictive modeling in business contexts.

RANK_REASON The item is an academic paper published on arXiv discussing machine learning techniques. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• Bank Telemarketing
• logistic regression model
• mutual information
• sensitivity analysis

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →