Aperçu des sections
Généralités
Communication and Availability :
Instructor: Dr.Allaoui.
Contact: soumaya.allaoui@univ-annaba.dz.
Reaserchgate: https://www.researchgate.net/profile/Soumaya-Allaoui
Linkedin: https://www.linkedin.com/in/soumaya-allaoui-55060119a/
In-Person Availability: Sunday and Monday from 8 AM to 12 PM at Department of Financial Sciences.
In Case of Emergency:
• Email: For access issues to the platform or other emergencies, please send me an email. I willrespond within 24 hours, except in cases of unforeseen circumstances.Other Situations: If you do not receive a response within the specified timeframe or if a phone conversation is needed,please contact the department’s secretary or your group leader.Plan of the first part: (Factorial Analysis Techniques) :
1- Introduction to the Methods Define each method (CPA, CA, MCA).
2-Explain their main applications in data science and statistics.
3-Theoretical Foundations.
6. Summary and Choosing the Right Method When to use CPA vs. CA vs. CAM? Summary table with key differences.
To get an idea: Principal Component Analysis
PCA : Used for dimensionality reduction on continuous numerical data. Identifies principal components to explain variance in a dataset.
Correspondence Factor Analysis
Mathematical Differences:
-CPA: relies on eigenvalue decomposition of the covariance matrix.
CA: uses the chi-square distance and singular value decomposition (SVD) to analyze categorical data.
CAM: extends CA to more than two categorical variables, using the Burt matrix.
4. Practical Applications in Data Science
CPA: Used in feature extraction, image compression, and financial modeling.
CA: Used for market segmentation, text mining, and social sciences.
CAM: Used in survey analysis, customer behavior studies, and recommendation systems.