EFA for Structure Detection in Image Data : Empirical Results on Two Datasets of Different Perspective
Structure detection discovery from image data is scarce. Hence, we attempt to explore and uncover the underlying structure from two datasets of different perspective through statistical procedures commonly used in psychology, social science, health and business. Firstly, distinction between princip...
| Main Authors: | , , , |
|---|---|
| Format: | Proceeding |
| Language: | English |
| Published: |
2015
|
| Subjects: | |
| Online Access: | http://ir.unimas.my/id/eprint/13443/ http://ir.unimas.my/id/eprint/13443/1/EFA%20for%20Structure%20Detection%20in%20Image%20Data%20%28abstract%29.pdf |
| Summary: | Structure detection discovery from image data is
scarce. Hence, we attempt to explore and uncover the underlying structure from two datasets of different perspective through statistical procedures commonly used in psychology, social science, health and business. Firstly, distinction between principal component analysis and exploratory factor analysis are briefly described; along with a simple test on the growth of publications on both techniques and datasets tested in this paper. Exploratory factor analyses results with and without data screening are summarized. 3-factor structures are derived from both datasets where texture features seem to be dominant than
others. Some critical issues concerning the appropriateness of methods are also discussed. The systematic procedures described in this paper are applicable to any other object type with similar characteristics as the ones tested. |
|---|