Multidimensional scaling /
Outlines a set of techniques that enable a researcher to discuss the 'hidden structure' of large databases. These techniques use proximities, measures which indicate how similar or different objects are, to find a configuration of points which reflects the structure in the data.
| Основен автор: | Kruskal, Joseph B., 1928- |
|---|---|
| Други автори: | Wish, Myron. |
| Формат: | Електронна книга |
| Език: | English |
| Публикувано: |
Beverly Hills, Calif. :
Sage Publications,
℗♭1978.
|
| Серия: |
Quantitative applications in the social sciences ;
no. 07-011. |
| Предмети: | |
| Онлайн достъп: |
http://search.ebscohost.com/login.aspx?direct=true&scope=site&db=nlebk&AN=24751 |
| Подобни документи: |
Print version::
Multidimensional scaling. |
Съдържание:
- 1. Basic concepts of multidimensional scaling
- Some methods for obtaining proximities data
- Relationship between the proximities and the spatial distances
- Notation and terminology
- Different types of MDS
- Defining an objective function
- Computational procedures
- Looking at the scatter diagram
- 2. Interpretation of the configuration
- Application to perceptions of nations
- Linear regression for dimensional interpretation
- Neighborhood interpretation of MDS configurations
- Embedding other information in the MDS configuration
- 3. Dimensionality
- Goodness-of-fit
- Other considerations concerning dimensionality
- 4. Three-way multidimensional scaling
- The INDSCAL model
- INDSCAL analysis of the nation similarities data
- Fitting the friendship ratings into the similarities space
- INDSCAL analysis of profile proximities
- 5. Preparing the input for multidimensional scaling
- Preprocessing of data
- Metric versus nonmetric MDS, and transformation of the proximities
- Some available computer programs for MDS.