SegmentSolve™
Feature Document
|
| ZZZ |
SegmentSolve™
8.0 |
Unique to SegmentSolve™ |
| Types of Data
That Can Be Analyzed |
| Preference |
ü |
ZZZ |
| Performance |
ü |
ZZZ |
| Expectation |
ü |
ZZZ |
| Gap data (between expectation and performance) |
ü |
ZZZ |
| Technical Criteria
Related to Clustering |
| Main Algorithms |
| IK means-simple (Forgy) |
ü |
ZZZ |
| IK means-reflected (Jancey) |
ü |
ü |
| Hierarchical agglomorative |
ü |
ZZZ |
| Q-Factor analysis |
Future |
ZZZ |
| Initialization
methodology (IK means) |
| Distance based |
ü |
ü* |
| Density based |
ü |
ü* |
| Hierarchical single linkage |
ü |
ü* |
| Hierarchical complete cinkage |
ü |
ü* |
| Hierarchical WARDs |
ü |
ü* |
| Agglomeration
methodology |
| Nearest neighbor SLINK |
ü |
ZZZ |
| Farthest neighbor CLINK |
ü |
ZZZ |
| Minimal variance WARDs |
ü |
ZZZ |
| Edit Reporting
Preferences |
| Select sub-set of variables |
ü |
ü |
| Display list of all replications used for analysis |
ü |
ZZZ |
| Weight variables |
ü |
ZZZ |
| Output significance of deviations of grand means |
ü |
ZZZ |
| Customize significance cutoff for reports |
ü |
ZZZ |
| Create report as text file |
ü |
ZZZ |
| Create report as Objective Grid spreadsheet |
ü |
ü |
| Save report as Excel file |
ü |
ZZZ |
| Report customized set of solutions |
ü |
ü |
| User-specified solution and cluster names |
ü |
ZZZ |
| Cluster Analysis
Data Input/Transform Options |
| Select sub-set of variables |
ü |
ZZZ |
| Read data directly from SPSS or text file |
ü |
ü |
| Read data directly from Excel file |
ü |
ü |
| Customized data filtering |
ü |
ü |
| Center variables |
ü |
ZZZ |
| Weight variables |
ü |
ZZZ |
| Ipsatize (remove all scaling effects) |
ü |
ü |
| Standardize variables |
ü |
ZZZ |
| Use N standardization |
ü |
ZZZ |
| Use N-1 standardization |
ü |
ü |
| Standardize variables arithmetically |
ü |
ü |
| Exporting solutions |
| Select cluster solutions to export individually
or collectively |
ü |
ü |
| Export solutions to PrefSolve™ for customer
satisfaction analysis |
ü |
ü |
| Export solutions to SPSS using syntax file |
ü |
ü |
| Export solutions to ascii text file |
ü |
ZZZ |
| Specify cluster and solution names |
ü |
ZZZ |
| Cluster Analysis
Run Options |
| Set minimum and maximum cluster size |
ü |
ZZZ |
| Define and store cluster method sequences |
ü |
ü |
| Select cluster methods (algorithms) and specifications |
ü |
ZZZ |
| Define distance metric style (Euclidian, Squared
Euclidian, and City Block) |
ü |
ü |
| Define variables to be used in analysis |
ü |
ZZZ |
| Use separate input files for respondent data and
variable definitions |
ü |
ZZZ |
| Use single SPSS file for clustering and filtering
variables |
ü |
ZZZ |
| Use Cluster segment methods |
ü |
ZZZ |
| Use CHAID segment methods |
Future |
ZZZ |
| Use Neural segment methods |
Future |
ZZZ |
| Visualization |
| Display cluster solutions with OpenGL 3-D visualization |
ü |
ü |
| Rotation ability for all 3D charts and graphs |
ü |
ü |
| Chart display customization options: Change report
and axis titles, colors, and properties, and change data series display
properties |
ü |
ZZZ |
| Geometry customization options: Change rotation,
spacing, and distance settings |
ü |
ü |
| View 3D relationships of any three attributes
by any replication method by any cluster solution |
ü |
ü |
| Expert System
Best Cluster Identification |
| Automatically identify best cluster method based
on weighting of technical criteria |
ü |
ü |
| Display cluster allocation frequencies |
ü |
ZZZ |
| Customize weight of technical criteria |
ü |
ü |
| Output cluster reproducability |
ü |
ZZZ |
| Output best replication method |
ü |
ZZZ |
| Output cluster means |
ü |
ZZZ |
| Output deviation from grand means |
ü |
ZZZ |
| Output sorted deviation from grand means |
ü |
ZZZ |
| Calculate significance between clusters |
ü |
ZZZ |
| Two group: T-test |
ü |
ZZZ |
| Multiple group: ANOVA |
ü |
ü |
| Identify best methods overall across solutions |
ü |
ü |
| Decision
Support Capabilities |
| Segmentation is automatically characterized against
categorical data |
ü |
ü |
| Significant associations between the clusters
and categorical data are automatically identified |
ü |
ü |
| A best recommended segmentation is automatically
selected. |
ü |
ü |
| Ease
of Use |
| Runs in either advanced or beginner mode via a
selectable wizard |
ü |
ü |
| Configuration files are saved to allow new data
to be segmented identically to a previous segmentation |
ü |
ü |
| Software Platform |
| Visual C++, 32 bit, MFC |
ü |
ZZZ |
| Open GL |
ü |
ZZZ |
| Stingray Objective Grid for Tables |
ü |
ZZZ |
| ZZZ |
| *Comprehensiveness of algorithms used
(SegmentSolve™ uses all) |
