This qualification is designed for Company
Associates with a starting knowledge of IBM SPSS Modeler edition 14 or higher
working in govt., universities, or business that use the IBM SPSS Modeler item
to execute information exploration actions such as information knowing,
information planning and modeling. The IBM Certified Associate may implement
the IBM SPSS Modeler item for programs such as client control, scams
recognition and turn, risk control & so forth.
Certification
Exams
Test
SPS-200: IBMSPSSMPRO
Section 1 - Business Knowledge (5%)
- CRISP-DM process methodology
- Identifying business objectives
- Translating business objectives to data mining goals
Section 2 - Data Knowledge (25%)
- Reading data from various sources - Source Nodes
- Using data visualization - Graph Nodes
- Understanding distributions and summary statistics
- Identifying data quality issues
- Identifying and understanding outliers
- Identifying anomalies - anomaly node
- Understanding relationships among variables
Section 3 - Data Preparation (35%)
- Using the Merge and Append nodes to combine datasets
- Deriving new fields - Fields Pallet Nodes
- Aggregate and restructure datasets
- Using the Select node
- Sampling and Balancing datasets
- Methods for reducing the dimensionality of the dataset
- Understanding SQL pushback
- Understanding use of data caching
- Methods for Missing Value Replacement
Section 4 - Modeling (20%)
·
Partitioning the dataset
·
Understanding which models to use for sets or
binary outcomes
·
Understanding which models to use for numeric
outcomes
·
Understanding model types and basic operations
·
Combining models using the Ensemble node
·
Auto modeling nodes
Section 5 - Evaluation of Results (10%)
·
Using the Analysis node
·
Producing and interpreting Evaluation charts
1.
Using data visualizations (charts) and classification
tables to interpret model results
2.
Interpreting Generated Model Nuggets
Section 6 - Deployment of Results (5%)
- Using the Export Nodes
- Scoring new data using generated models
- Understanding monitoring of deployed models
To be a successful certified as an IBM SPSS Modeler,
Business Partner Data Analyst, Associate, IBM recommend that you have
already finished the following courses or have the equivalent understanding of
the topics covered.
Business Understanding (5%)
- Review of the CRISP methodology
General operations in Modeler (20%)
·
Building Streams
·
Running Streams
Data Understanding (30%)
·
Extent of Missing Data
- Outliers
- Field Distribution and Summary Statistics
Data Preparation (40%)
·
Combining data files
·
Restructuring data
·
Aggregating data
·
Removing Duplicates
·
Sampling cases
Modeling (5%)
·
Predictive Models
·
Cluster Models
·
Association Models
Test
SPS-202: IBMSPSSMBPDM: IBM SPSS Modeler Business Partner Data Mining Associate
Business Understanding (5%)
- Review of the CRISP methodology
General operations in Modeler (20%)
·
Building Streams
·
Running Streams
·
Reading different file types into Modeler
including delimited information, statistics information, spreadsheet
information and databases
Data Understanding (20%)
·
Extent of Missing Data
·
Outliers
·
Field Distribution and Summary Statistics
·
Auto checking for missing and out of bounds data
·
Looking at vicarage relationships between
variables
Data Preparation (30%)
·
Automated Data Preparation
·
Feature Section
Modeling (20%)
·
Auto Classifier
·
Auto Numeric
Deployment (5%)
- Scoring new data with a deployed model
- Exporting Model outcomes
No comments:
Post a Comment