Certification Overview
An Associate Data Scientist has fundamental data science skills and knowledge using IBM Watson X AI to solve business problems with machine learning solutions. This includes the ability to connect machine learning solutions to enterprise requirements and understand when to apply an enterprise AI workflow.
Concepts in this Associate level exam include:
- Problem scoping and tool selection
- Exploratory data analysis
- Feature engineering
- Model training and selection
- Model evaluation
Recommended Skills
- Python
- R
- Descriptive statistics
- Predictive analytics
Requirements
Exam C1000-177: Foundations of Data Science using IBM WatsonX
Exam Objectives
During exam development, the Subject Matter Experts (SMEs) define all of the tasks, knowledge and experience that an individual would need in order to successfully fulfill their role with the product or solution. These are represented by the objectives below and the questions on the exam are based upon this objective.
-
Section 1: Evaluate the Business Problem16%
- Translate business objectives into Data Science/ML/AI solutions
- Formulate the hypothesis to be tested
- Identify appropriate tools for analysis
-
Section 2: Perform Exploratory Data Analysis21%
- Visually examine the data for data understanding
- Assess data characteristics to guide future processing
- Conduct statistical analysis of data
- Visualize data to identify patterns/trends
- Deselect features that have minimal predictive value
-
Section 3: Development Tools and Techniques13%
- Assess which modeling and statistical techniques are best suited
- Select the appropriate environment and libraries
-
Section 4: Pre-Processing and Feature Engineering33%
- Integrate data from different sources and formats
- Normalize data
- Mitigate imbalanced data
- Handle data anomalies and missing values
- Identify the Best Categorical Data Encoding Techniques
- Transform Features
- Select Relevant Features
-
Section 5: Model Selection, Training, Evaluation, and Presentation17%
- Identify an adequate Machine Learning Model
- Split the data to support model evaluation
- Choose appropriate model metrics to assess model performance
Reviews
There are no reviews yet.