Data Analytics Graduate Certificate

Graduate Certificate Program in Data Analytics

The Graduate Certificate Program in Data Analytics is a twelve-graduate-credit sequence that graduate students can take instead of or in combination with a master's program at Kutztown University, such as the Master of Computer Science or MBA. This program prepares students for practical, application-oriented work in data science in the industry or  in academia.

Many modern academic and commercial endeavors apply the techniques of data mining and predictive analytics to their data sets. Students who complete this certificate program will learn concepts, methods, and software tools for locating and obtaining data of interest, for preparing data for semi-automated analysis, for interacting with software tools in analyzing data for patterns, for visualizing structural and dynamic patterns in data, and for designing systems that respond to patterns in data. Successful completion of the program requires a capstone independent study course or internship, in addition to the required courses. After successful completion of program coursework, students will be able to demonstrate the following knowledge and skills:

1. Demonstrate the ability to locate useful massive data sources, download targeted data, evaluate it for integrity and completeness, and format it for use in data analysis software tools.
2. Filter, organize, and manage prepared data for use in the analysis.
3. Construct formal analyses of data using multiple machine learning mechanisms and processes, including standard approaches from the following list.
        a. Data association rules
        b. Data classification trees
        c. Statistical regression techniques
        d. Data clustering techniques
        e. Instance-based machine learning
        f. Ensemble machine learning
        g. Neural networks
4. Create meaningful graphical visualizations of patterns in data relationships.
5. Evaluate the integrity and completeness of the results of data analyses.
6. Communicate the results of analyses using written, graphical, and oral communication
techniques.
7. Integrate the results of analyses into automated mechanisms for responding to patterns in
data.

Students can attend classes in person or via interactive, streaming classroom presentations at class times.

Master's candidates must register for the Certificate Program in addition to registering for their MS programs. Certificate Program students who are not pursuing a Master's degree are not available for federal financial tuition aid.

The courses in the Data Analytics Certificate can be applied to two Master’s Degrees, MS in Computer Science and MBA. The following pages will provide additional information on these programs:

• MS in Computer Science/Software Development
• MS in Computer Science/Information Technology
• MS in Computer Science/Interdisciplinary
• MBA

Requirements for the Certificate Program:

   1.  CSC458: Data Mining and Predictive Analytics I
   2. CSC459: Introduction to Big Data
   3. CSC558: Data Mining and Predictive Analytics II
   4. Select one of the following as approved by the student's program advisor.
            - CSC523: Advanced Scripting for Data Manipulation, Analysis, and Machine Learning
            - CSC 570 - Independent Study and/or Projects in Computer Science
            - CSC 590 - CS Cooperative Internship

Course Descriptions for the Courses in the Certificate Program:

CSC 458 - Data Mining and Predictive Analytics I (3 sh)
Many academic and commercial endeavors apply the techniques of data mining and predictive analytics to their data sets. Students taking this course will learn methods and software tools for locating and obtaining data of interest, for preparing data for semi-automated analysis, for interacting with software tools in analyzing data for patterns, for visualizing structural and dynamic patterns in data, and for designing systems that respond to patterns in data. Data cleaning and formatting requires some programming in a modern scripting language. Other course activities include learning to use off-the-shelf software tools to accomplish the tasks of data analysis.

CSC 459 - Introduction to Big Data (3 sh)
This course explores key data management and analysis techniques, which deal with massive datasets to enable real-time decision-making in distributed environments, business intelligence in the Web, and scientific discovery in a large scale. In particular, map-reduce parallel computing paradigms and associated technologies, such as distributed file systems, noSQL databases, and basic machine learning methods, will be explored.

CSC 558 - Data Mining and Predictive Analytics II (3 sh)
This course covers advanced study and practice in data mining and predictive analytics. Topics include understanding, configuring, and applying advanced variants of data association, classification, clustering, and statistical analysis engines, analyzing and applying underlying machine learning algorithms, exploring instance-based, support vector, time-series, ensemble, graphical, and lazy learning algorithms, meta-learning, neural nets, genetic algorithms, and validating results. The course examines topics specific to very large data sets. Data cleaning and formatting require some programming in a modern scripting language. Other course activities include using, extending, and customizing off-the-shelf machine learning software systems to accomplish the tasks of data analysis.

CSC 523 - Advanced Scripting for Data Manipulation, Analysis, and Machine Learning (3 sh)
This course covers advanced study and practice in using a modern scripting language to integrate off-the-shelf code libraries for the retrieval of unstructured and partially structured data, and for the cleaning, integration, formatting, storage, analysis, and visualization of large data sets. Modern scripting languages include powerful built-in features for storing, retrieving, mapping, and integrating data; code libraries extend such features greatly. Libraries include those for regular-expression based extraction of textual data, data integration, statistical analysis and correlation, machine learning, natural language processing, machine vision and listening visualization, and storage in files and database systems. Emphasis is on using a scripting language to glue together off-the-shelf library modules without writing the complex, underlying library code.