# Data Science Accelerators

Not your typical boot camp!

Not your typical boot camp!

Accelerators

The Data Science accelerators are intense non-credit training courses given daily over 5 weeks. All accelerators include a theoretical component and a significant hands-on component, including in-class problem-solving using popular software packages on real-world data sets. All work is completed in the classroom.

The accelerators are targeted to practitioners with diverse backgrounds who do not wish to pursue a formal degree program. The accelerators cover a wide range of topics, from refresher courses to theoretical and practical aspects of data science.

**Accelerator programs **

(Data Sheet)

Refresher

Data Analytics

Data Engineering

Basic Data Science

**Schedule:** Accelerators are offered:

Mornings: Mon-Fri: 9 a.m. to 12 p.m.

Evenings: Mon-Friday: 6 p.m. to 9 p.m.

**Cost**: $7,500 for a 5-week session. A significant discount is available for early registration.

**Register**: https://executiveeducation.njit.edu/execleadership/

**More information:** contact Ryan Mass, 973-596-3178, ryan.m.mass@njit.edu

In this accelerator students will become familiar with the math and computer science fundamentals needed to continue to a career in Data Science. The material includes basic calculus, linear algebra, statistics and Python programming. The accelerator includes 75 hours of instructor-led training and in-class hands-on work.

**Course Outline**

- Calculus
- Functions
- Derivatives
- Integrals

- Linear Algebra
- Vector spaces
- Linear independence, bases and dimension
- Matrix representation of linear functions
- Matrix algebra and solving linear equations
- Eigenvectors and eigenvalues

- Probability
- Discrete and continuous random variables
- Histograms and probability distributions
- Expectation, variance, covariance
- Normal distributions

- Statistics
- Common statistical tests
- Linear regression
- Methods for curve fitting

- Programming
- Basic programming constructs
- Basic data structures
- Web programming
- Basic database design

- Algorithms
- Sorting and searching, divide and conquer, greedy algorithms
- Complexity
- Graph algorithms
- String algorithms

**Software tools**

- Python
- HTML
- XML
- JSON
- SQL

**Prerequisites**: None

**Outcomes**: Become familiar with math and computer science fundamentals required for a career in Data Science.

**Register**: https://executiveeducation.njit.edu/execleadership/data-science-refresher-accelerator/

This accelerator provides the essential skills required to extract actionable intelligence from data resources. Topics covered include principles of information-retrieval system design and management, and basic tools and analytic techniques to extract, report and visualize information from the data. The course includes 75 hours of instructor-led training and in-class hands-on work.

**Course Outline**

- Data Collection and Organization
- Describing and visualizing data
- Pivot tables
- Summary statistics

- Data preprocessing
- Data cleaning
- Data integration
- Data transformation and normalization
- Data reduction
- Correlation analysis

- Data Structures
- Spatial
- Time series
- Multi-dimensional

- Data Visualization
- Dashboard and scorecard
- What if analysis
- Parameters for reports/analysis
- Graphs

- Big Data Storage
- Data base
- Key-value stores
- Star schema and snowflake
- Data lakes

- Data Mining
- Multi-dimensional modeling
- Data warehousing
- Online Analytic Processing (OLAP)
- Association rule mining
- Clustering

- Supervised Learning
- Decision trees
- Naïve Bayesian methods
- Neural networks
- Support vector machines

**Software tools**

- Java
- WEKA
- Tableau
- Tez
- Hive
- Pig
- SAP

**Prerequisites**: None

**Outcomes**: Acquire the skills and learn the tools needed to perform data analysis, reporting and design visual dashboard solutions.

**Register**: https://executiveeducation.njit.edu/execleadership/data-analytics-accelerator/

This accelerator will focus on the practical application of data collection, storage, organization and processing. Students will learn how to build the big data infrastructure, interfaces and mechanisms needed to provide data to be analyzed by data scientists. In addition to big data infrastructure, course work will also focus on data modeling, integration and pipeline processing. Basic programming skills and working knowledge of data structures is required. The course includes 75 hours of instructor-led training and in-class hands-on work.

**Course Outline:**

- The Big Data Ecosystem
- High performance computing
- Grid and cloud computing
- Mobile computing

- Big Data Storage
- Hadoop
- HDFS

- Big Data Management
- Key-value tables
- Documents
- Graphs

- Databases
- Relational databases
- NoSQL

- Big Data Computing
- MapReduce and Spark
- YARN

- Big Data Workflows
- Tez and Storm
- Oozie

- Big Data Analytics
- Clustering & classification
- Recommendation engines
- Machine learning

- Data Cleansing Techniques
- NA values/format correction
- Encoding & Normalization

- Big Data Visualization
- Challenges
- 2D, multidimensional data
- Hierarchical/network data
- Tableau and PowerBI

- Mahout
- K-means clusters, Bayesian classification

**Software tools**

- Hadoop
- MySQL
- HBase
- MapReduce
- Spark
- Storm
- Pig
- Hive
- Tez
- Oozie
- Java/Python
- PowerBI
- Tableau
- Mahout

**Prerequisites**: Basic programming skills, data structures

**Outcomes**: Ability to prepare big data infrastructure and provide data to be analyzed by data scientists.

**Register**: https://executiveeducation.njit.edu/execleadership/data-engineering-accelerator/

This accelerator covers methods of statistical inference, machine learning, predictive modeling and data visualization, data mining and big data, all of which are key for the daily work of a data scientist. During the course, students will apply methods in Python Scikit-learn library to perform typical data processing, e.g. classification, regression and clustering of data. Basic Python programming skills and working knowledge of data structures and algorithms is required, as is fundamentals of calculus and linear algebra, probability and statistics. The course includes 75 hours of instructor-led training and in-class hands-on work.

**Course Outline:**

- Machine learning basics
- Linear algebra
- Probability and statistics
- Python programming

- Bayesian learning
- Gaussian classification
- Nearest centroids
- Naive Bayes

- Linear classification
- Least squares
- Support vector machines (SVM)
- Logistic regression

- Non-linear classification
- Kernel methods
- Neural networks

- Feature engineering
- Feature selection
- Dimensionality reduction
- Feature learning

- Trees
- Decision trees
- Random forests
- Boosting

- Unsupervised learning
- k-means
- Spectral clustering

- Regression
- Linear regression
- Support vector regression

- Time series
- ARIMA
- Regression
- Long short term memory

- Text processing
- Encoding and classification

**Software tools**

- Python
- Scikit-learn
- Pandas

**Prerequisites**: Basic Python programming, data structures and algorithms, calculus and linear algebra, probability and statistics

**Outcomes**: Able to apply methods in Python Scikit-learn library to perform classification, regression and clustering of data.

**Register**: https://executiveeducation.njit.edu/execleadership/basic-data-science-accelerator/

For more information, contact Ryan Mass, 973-596-3178, ryan.m.mass@njit.edu