Data Science Methodology

Data Science Methodology 
Course Overview

Reviewers' Background

We are a group of graduate students who are currently taking a business analytics course and looking to learn more about Data Science Methodology.  We chose this Coursera course because the instructors are IBM data scientists.  We will attempt to give you an unbiased review.

Coursera states that 30% started a new career after completing the course, 32% got a tangible career benefit and 10% received a pay increase or promotion after completing this course.

Course Information

About:

The ability to utilize data within the decision making process has been a select skill that many users fail to employ to its maximum potential.  This can be attributed to a lack of understanding of the problems at hand and on navigating how to apply the data being collected and researched, correctly to the situation.

Basic Information - Beginner Level & approximately 10 hours to complete

Instructors:

   Alex Aklson, Ph.D., Data Scientist  

  Polong Lin, Data Scientist

Objectives:

• Identifying a research problem and developing a business plan to address the issue.
• Collecting and analyzing data in order to apply to the problem being identified.
• Creating data models for analysis to draw conclusions and interpret findings.
• Reviewing and understanding feedback being presented on data analysis, model development and the deployment of solutions. 

Week 1 - In this module it will cover the importance of the methodological process that occurs within data science.  Next, a business understanding of the problem is established in order to begin an analytic approach through research.  Data requirements and collection parameters are set to adequately collect and apply findings. 

   Week 2 - In module two, determines the means of how to understand, prepare, and clean data.  Furthermore, it will convey the importance of the data modeling process step-by-step in order to condone research and obtain information.  Once the modeling process is completed, users will enter the evaluation phase to draw conclusions and apply the information gained to the problem at hand.

   Week 3 - In module three, the impacts of model deployment are reviewed and feedback from the analysis is obtained.  Here, results are recorded and studied to apply to the problem identified in the beginning stages of the process.  Addressing feedback and making necessary adjustments is also essential for future development. 

Course recommendation and experience 

We would recommend this course to anyone who is inexperienced with using data to solve problems and wants to learn more about data science.  The course gives a good overview of the methods data scientists use to apply data to problem solving.  Starting with the basics of defining the problem and deciding what data needs to be gathered.  Going on to how to collect this relevant data and prepare it to be used in a predictive model.  Continuing with producing a model and evaluating the results and finally the course finishes with deploying the model you created for broader use.  All of these steps provide good building blocks for someone who is aspiring to learn more about data science and wants to learn how to apply it to their business problems.  

 

After taking this course someone would have a basic understanding of how to collect and prepare data to create models and solve problems.  The next step would be to take more technical courses to learn how to create these models using data science programs such as Python or Frontline Solver.  Learning these skills would give aspiring data scientists the necessary tools and knowledge to take the next steps from knowing data science methodology to be able to apply this methodology in real world situations. Combining the building blocks taught in this course along with some more technical knowledge would prepare someone to apply data science to their business problems and give them the tools to create their own predictive models.