A Thorough Review of Boston University’s MS in Applied Data Analytics Program

https://miro.medium.com/max/1200/0*-j_TYLBpMJxIapYC

Original Source Here

A Thorough Review of Boston University’s MS in Applied Data Analytics Program

Before I started this MS program, I was looking for course curricula of different Masters programs and trying to find reviews of other people to understand which program is suitable for me. Now, as I am almost done with my MS, I thought I should write a review to help other learners who are looking for an MS program in Data Science or Analytics.

Before I dive into the MS program, here is my background. I have a Bachelor’s in Civil Engineering and a master’s in Environmental Engineering. So, I am not from a computer science background. I did a six-month-long Bootcamp on programming and did a lot of online courses in Coursera on Python, Applied Data Science, Machine Learning, and Statistics. So, I had a pretty good base to start with even though I did not have a CS background.

I will try to explain in details what you will get if you choose to do this without being biased or judgemental

The MS Program

This is a complete course-based master’s program. But there will be a lot of projects throughout the program. There are a total of eight courses. Six essential courses and 2 electives. Six essentials are Analysis of Algorithms, Foundation of Analytics with R, Data Analysis and Visualization with R, Web Analytics and Mining, Data Mining, and Data Science with Python. There are several choices for electives such as Machine Learning, Big Data Analytics, Advanced Database Management, Designing and Implementing a Data Warehouse, Mathematical Statistics.

I definitely took all the essentials and choose Machine Learning and Big Data Analytics for electives. I will start with the six essentials one by one.

Analysis of Algorithms

This is an intensive course. It works on the time and space complexity of the algorithms and popular algorithms such as searching and sorting algorithms, graph algorithms, dynamic programming, memoization, etc. This is not possible to really master all of this in these seven weeks but it provides a great base to build on. Sometimes all the materials can be overwhelming for a non-CS background student but it is manageable. Specially, we had great facilitators sessions every week that helped with the homework.

Foundation of Analytics with R

This course teaches the basics of R programming from the beginning. If you do not have any R programming knowledge at all, no problem. It starts from the beginning. But it moves very fast. Because I already knew Python very well, I picked up pretty fast. So, if you know another programming language, you are good. This course teaches different libraries for data analysis and data visualization with base R. Later in the course it moves towards statistical topics like probability distribution and sampling methods. This course also touches on regular expressions and text mining.

Data Analysis and Visualization with R

This is a sequel to the previous course. It picks up where the previous course ends and keeps moving with more advanced topics of statistics. This course teaches some more advanced libraries for data analysis, hypothesis testing, and data modeling. After these two courses on R, anyone should feel comfortable performing data analysis and modeling with R.

Web Analytics and Mining

This course covers text mining, web data mining, modeling of text data, data extraction from social media, and generating dashboards with text data. You can see, a lot of work on text data. This used the R programming language most of the time. Using Python was allowed in the first two weeks only.

Data Mining

It feels odd to confess that I got to know for the first time in this course that Machine Learning models fall under data mining. There was almost no coding in this course. It covered some basic building blocks for machine learning and we had to do all the calculations manually. I thought that was very helpful because it taught me a lot of the concepts really well. Even some popular machine learning models were done manually with simple cases of course. It is almost impossible to work on complex datasets manually. After learning with the simple dataset, we used Weka and JMP Pro to work with the bigger and more complex datasets. I have done almost all the things I did in this class before using Python and its libraries without knowing what happens behind the scene. This time I learned that behind the scene part.

Data Science with Python

Knowing Python is necessary to take this course. Because it doesn’t teach python. It teaches Python’s data science libraries such as Numpy, Pandas, Matplotlib, Seaborn, and scikit-Learn. That means Data analysis, Visualization, and Machine Learning using different libraries in Python. It’s an intensive course because it is a lot of material to cover in 6 weeks. This wasn’t hard for me because I learned most of the material before from online courses. But if it is new to a student, it should take a lot of time to grasp each week’s materials.

Machine Learning

As you saw from the previous description that several courses focus on machine learning and data modeling already. So, it is expected that you already know machine learning before coming to this class. Though the name of the course is ‘Machine Learning’ it actually focuses on Neural Networks and Deep Learning. Starts with developing basic neural networks from scratch and then the use of Tensorflow to perform more advanced topics. This course focuses on learning the concepts behind the functions of TensorFlow and Keras and use TensorFlow and Keras to build the deep learning models.

These are all the courses I could talk about. I haven’t taken big data analytics yet. That’s the only course I have left with. So, I cannot talk about that.

Conclusion

I am not talking about the pros and cons here. I only wanted to share what you get in the course in detail. Because looking at the curriculum in the course website may not be as clear. I tried to explain some more here. If you notice, four of the six essential courses cover some or a lot of predictive modeling. So, if that’s what you want, this is great!

Lastly, I should mention that each course demands a lot of time. There are assignments every week, in some courses assignments and quizzes every week. Most courses have a lot of material covered every week. I am currently not working, still, it feels hectic towards the end of the semester. I take it really seriously though. Because I am trying to change my career path. It may require at least 20 hours per week or more of study time depending on your level.

That’s all I could share!

Please feel free to follow me on Twitter.

More Reading

AI/ML

Trending AI/ML Article Identified & Digested via Granola by Ramsey Elbasheer; a Machine-Driven RSS Bot

%d bloggers like this: