IBM Data Science Professional Certificate on Coursera: job ready?
An honest review from a previously certified data student
An honest review from a previously certified data student
Introduction
Last year I took and completed the 9 main courses which make up the IBM Data Science Professional Certificate offered by IBM on Coursera (link to certificate here.
I have been wanting to certify my Data Science skills during the last few years, and thus jumped at the opportunity to take this series of courses when I discovered that IBM put out a massive catalog on the famous e-learning platform, which I found myself using quite a lot for other courses and specializations, some of which I have also reviewed here on Medium.
Coursera’s Data Engineering with GCP Professional Certificate: better content or better marketing?
An honest review from a recently certified data studenttowardsdatascience.com
Overall Experience
Taking the Certificate has been definitely an overall positive experience, gave me a solid conceptual understanding of a good portion of the modern day Python data science stack and allowed me to get my hands dirty with capstone projects which take you away from the more theoretical lectures.
I particularly appreciated the broad and modular approach of the course series, which starts from the theoretical foundation rooted in the data science methodology and then proceeds to guide you to the more practical coding sessions and capstone projects. While at it, the courses also introduce you to using IBM cloud resources such as IBM Watson Studio so that you learn how to integrate IBM products into your workflow.
Another aspect of the Certificate which I appreciated was the fact that its entire curriculum is structured as if it were a project, taking you from little knowledge to project delivery in a smooth manner, never making you feel too stretched or too bored. Thus, I experienced a smooth transition between the various topics and components.
Wanting you to judge for yourself, I have summarized below what to expect if you decide to take IBM’s Data Science Professional Certificate on Coursera, to enable you to take a decision of whether to proceed with the time and investment (29£/month subscription until completion- prices may vary depending on your location)
Structure, Course Topics and Tech Stack
IBM’s Data Science Professional Certificate is structured across 9 courses.
Total indicative duration is 10 months at a pace of 5 hours per week. If you are keen on dedicating constant time to studying and going through the materials, you can definitely cut that to a couple of months, especially if it’s not your first Data Science course series.
The course list is the following:
There are no pre-requisites for following and completing the series, which will enable to work around the following:
Tools: Jupyter / JupyterLab, GitHub, R Studio, and Watson Studio
Libraries: Pandas, NumPy, Matplotlib, Seaborn, Folium, ipython-sql, Scikit-learn, ScipPy, etc.
Projects: random album generator, predict housing prices, best classifier model, battle of neighborhoods
The Pros
Engaging learning experience
I never got bored following the lectures, and this has to be a major advantage, especially if one has the objective of going through the entire set of courses which make up the Professional Certificate.
There is a good mix of video lectures and practice, but also the approach and the messaging with which the materials are conveyed feels very natural and smooth, making you want to jump from lecture to lecture to know more and thus build on the previously seen materials.
The learning experience is of course subjective, but comparing these courses to other ones I took on the topic, I found the delivery of the content more effective in this case, which led me to stick around until the end.
The teachers are all IBM Data Science practitioners, and this heavily influences the focus on practicality of the series, leaving a perhaps purely academic approach to teaching a bit behind.
With online course completion rates being quite low in the e-learning industry, finding courses that motivate you and inspire you to want to know more and to consequentially go through all the materials is quite a rare feat, and for this reason this IBM Professional Certificate definitely deserves praise.
2. Build confidence to get started with personal projects
The Certificate’s highlighted practical approach is best exemplified by the final Capstone project, centered around the use of RESTFUL APIs to fetch geo-location data then used in a clustering algorithm.
Although there is naturally a bit of student handholding, closing the course series with a project does well in terms of leaving the curious student wanting to explore more project work centering around the intersection of the newly acquired data science skills and personal interests, which will inevitably lead to more project-based learning.
In this regard, the Certificate certainly puts the student in a very “hands-on” frame of mind, which is definitely appreciated.
3. Great introduction to more advanced courses
I like to view this course as a great “enabler” for the student to then proceed to deeper training grounds. It teaches you just enough around Python, SQL and the Data Science approach/methodology for you to move forward with your learning.
Especially on Coursera, said learning grounds are very fertile and can lead you down all sorts of interesting paths.
Follow-on courses/specializations I would recommend are:
Machine Learning (by Andrew NG): an evergreen classic and a must (even though it’s MATLAB/Octave based) → For transparency, I am linking my course completion certificate here
IBM AI Engineering Professional Certificate: another series of courses offered by IBM, this time more focused around practical applications of deep learning. Next on my list.
Deep Learning Specialization: offered by DeepLearning.ai, it aims to cover all important aspects of deep learning algorithms and applications, with a sound theoretical approach. General recommendation is to have completed Machine Learning (by Andrew NG) before attempting this one.
4. Great sell for entry level data science positions
Overall, this is a great Certificate for applying to entry level data science/data analytics positions, especially if you are re-skilling and hopping on the data science wave coming from neighboring and related fields.
Showing your credentials demonstrates you have taken an active interest in learning and practicing your skills in order to get your foot in the door as a Junior Data Scientist practitioner who can add value for all kinds of companies and projects.
The problem I have with the course is that the positioning in this regard is a bit confusing.
IBM promises they will enable you to reach “proficiency” in Data Science, which I think is more of a marketing ploy (if not a plain overstatement) than a realistic student expectation, since I would refrain from considering myself as fully proficient in such a broadly-defined professional field.
Hence I would be careful in positioning myself for more advanced data science positions, and would focus more on leveraging such certificate to build a foundational toolkit and introductory knowledge of the field which shall then be complemented by further training and learning on the job.
The Cons
Light on stats and maths
While a highly practical course, I would have definitely appreciated a bit more background information on some of the fundamental building blocks that are so vital to Data Science, such as essential statistical concepts and theorems, as well as linear algebra transformations and derivation.
Instead, the series focuses more on arming the student with an entry-level and out-of-the-box data science toolkit, while sometimes glossing over some theoretical aspects the more inquisitive student may want to understand rather than give for granted.
In a sense, the courses are designed to show you the “What” and not necessarily always the “Why” behind Data Science concepts and methodologies.
While it’s impossible to ask for an all-encompassing Certificate, a few hints at some essential intuition would have definitely been appreciated.
2. Reliance on IBM Cloud Resources
Being IBM the provider of the series, there is a good amount of time dedicated to making use of IBM resources which make part the IBM Cloud product suite, such as Watson studio.
Therefore, the student looking for a more open source approach to learning may be dissatisfied, as the student is dependent on adopting the IBM suite to head towards the Certificate’s completion.
This is a general (marketing?) practice adopted by organizations who set up learning programs open to the public while at the same time using the screen time to have the secondary goal of enhancing product utilization. This aspect is definitely overtaken by the overall great course quality, and therefore this cannot be considered a major fault.
Where IBM might suffer, at least at the time of writing, is that a student may want to pair up learning Data Science skills with other Cloud platforms more widely adopted in the market (GCP, AWS, Azure), to maximize immediate skill-platform applicability.
Final Verdict
All in all, I would still recommend taking this course if you are:
an accomplished BI Analyst looking to take the next steps in your Analytics career
looking to get certified and prepare/apply for an entry level Data Science position
looking for a broad overview of the modern data science stack and project approach, while not getting stuck working only in Jupyter Notebooks.
I would not still recommend taking this course if you are:
expecting to become a proficient Data Scientist after completion of the Coursera’s Professional Certificate
looking for a deep dive on the most advanced machine learning models, especially in the Deep Learning space. More deep learning and neural network-heavy courses are likely to prove way more effective for you.
Access my free Data Science resource checklist here
Join Medium with my referral link - Edoardo Romani
As a Medium member, a portion of your membership fee goes to writers you read, and you get full access to every story…edo-romani1.medium.com