You killed part of my enthusiasm and interest for this Data Science program because of your lame videos, it feels like you went back in the early 90's to make them and the voice is so hypnotic that I couldn't keep my focus, I don't think that I learned much here, I just wanted to go through it as fast as possible.

This course is terrible. The general questions are bad and check video recall over understanding, which doesn't aid in the end assignment, that is ridiculous and open ended.

The design of the course was fine. It is great that it was one instructor for the whole course and it was pleasant listening to him. The slides were simple...but the structure of the presentations was good. It was nice seeing the case studies too.

I can't give it five stars though. There are labs, but not everything works properly. If you look at the discussions, you'll see that this has been a problem since the course was developed. It's great that we can see the code, but it's useless if we can't run it. There should definitely be some instructions on that in Week 1 about how to do it!

It would be good if the lab has more explanation on each code, better with a dictionary of the syntaxes.

Data science methodology refers to the process of identifying business problems, collecting relevant data, cleaning and preprocessing the data, analyzing it using statistical and machine learning techniques, and presenting insights to stakeholders.

The data science methodology usually consists of the following steps:

Define the Problem: In this step, the business problem is identified and clearly defined. It is important to understand the problem thoroughly in order to identify the relevant data and analysis techniques.

Collect the Data: In this step, the relevant data is collected from various sources. The data can be structured or unstructured and may require cleaning or preprocessing before analysis.

Data Preparation: This step involves cleaning and preprocessing the data to ensure its quality and consistency. This includes removing missing values, handling outliers, and transforming the data as needed.

Data Analysis: In this step, statistical and machine learning techniques are used to analyze the data and uncover patterns and relationships. This can include techniques such as clustering, regression, and classification.

Model Development: Once the analysis is complete, models are developed to make predictions or classify new data. This can involve building machine learning models such as decision trees, neural networks, or support vector machines.

Model Evaluation: In this step, the models are evaluated using various metrics to determine their accuracy and effectiveness.

Deployment: Finally, the insights and models are presented to stakeholders and deployed in the business process to improve decision making and operations.

Overall, the data science methodology provides a structured approach to solving business problems and extracting insights from data. By following this methodology, data scientists can ensure that their analysis is thorough, accurate, and relevant to the business problem at hand.

Very interesting course. It shed a light on what the structured approach really is. It's worth to pause for a moment with every step of the methodology and think how to apply it in real life. Thanks!

The example should be easier than CHF in the videos like the example of cuisines was. The CHF example was good but it was not self explanatory and it might be hard for some people.

The course tells a story of the foundational stages (methodology) in data science,

Identify appropriate data sources to address a business problem

This was one of my favorite modules so far in the Data Science coursework. It really breaks down the methodology used to acheive the results wanted.

Great course, but it goes over some key concepts very quickly. It wasn't a problem for me because I'm familiar with statistics and I conduct social science research. But for someone who is completely new to these topics, I think this course would lack enough detail in order to be useful.

The material is quite interesting and assignment was challenging too. However, I think that this course would be more effective after we learn some python, sql and AI courses. After that it will be more helpfull to implement theories of methodology, described in this course.

this course should be a little later in the IBM sylalbis

I have completed the 2 courses of the certificate before this "What is Data science?" and "Tools for data science". This course was of lower quality than the two before. There was ambiguity between terms. Some of the content was vague. Overall there were a lot of new terms and sufficient explanations and definitions were not provided. The study content had a lot of typos. The man commenting the videos had bad presenting skills. He just read the words without thinking what he is reading about. I had to think hard about what he is even trying to say. I would suggest to make the study content more coherent and consistent (no ambiguity, no low clarity).

The CHS case was very hard to follow. I feel that with a simpler case, the course would've been easier to understand. The quizzes weren't really all that helpful either and a lot of the terms weren't well explained. There should've been clear definitions of what the different stages of the methodology were. I had a lot of trouble differentiating between the different stages like data preparation and data understanding for example. Overall, I felt I learned very little. Btw, this is not a beginner course... This is like a beginner course from someone who already knows data science.

A good overview of data science methodology, with appropriate emphasis on the fact that it is a continuous process with many repetitions that involves stakeholder feedback, thoughtful planning ahead and constant adjustment.

But, I felt there was too much time and emphasis on the details of the specific examples given, and not enough focus on the actual concepts and methods, which could be better explained and their importance better illustrated. The python lab in particular is a well-made example but not very educational from the student perspective.

Videos are short, but full of complicated terms that are difficult to grasp at once. Too many terminologies not only from the Data Science itself, but also from the chosen example make the concepts even more difficult to remember. On top of that, the slides have many texts that cannot be read, since the narrator talks different sentences than what is written. Since there are no written text about the concepts that we are supposed to remember, I had to go back to the videos many times to find/remember the answers to the questions during the Quizz.

The videos and slides are really aweful, but the Jupyter notebooks were interesting

[Reviewing the entire specialization but points are applicable for each course]

I signed up for the IBM Data Science specialization and I was genuinely excited to start it for some 4-5 weeks (I had a GCP exam coming up). I eventually started the specialization beginning of August `20 and started making my way though it and I was amazed … amazed of how much a pile of bullshit this specialization is. I made it though the first 4 courses and at the end of the SQL for data science I couldn’t take it anymore. Here’s why:

1. First and foremost, the entire specialization (all 4 courses I have taken at least) were full of typos and broken URLs which a lot of other students confirm as well. This does not speak professionalism to me but whatever, lets move on.

2. The in-video quizzes and following tests are simply ridiculous … you are expected to have memorized content word by word rather than understand thing for your own and be able to explain them. Some of the question were so far away from tech courses it is not even funny.

3. The final assignments are a total joke. We are asked to review each other which IMHO is a terrible idea since we are all just starting up. Nothing stops you from giving top marks to a bad assignments and vice versa.

4. We eventually got to the more techy part and even got code snippets and jupyter notebooks to look through but they were still bad. There was no proper order in which information was presented i.e. you would read python and seaborn code in the SQL course’s tasks even though python and matplotlib/seaborn are discussed in the following courses.

5. And my final and biggest problem with this whole specialization is that it all feel like an extended advertisement of this piece-of-dodo tech inbred excuse-of-a-software called IBM cloud. There are constants up-sells here and there how almighty IBM is and how great their cloud and IBM Watson Studio are … they are not. I had to spend 2+ hours fixing problems with jupyter notebooks and their cloud just to complete my assignments which both took me 30ish minutes. They mention open source and even though there are open source equivalents to jupyter they insist using IBM cloud. I kept having the feeling they are more focused on promoting IBM products than actually bringing quality content.

6. Now after finishing the SQL course there was a 1min survey which I gladly filled in basically letting them know their specialization if terrible and is doing more harm than good in my opinion. I even sent them a quick challenge because I do not think IBM maintains this course at all or even reads the reviews. You can see my challenge to IBM here: https://bit.ly/3geOyfb

I was very saddened by the quality of the specialization and the content and was wondering whether I should even try and finish the remaining courses but after reading some reviews on the remaining courses I figured out it was just more of the same. If you are in the same boat I would recommend the kaggle micro-courses which I will focus on starting next week.

In conclusion, I got this whole specialization for free via financial aid and I have to say even though I did not pay a dime I feel I need to be compensated by IBM and refunded real money for torturing myself with their courses.

Hi,

I completed entire program and received the Professional Certificate. On the Coursera link of my certificate "3 weeks of study, 2-3 hours/week average per course" is written. This information is not correct at all, it takes approximately 3 times of that time on average! I informed Coursera about it but no correction was made. It should be corrected with "it takes approximately 19 hours study per course" or "Approx. 10 months to complete Suggested 4 hours/week for the Professional Certificate".

Here is the approximate duration for each course can be found one by one clicking the webpages of the courses in the professional certificate webpage: (*)

Course 1: approximately 9 hours to complete

Course 2: approximately 16 hours to complete

Course 3: approximately 9 hours to complete

Course 4: approximately 22 hours to complete

Course 5: approximately 14 hours to complete

Course 6: approximately 16 hours to complete

Course 7: approximately 16 hours to complete

Course 8: approximately 20 hours to complete

Course 9: approximately 47 hours to complete

This makes in total approximately 169 hours to complete the Professional Certificate. As there are 9 courses, each course takes approximately 19 hours (=169/9) to complete.

(*): https://www.coursera.org/professional-certificates/ibm-data-science?utm_source=gg&utm_medium=sem&campaignid=1876641588&utm_content=10-IBM-Data-Science-US&adgroupid=70740725700&device=c&keyword=ibm%20data%20science%20professional%20certificate%20coursera&matchtype=b&network=g&devicemodel=&adpostion=&creativeid=347453133242&hide_mobile_promo&gclid=Cj0KCQjw0Mb3BRCaARIsAPSNGpWPrZDik6-Ne30To7vg20jGReHOKi4AbvstRfSbFxqA-6ZMrPn1gDAaAiMGEALw_wcB

Unfortunately, this particular course dissonates a lot from the previous ones made by IBM on Coursera. The material is very poor, the narration is very fast (we're not all native English speakers!) and most of the time it doesn't match what we see on the screen. It's completely confusing, it's impossible to aprehend any information on these videos. The study case is far from a good example to be understood, it only makes the classes even more confusing than they are.

The notebooks are of very poor interaction, and even the quizzes and exams are not pedagogic. I really felt very much frustrated with this particular course and I hope no other will be as bad as this one, as I felt I had just wasted time and money doing it. I really felt like I've learned nothing from it. Reading the "IBMOpenSource_FoundationalMethologyforDataScience" 3 times and then making an exam about it would be 10 times more effective learning than wasting hours on this terrible course.

I am really shocked with the lack of quality of this particular course, comparing to the other which are simply amazing. Please, substitute this course ASAP for a good one, because I am sure it is lowering the overall quality perception of anyone who is following the 9 courses to reach the certificate.

Sorry about the honesty, but it was very hard to go through this course. I am still shocked about the difference between this one and the others IBM has offered.

Thanks!

Listen - I am the last person to give something a one-star review. Especially on Coursera where all the courses I did were very good at least. This debacle of a course is my first disappointment with a content offered by this platform.

You can see the drastic drop in quality between the first course in the certificate and this one. Poor videos which offered very little value - I felt like I was sitting on a corporate meeting and listening to a boring PowerPoint presentation.

The whole model is not very helpful to understand those things. Cooking metaphor is not very helpful as it feels very forced. The model has far too many stages to actually be useful for grasping the concepts - especially because the stages are very intertwined.

Poor-quality videos which look like a low-effort PP presentation, boring and monotone voice. Remove this from the certificate or improve this.

The resources for this course could be improved - the Skills Lab environment is not performant and there have been problems with the associated links. The Watson video needs updating to reflect the significant changes IBM have made to the interface. I found the use-case to be jargon-heavy. The overall delivery of content is extremely dry and could be made more engaging. Overall, a disappointing course. I am doing the specialisation in order and frankly I would give all 3 of the modules I've completed a very low rating - the presentation of material is outdated and the resources woeful. I expected more from IBM.

Did not find this course very helpful. Too much priority on the case study in favor of diving through each of the steps in the methodology first.

I would recommending restructuring course to more holistically define each step in the methedology and save the case study for the final 'week' before the assignment.

I found the below link through a google search, and reviewing this was more helpful than the time i spent in the course.

https://www.ibmbigdatahub.com/blog/why-we-need-methodology-data-science

This was one of the important modules, yet the most terribly organised one! The module covered very important topics very superficially? How come? The presentations were horrible and instructor - omg I can't find a right word to say. Now, I want to subscribe to another course because I don't think I master the data science methodology here.

Actually a really interesting topic but unfortunately made quite poorly. Visuals were complicated, had spelling mistakes and often went at 1 slide / 60 seconds which meant you had no idea which part of the wall of text the tutor was reading. The extremely monotone voice relapsed my attention in the first 10 seconds...