I think I was a tougher critic of my own project than anyone else was going to be; the rubric for peer-grading was almost entirely about presentation with little emphasis on the data analysis itself. The requirement to use Foursquare's API was a limitation on the possible topics for the project, and Foursquare's documentation of endpoints fails to disclose that some fields such as Rating are the result of proprietary, unusual calculations that are unlikely to correlate strongly with any simple data.

On overall IBM data science professional certificate track: Pros: Content is just good enough, instructors are good. Cons: IBM watson and the platform given to practise on is awful and has terrible performance and reliability issues, most often doesnt work and had an impact on my test deliverables. I personally overcame those issues to some extent with kaggle's and google colab jupyter notebook environments.

the idea and all that was great, but despite doing all the previous courses, it was very difficult for me to do the project ... I felt I had many cheese holes and got stuck many times ... note, I had no experience with DS, programming before, but coming from a finance background ... I would say the entire specialisation is not for a true beginner ... but the course package overall is very strong I think

Overall course is good but as you get closer and closer to the final project in the hints and assignments are not explained very well. I had to go through a lot of back and forth on the dicussion forum to solve problems, often over multiple day.

Good material but instructions are very unclear. The final project instructions were not at all well laid out.

The course is quite chaotic and could have been organised better

A laborious process.

This was for me the last course needed for the IBM Data Science Professional Certificate. To have a capstone really makes sense. However, having to use the Foursquare API is very restrictive and does not leave much room for different data science problems. The data provided by Foursquare for my location are only suited for certain types of problems and I had a lot of difficulty finding a rationale for a meaningful project using the Foursquare data. It would be very helpful if more data sources are explicitly provided to allow also ML projects with supervised learning. I also think that the 20h estimation for completing the final task is ridiculously optimistic. I needed at least 100h (time in total) to complete the final project, whereas in all the previous tasks of the 9 courses comprising the IBM Data Science Professional Certificate I needed less than the estimated time.

On one hand there is some encouragement to experiment and come up with an interesting problem and I actually took some time to come up with something non-trivial before starting. But practically, there is no need to -- you can just take the example notebook from one of the prior weeks and change the city. The evaluation doesn't really reward you for doing something original or extra.

Sadly, the certificate of the specialization as a whole is worthless.

I hoped this course would be more complex but it mostly consist of Foursquare API requests. I had to focus how to fit other datasets in addition to Foursquare data to find something interesting but instead of using fancy algorithms and testing them I had to focus on data and it was really time consuming, I spent 2x more time on each assignment. Whole IBM DS path was great but final course was a big disappointment.

A comprehensive recap of the skills learned on the course, but once again full of typos, errors and technical issues. The capstone project itself was an incredibly dull way to bring all the information together, leading to a presentation containing almost no insight. The choice sof data and visualisations made little sense, essentially turning this into brainless donkey work.

it was the worst Coursera course I've ever took. Course capstones pages are buggy, there are not adequate explanations at videos. The final project is very painful because there are not enough resources and you are forced to prepare "notebook", "report" and "presentation/blog post" can't we just prepare two of these items?

Issues with the lab sessions and untimely responses from the staff hindered my learning. Also, the students did not respond to my questions in the discussion forums, they seem to only be about themselves and having people grade their assignments.

I get now (8th of July) a message that I can only begin in August on my assignment although I have finished week 4 succesfully!!!

I find it unacceptable that Coursera tries to let you renew your monthly subscription over and over again.

Could you unlock my assignment week 5 please?

Also Coursera let you use IBM lite subscription for free in the beginning but if you use it a lot, the lite version is not enough. You can upgrade, if you pay...

If you enroll for subcourses, you are charged each time you submit another small part.

the classes itself are good but Coursera is not clear on how much you need to pay in the end. The prices they let you believe seem reasonable but there are hidden extra costs.

Like other courses of this program, it's just worthwhile for someone who is still learning Python, but even in theses cases, there are better course to take. The grading system is the worst I've seen in more than 40 courses at Coursera.

If your IBM Watson trial code expires it's impossible to get anyone to help you. You can't finish the course and you're stuck paying coursera while you try to resolve it. Pessimal. Feels like a scam.

Terrible. I was more than half way through this course and they changed the course and I had to start over. It was already unclear, now the instructions are much less than before. A waste of money.

There are technical issues in the Lab Projects and no sufficient help received from the instructor/tutors. Very bad experience!

It does add nothing to previous knowledge, the final assignment is a major waist of time

cannot unlock and continue studying after resting the deadline.

good

Bad

I recently enrolled in the Applied Data Science Capstone course as part of the IBM Data Science Professional Certificate specialization on Coursera, and I couldn't be more satisfied with the experience. This course provides an excellent opportunity to apply everything I had learned throughout the specialization in a real-world project. The hands-on approach makes the concepts clear and actionable, and I gain valuable experience with advanced tools and techniques used in the industry, such as data collection, cleaning, analysis, and visualization. The project-based learning style id both engaging and challenging, allowing me to develop a comprehensive solution to a real-world data problem. The guidance provided by the instructors is clear, and I appreciate the well-structured assignments that keep me on track. By the end of the course, I feel confident in my ability to tackle data science problems independently. This capstone truly tied together everything I learned in the previous courses and give me a sense of accomplishment. Highly recommended for anyone looking to solidify their data science skills with practical experience!

When I first started this capstone, it felt a bit disjointed compared with the rest of the courses. But after really biting into it, I realized the content makes sense: it allowed me to put to the test what I learned on the course. Being constrained to use the Foursquare API on the capstone feels a bit odd, but at the end using an API to get information works really well. I tried scrapping the information my self and the workload I put on my self became significant.

The only new machine learning tool introduced for this final part is K-means clustering, it's the most abstract concept on the entire specialization and I think it's the only one that could have been presented somewhat better. What I noticed while reviewing my peers is that for the assignments everyone (me included) would just copy the k-means clustering algorithm and repeat the same analysis used on the labs. Which is kind of a shame, I just wish K-means clustering could have been developed better, beyond copy-paste.

At then end it's you; the person taking this specialization, the one who decides how much work you're going to put into this.

A comprehensive step-by-step guided project with a real-world dataset to implement most of the required skills that a data scientist must be familiar with. Data collection thru API & web scraping, data cleaning, EDA with SQL & pandas/numpy, visualizations with matplotlib, pandas & seaborn plus geospatial data visualization with folium, creating interactive with plotly dash, classification with machine learning algorithms, and preparing a set of slides to present the story and a GitHub repo to share the outcomes with the peers are the steps that an individual must carry out to complete the course. Totally recommended! Thanks, IBM & thanks, Coursera!