I had "Data Science Specialization" course in Coursera https://www.coursera.org/specializations/jhu-data-science.
It is a basic level course to learn data science essential skills by using R language, from getting data to visualization and create prediction model. It starts from beginner level, but finally I could developed own application by using prediction model. So this course was nice for me to briefly learn the process for data scientist or machine learning engineer.
I listed up the useful skills and tools I learned in the below.
- R language skill Getting and cleaning data, exploratory data analytics and statistical inference, etc. This was the first time for me to use R language, however R studio was very powerful tool to analyzing the data.
- R markdown For data analysis, I need to report the result to others. Reporting modern data analyses in a reproducible manner is necessary to make our work a high productivity. R markdown, one of R packages, is notebook interface to write down text and code.
- Creating product by using Shiny Shiny is an R package that makes it very easy to develop interactive web applications using data analytics. Many typical widgets are prepared, so we can build dashboards.
In the final assignment(Data Science Capstone), I built my own simple web app, which is next word prediction based on the sentence input by users, which is natural language processing (NLP). For example, Swiftkey keyboard app is the same concept. I can learn to analyze data, build prediction model, evaluate precision and develop web application. And I need to care of trade-off between response time and accuracy. More data and more complex algorithm I use, Better the accuracy is, but the response time is getting slower. It is very tough but I could have 0 to 10 experience to build data science product. Here is my application https://norihiton.shinyapps.io/NextWordPrediction/. It is very simple app but I could learn so many things. If you have any advise to make it better, please send messages.
If you are a beginner for data science, I recommend this course for the first step.
Here, I got the certification in about 3 month. I can link the certification to Linkedin profile.