“Data Science” Project Cycle (Part 5)
The Data Science Project Cycle series consists of 5 separate articles, and this part is the last article in the series. In this part, we will talk about “Running on Production”.
Note: I usually will use some abbreviated words below:
- Data Science — DS
- Artificial Intelligence — AI
- Machine Learning — ML
- Big Data — BD
- Deep Learning — DL
- Statistical Learning — SL
Let’s continue…
As you can see from the above picture, everyone has a role in the DS, and this will lead to a more scissors opening and the DS to be subdivided in the future. One of them is the Running on Production expert.
Of course, it is not always possible to integrate the work into live systems. This will differ for many projects. This can be in the form of sending an SPSS output for a project to excel related units. If an e-commerce site has worked with people-related advice systems, the project outputs developed on any platform can be printed on the mysql tables and through this interface, the relevant results can be obtained. Similarly, in a web system, predictive results can be kept in mysql tables and recalled when needed. Or it may be necessary to run a more advanced instantaneous predictive model.
Similarly, job planning can vary according to every need. It may not be necessary to plan. Perhaps the most important item that can be expressed about this step is:
User movements that occur after the results produced by using models and the structures of these movements, that is, when the data is used for modeling, will cause the models into vicious circles and cause serious biases. In other words, the projects in live systems connected to Jobs should NOT be included in the jobs that will be reworked for modeling, with the user guidance and the data generated by the reactions of the users after these prompts. This situation is very critical in DS-ML processes.
In addition, the systems that come out live need to be tested and the process should continue alternately. After all this information, if the visual published by Microsoft is examined again:
Motivation, SL & ML, Programming Tools, what are you waiting for?
Now that we have more or less how a DS project should be handled, now we need to find a project to do!
This is all I have written about the “Data Science Project Cycle”. If you want to know more about DS and related others, you can check out my other serial articles. Sample:
What is Data Science(DS) and How can it be learned?
You can reach me from my Linkedin account for all your questions and requests.
Hope to meet you in other series articles and articles…🖖🏼
References
1. https://twitter.com/kashthefuturist/status/1024456825708662784
2. https://www.veribilimiokulu.com/blog/veri-bilimi-proje-dongusu/
3. https://www.datasciencecentral.com/profiles/blogs/data-science-opportunities-in-the-age-of-covid
4. https://medium.com/analytics-vidhya/data-science-project-lifecycle-a-primer-f4bf5f79eef8
5. https://cdn.mos.cms.futurecdn.net/kMFJjsRDbNLmQ3MGCwbhsL-970-80.jpg.webp
