Note of data science training EP 14 END – Data scientists did their mistakes

In this series

1. Note of data science training EP 1: Intro – unboxing
2. Note of data science training EP 2: Pandas & Matplotlib – from a thousand mile above
3. Note of data science training EP 3: Matplotlib & Seaborn – Luxury visualization
4. Note of data science training EP 4: Scikit-learn & Linear Regression – Linear trending
5. Note of data science training EP 5: Logistic Regression & Dummy Classifier – Divide and Predict
6. Note of data science training EP 6: Decision Tree – At a point of distraction
7. Note of data science training EP 7: Metrics – It is qualified
8. Note of data science training EP 8: Ensemble – Avenger's ensemble
9. Note of data science training EP 9: NetworkX – Map of Marauder in real world
10. Note of data science training EP 10: Cluster – collecting and clustering
11. Note of data science training EP 11: NLP & Spacy – Languages are borderless
12. Note of data science training EP 12: skimage – Look out carefully
13. Note of data science training EP 13: Regularization – make it regular with Regularization
14. Note of data science training EP 14 END – Data scientists did their mistakes

This is the last class of the training and I’d like to thanks Mr.James Larkin, Mr. Greg Baker, and Mr. Attapol Thamrongrattanarit for devoting their efforts to share knowledge to me and friends.

And here are the advices for frequent mistakes committed by data scientists.

---

Question

What mistakes do data scientists often make?

---

Mr. Larkin answers

There are mistakes generally made by data scientists new to the field. Some high level issues I’ve seen in work environments are:

1. Focusing more on the algorithms than the data itself
   Success is needed but data insights or something lies in it are prior.
2. Picking unfit algorithms for their work
   This often occurs because the data scientists want to test out some new, powerful but black box options. It may not be fit or suitable for data of the work.
3. Poor data understanding
   First steps are data exploring, cleansing, and visualizing. There are 99% of the time that we attempt to skim through data and choosing an algorithm. A veteran data scientist tends to efficiently dive deep into the dataset.
4. Experience-wise
   Experiences are good. Nevertheless sometimes we should focus more on data not the learned experience. The data can be changed overtime and we data scientists need to chase it.
5. Team and communication
   A lot of fairly good works can go nowhere in an organization. This is because they were not explained clearly and effectively to the appropriate stakeholders.

---

They are benefit to us and some are able to be applied in other works.

This is the end of this series, I would say.

And I will pick it up for you for what’s next. Stay tuned.