Quora is a regular haunt of mine these days, and a lot of my activity there is centered on topics of deep interest – usually data, engineering, aviation and technology. Here’s the first version of the Quora data science answers roundup that I posted in January 2017, soon after I was designated Quora Top Writer for 2017.

What you see below are some more of my answers from 2017, on data and related areas, from Quora.

Data science, data analysis, simulation, probability, statistics and machine learning answers:

1. Some hard truths about becoming a data scientist
2. The best thing about working in data science
3. Important qualities for data scientists. Related posts here and here
4. Relevance of the basics of ML given the presence of machine learning APIs
5. Expensive boot camps for data science and justifying spending
6. Nontrivial ideas from probability and statistics required for data science
7. Thoughts on Andrew Ng’s deep learning course (which led to a blog post here too)
8. On new and interesting research ideas in the AI space
9. Managing unstructured text data and feature extraction – more here
10. Managing missing data fields and null values in data science problems
11. On linear programming versus stochastic searches for hyperparameter optimization
12. Differentiating between fitness and loss functions
13. On model interpretability in machine learning
14. Characteristics of a good regression model
15. Distribution modeling and probability – 1 , 2 , 3 , 4
16. On data analysis and its use in the manufacturing industry
17. Optimization techniques in data analysis and data science
18. On the philosophy of deep learning – related answers on how deep learning algorithms learn , on weight initialization in deep neural networks
19. On time series models in data analysis – more here , here , here , here , here , here and here
20. Convex optimization and the use of gradient descent
21. On Genetic Algorithms – 1 , 2 , 3
22. Anomaly detection in financial time series data – related answer here
23. Significance and difference in significance testing
24. Agent based modeling for traffic simulations

Technology-specific answers on data science and analysis:

1. On big data technology courses, and the lack of architecture, strategy and such courses
2. On the continuing relevance of SQL/RDBMS technologies
3. The develop-vs-use conundrum for building data and machine learning systems – more here
4. Advice on career and certifications  – 1 , 2 , 3 , 4
5. Programming language specific answers – 1 , 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11
6. General data science books, resources, skills – 1 , 2 , 3
7. On big data ecosystems and components
8. Perspectives on data warehousing and big data technologies
9. Contextualizing tools like Excel in the context of data analysis

Data science and management:

1. The importance of BI and decision enablement tools in the data space
2. Andrew Ng’s venture and how it could be differentiated from others
3. Managing data science projects

Hope you enjoy reading through them and find them interesting and informative!