1. Why did you use tree based algorithms for financial prediction modelling? 2. What are some methods used for categorical feature engineering? 3. How is big data modelling done? What does your dataset look like? 4. How to transform textual big data into training data ? { Explain NLP terminologies] 5. Can we replace precision or accuracy instead of Cost function while training deep learning models? 6. If data is imbalanced, then what are some methods to make data balanced before model training? 7. Why is PCA used for dimensionality reduction when we can undersample the data? If we are using PCA in small datasets, to visualize linearity and correlation or reduce dimensions. Then would just correlation function and manual feature engineering be good? 8. What are some methods to balance the data if model is underfitting? 9. What are features and attributes in your dataset? Did you use classification or regression? What kind of statistical feature engineering used? 10. How does your final prediction look like? How did you engineer your target variable in live streaming big dataset?
