ICICI Securities Data Scientist Interview Questions and Answers

Q1. Projects and Data Science concepts
Add your answer

Q1. Python and coding skills
Add your answer

Medium General Aptitude questions and technical(Big Data, Python etc.)

Q1. ML Algorithms (SVM, Random forest, bagging boosting, ridge, etc)
Add your answer

Q1. Deep equations and understading of DL and ML Algorithms
Ans. 
Understanding deep equations and algorithms in DL and ML is crucial for a data scientist.
• Deep learning involves complex neural network architectures like CNNs and RNNs.

• Machine learning algorithms include decision trees, SVM, k-means clustering, etc.

• Understanding the math behind algorithms helps in optimizing model performance.

• Equations like gradient descent, backpropagation, and loss functions are key concepts.

• Practica...read more
Answered by AI
Add your answer

Many Mcq,s.Similar to cat exam

Ml case study . Eg loan default prediction

Q1. Tell me about yourself?
Add your answer
Q2. What are the GenAI industry trends? ChatGPT vs DeepSeek high level discussion.
Ans. 
GenAI trends focus on advancements in AI models, with ChatGPT and DeepSeek showcasing different applications and capabilities.
• Increased adoption of conversational AI in customer service, exemplified by ChatGPT's integration into various platforms.

• DeepSeek focuses on specialized knowledge retrieval, enhancing search capabilities in niche domains like legal or medical fields.

• Emergence of hybrid models combining generativ...read more
Answered by AI
Add your answer
Q3. ML question: Random Forest vs XGBoost
Ans. 
Random Forest is an ensemble method using bagging, while XGBoost uses boosting for improved accuracy and speed.
• Random Forest builds multiple decision trees and merges them for better accuracy.

• XGBoost optimizes the model by sequentially adding trees that correct errors of previous ones.

• Random Forest is less prone to overfitting compared to individual decision trees.

• XGBoost includes regularization techniques to prevent o...read more
Answered by AI
Add your answer
Q4. ML question: Classification vs Regression
Ans. 
Classification predicts categories, while regression predicts continuous values in machine learning tasks.
• Classification: Assigns labels to data points (e.g., spam vs. not spam).

• Regression: Predicts numerical values (e.g., house prices based on features).

• Classification algorithms include logistic regression, decision trees, and SVM.

• Regression algorithms include linear regression, polynomial regression, and regression t...read more
Answered by AI
Add your answer
Q5. ML question: How does one handle data imbalance in a dataset?
Ans. 
Data imbalance can skew model performance; various techniques can help mitigate its effects.
• Resampling techniques: Use oversampling (e.g., SMOTE) or undersampling to balance classes.

• Use different evaluation metrics: Focus on precision, recall, or F1-score instead of accuracy.

• Implement cost-sensitive learning: Assign higher misclassification costs to minority class errors.

• Utilize ensemble methods: Techniques like Random...read more
Answered by AI
Add your answer
Q6. ML question: When to use accuracy, precision, recall? How to identify the classification model's performance using above three? F score, ROC area under the curve.
Add your answer
Q7. ML question: What are the X and Y axis of ROC curve?
Ans. 
The ROC curve plots true positive rate against false positive rate to evaluate classifier performance.
• X-axis: False Positive Rate (FPR) - the ratio of negative instances incorrectly classified as positive.

• Y-axis: True Positive Rate (TPR) - the ratio of positive instances correctly classified as positive.

• Example: A model with a TPR of 0.9 and FPR of 0.1 indicates high sensitivity but some false alarms.

• The ROC curve help...read more
Answered by AI
Add your answer
Q8. SQL questions: Employee and department table; 1. Salary of employees with salary > avg salary 2. Salary of employees with salary > avg salary of department 3. Salary of top 3 employees with salary > avg sa...read more
Add your answer
Q9. Any other questions for us?
Ans. 
I'm curious about the team's data science projects and how they align with the company's goals and vision.
• What are the current data science projects the team is working on?

• How does the data science team collaborate with other departments?

• Can you share examples of how data-driven decisions have impacted the company?

• What tools and technologies does the team primarily use?

• How does the company support continuous learning a...read more
Answered by AI
Add your answer

Q1. What is Gradient Descents?
Ans. 
Gradient descent is an optimization algorithm used to minimize a function by iteratively moving in the direction of steepest descent.
• Gradient descent is used to find the minimum of a function by taking steps proportional to the negative of the gradient at the current point.

• It is commonly used in machine learning to optimize the parameters of a model by minimizing the loss function.

• There are different variants of gradie...read more
Answered by AI
Add your answer
Q2. What is LSTM?, and what are the gates in it?
Ans. 
LSTM (Long Short-Term Memory) is a type of recurrent neural network designed to handle long-term dependencies.
• LSTM has three gates: input gate, forget gate, and output gate.

• Input gate controls the flow of information into the cell state.

• Forget gate decides what information to discard from the cell state.

• Output gate determines the output based on the cell state.
Answered by AI
Add your answer
Q3. They gave me a link to dataset and started saying the operations to apply on that. E.g, value_counts, null_values, fill the values with mean,etc.
Add your answer
Q4. What is t-test? What is Mean, Median and Mode and where to use these?
Ans. 
T-test is a statistical test used to determine if there is a significant difference between the means of two groups.
• Mean is the average of a set of numbers, median is the middle value when the numbers are ordered, and mode is the most frequently occurring value.

• Mean is sensitive to outliers, median is robust to outliers, and mode is useful for categorical data.

• T-test is used to compare means of two groups, mean is used...read more
Answered by AI
Add your answer
Q5. What is RANDOM FOREST ?
Ans. 
Random Forest is an ensemble learning method used for classification and regression tasks.
• Random Forest is a collection of decision trees that are trained on random subsets of the data.

• Each tree in the forest makes a prediction, and the final prediction is the average (regression) or majority vote (classification) of all trees.

• Random Forest helps reduce overfitting and improve accuracy compared to a single decision tre...read more
Answered by AI
Add your answer