Data Science Interview Questions

Python is the most popular programming language for Data Science because it is easy to learn, has a simple syntax, and offers powerful libraries like NumPy, Pandas, Matplotlib, Scikit-learn, and TensorFlow. It supports data analysis, machine learning, and visualization.

name = "Python"
print(name)

Output
Python

Machine Learning is a branch of Artificial Intelligence where computers learn patterns from data and make predictions without being explicitly programmed for every task.

Data Science is a broad field that includes collecting, cleaning, analyzing, and visualizing data. Machine Learning is a subset of Data Science focused on building predictive models using algorithms.

A Data Science project generally follows these steps:

1. Data Collection

2. Data Cleaning

3. Data Exploration

4. Feature Engineering

5. Model Building

6. Model Evaluation

7. Deployment

Data Cleaning is the process of fixing incorrect, incomplete, duplicate, or inconsistent data before analysis. Clean data improves the accuracy of machine learning models.

Exploratory Data Analysis (EDA) is the process of understanding data using statistical summaries and visualizations before building machine learning models.

EDA helps identify:

• Missing values

• Outliers

• Correlation

• Trends

• Patterns

import pandas as pd
marks = [80,90,75,95]
df = pd.DataFrame(marks)
print(df.describe())

Output
count     4
mean     85
min       75
max      95

NumPy is a Python library used for numerical computing and mathematical operations on arrays.

Pandas is a Python library used for handling and analyzing tabular data using DataFrames.

import pandas as pd

data = {
    "Name":["Rahul","Aman"],
    "Age":[22,25]
}

df = pd.DataFrame(data)

print(df)

Output
    Name  Age
0  Rahul   22
1   Aman   25

Supervised Learning

• Uses labeled data.

• Predicts known outputs.

• Examples: House Price Prediction, Email Spam Detection.

Unsupervised Learning

• Uses unlabeled data.

• Finds hidden patterns.

• Examples: Customer Segmentation, Clustering.

Overfitting

A model memorizes training data and performs poorly on new data.

Underfitting

A model is too simple and fails to learn important patterns from the data.

Training Accuracy = 99%
Testing Accuracy = 62%