Data-Science Interview Questions

CRISP-DM (Cross Industry Standard Process for Data Mining) is a methodology for data science projects.

Phases include Business Understanding, Data Understanding, Data Preparation, Modeling, Evaluation, and Deployment.

It provides a structured approach to solving business problems with data science.

Following CRISP-DM ensures alignment with business objectives and systematic project execution.

It is widely adopted in industry and academia for managing data science initiatives.

Exploratory Data Analysis (EDA) is an approach to summarizing data characteristics, often using visual methods.

It involves checking for missing values, outliers, distributions, and relationships between variables.

Tools like Pandas, Matplotlib, Seaborn, and Jupyter Notebooks are commonly used for EDA.

EDA informs feature engineering, model selection, and data quality before modeling begins.

It is a foundational step in the data science pipeline for informed model building.

Data cleaning involves detecting and correcting errors and inconsistencies in datasets.

It includes handling missing values, removing duplicates, fixing structural issues, and normalizing data formats.

Dirty data leads to inaccurate models and flawed insights.

Pandas, NumPy, and Scikit-learn are commonly used for data cleaning in Python.

Automated tools and scripts can be written for recurring data quality issues.

Clean data is essential for building accurate and reliable models.