Data Science Viva Notes

Q: What is Exploratory Data Analysis (EDA)?

A: EDA is the process of exploring and visualizing data to understand its structure, patterns, and relationships

before applying any model.

Q: Summary statistics means?

A: Summary statistics are basic values that describe a dataset like mean, median, mode, min, max, standard

deviation.

Q: Histogram displays what?

A: A histogram shows the frequency distribution of a numeric variable.

Q: Box plot?

A: A box plot shows the spread of data using median, quartiles, and outliers.

Q: How to conclude after seeing boxplot?

A: You can see if data is symmetric, skewed, and whether there are outliers.

Q: Whiskers means?

A: Whiskers in a boxplot show the minimum and maximum values within 1.5 IQR from the quartiles.

Q: Scatter plot?

A: A scatter plot shows the relationship between two numeric variables.

Q: Data cleaning?

A: Data cleaning means fixing or removing wrong, incomplete, or inconsistent data.

Page 1
 Data Science Viva Notes

Q: Handling inconsistencies mean?

A: It means correcting values that are wrongly formatted or mismatched in the dataset.

Q: How to apply imputation?

A: Imputation is filling missing values using mean, median, mode, or predictive models.

Q: How to remove duplicates?

A: Use tools or code (like `.drop_duplicates()` in Python) to delete repeated rows.

Q: Data transformation and feature engineering means?

A: Data transformation changes the data format or scale. Feature engineering creates new useful features for

the model.

Q: Normalization means?

A: Scaling all numeric data to a common range (like 0 to 1) to treat all features equally.

Q: Data transformation: converting categorical variables?

A: Convert them into numbers using encoding like One-Hot Encoding or Label Encoding.

Q: Binning?

A: Binning means converting continuous data into fixed intervals or categories.

Q: Polynomial feature creation?

A: Creating new features by raising existing numeric features to powers (like x², x³).

Q: Statistical analysis?

Page 2
 Data Science Viva Notes

A: It involves using mathematical techniques to summarize, understand, and draw conclusions from data.

Q: Hypothesis testing means?

A: It tests if a statement about a population is likely true using sample data.

Q: Regression analysis?

A: A technique to study relationships between variables and predict one based on others.

Q: Linear regression model means?

A: A model that predicts an output using a straight-line relationship with input(s).

Q: T-test means?

A: A test to compare the means of two groups to see if they are significantly different.

Q: Chi-square test?

A: A test to check the association between two categorical variables.

Q: P-value means?

A: It shows the probability that the result happened by chance. A small p-value (<0.05) means the result is

statistically significant.

Q: Logistic regression?

A: A model used for classification problems (like yes/no) by predicting probability.

Q: Accuracy means?

A: Accuracy is the percentage of correct predictions made by a model.

Page 3
 Data Science Viva Notes

Q: Accuracy, Precision, and Recall?

A: Accuracy: Overall correct predictions. Precision: Correct positive predictions. Recall: All actual positives

correctly predicted.

Q: ROC AUC curve?

A: A graph that shows model performance. AUC score near 1 is best.

Q: Clustering means?

A: Grouping similar data points together based on features.

Q: Segmentation?

A: Dividing data into meaningful groups (like customer segments).

Q: K-means clustering?

A: A method that divides data into 'k' clusters based on similarity.

Q: Difference between clustering and segmentation?

A: Clustering is the technique, segmentation is the result or goal.

Q: Churn prediction model?

A: A model that predicts which customers are likely to leave (churn).

Q: Time series analysis?

A: Analyzing data collected over time to find patterns and trends.

Page 4
 Data Science Viva Notes

Q: Trend?

A: Long-term movement in data (upward or downward).

Q: Seasonality?

A: Repeating patterns at regular intervals (like monthly or yearly).

Q: Noise components?

A: Random or irregular variations in data that cannot be explained.

Q: Outliers mean?

A: Unusual values far from most of the data.

Q: ARIMA?

A: A forecasting model using past values and errors. It stands for AutoRegressive Integrated Moving

Average.

Q: Forecasting means?

A: Predicting future values based on past data.

Q: Exponential smoothing?

A: A method to forecast data by giving more weight to recent observations.

Q: Anomalies?

A: Unusual or unexpected data points that don?t fit the pattern.

Q: Z-score?

Page 5
 Data Science Viva Notes

A: A value that shows how far a data point is from the mean, in standard deviations.

Q: Isolation Forest model?

A: A model used to detect anomalies by isolating them from the rest of the data.

Q: Profiling means?

A: Creating a summary of data to understand its structure, quality, and patterns.

Q: Correlation matrix?

A: A table showing how variables relate to each other (with values between -1 and 1).

Q: Correlation coefficient?

A: A number that shows the strength and direction of the relationship between two variables.

Q: ML, AI, and Deep Learning?

A: AI is the broad field. ML is a part of AI that learns from data. Deep Learning is a type of ML using neural

networks.

Q: Supervised and Unsupervised learning?

A: Supervised: learns with labeled data (has answers). Unsupervised: finds patterns from unlabeled data.

Page 6