#18 Exploratory data analysis | Python for Data Science

Welcome to 'Python for Data Science' course !

This video focuses on uncovering patterns and relationships within your data. The lecture introduces frequency tables for understanding the distribution of categorical variables, like fuel type in the car dataset (). You'll learn how to handle missing values when creating these tables.
It then covers two-way tables for exploring relationships between two categorical variables, for example, fuel type and gearbox type. You'll see how to calculate joint, marginal, and conditional probabilities from these tables to make inferences about variable relationships.
Finally, the video introduces correlation as a measure of association between two numerical variables, using the car dataset to demonstrate its calculation and interpretation. It emphasizes that understanding these exploratory data analysis techniques is crucial for generating hypotheses and deriving meaningful insights from your data.
NPTEL Courses permit certifications that can be used for Course Credits in Indian Universities as per the UGC and AICTE notifications.

#ExploratoryDataAnalysis #FrequencyTables #CrossTabulation #Correlation

[heisenbug]

For selecting numeric columns using exclude = ['object'] is kinda indirect. Use include = [np.number] both will do same thing but later one is preferred.

[Abhi-qiwm]

The data at 13:37 needs to be feature scaled it seems

[monexsharma]

19:05 please explain how rows 'All' values come

[deepanjanmitra]

Though many of the values of category 'FuelType' are missing in my dataset yet when I set dropna=False I get the same result as I get when dropna=True. Why is this happening?

[kodiaeudes]

Thank you so much for sharing this marvelous course. Could you please share the dataset used in this video, so I can try to do the same thing.

[bikdigdaddy]

terrible lecture. she just reads, doesn't know herself what she's teaching.