Create pages/3_EDA.py
Browse files- pages/3_EDA.py +33 -0
pages/3_EDA.py
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import streamlit as st
|
| 2 |
+
|
| 3 |
+
st.title("π Exploratory Data Analysis (EDA)")
|
| 4 |
+
|
| 5 |
+
st.markdown("""
|
| 6 |
+
## π Data Exploration:
|
| 7 |
+
The dataset was carefully analyzed to understand:
|
| 8 |
+
- Distribution of ticket prices across different airlines.
|
| 9 |
+
- Impact of number of stops on ticket price.
|
| 10 |
+
- Influence of departure and arrival times on pricing.
|
| 11 |
+
- Effect of travel class on ticket cost.
|
| 12 |
+
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
## π Key Observations:
|
| 16 |
+
- **Air India** and **Business Class flights** tend to have higher ticket prices.
|
| 17 |
+
- **Zero-stop flights** are usually more expensive compared to those with multiple stops.
|
| 18 |
+
- Flights booked closer to the travel date tend to be significantly more expensive.
|
| 19 |
+
|
| 20 |
+
---
|
| 21 |
+
|
| 22 |
+
## π Visualization Techniques:
|
| 23 |
+
- Countplots to understand the frequency of different categories.
|
| 24 |
+
- Boxplots to analyze outliers and distribution.
|
| 25 |
+
|
| 26 |
+
These insights guided the feature engineering and model building processes.
|
| 27 |
+
""")
|
| 28 |
+
|
| 29 |
+
if st.button('Next'):
|
| 30 |
+
st.switch_page('pages/4_Feature_Engineering.py')
|
| 31 |
+
|
| 32 |
+
if st.button('Go to Data Understanding'):
|
| 33 |
+
st.switch_page('pages/2_Data_Understanding.py')
|