Capstone Project
Social Media Tourism
- M. Brahma Chari
5th April 2024
"Navigating New Horizons:
Transforming Aviation Marketing
Through Digital Insights and Targeted
Advertising"
SCOPE & BUSINESS OBJECTIVE
Business Problem:
The aviation industry seeks to replace traditional marketing methods with
targeted digital strategies, leveraging social networking platforms for
advertising. By analyzing the digital and social behavior of potential customers,
the company aims to deliver personalized advertisements to individuals most
likely to purchase their services.
Objective:
To effectively implement the targeted digital advertising strategy of the aviation
company we can conduct several analysis:
Customer Segmentation
Predictive Modeling
Sentiment Analysis
Behavioral Targeting and many more.
3
NEED OF THE STUDY
Influence of digital media on tourism
Understanding the importance of online
reviews.
Analyzing customer behaviour and creating
strategies based on their wants.
4
DATA DESCRIPTION
About the Data:
The data provided for analysis is the past
customer data from aviation tourism sector, their
preferences in selecting destinations, Number of
members in the family, their social media
activism about trip planning and interest in
travelling. Here are the variables present in the
provided Dataset
Constraints in the Data:
-Few columns which should have been Integer
data type are of Object datatype.
-Missing values.
-Presence of special characters (Eg: *)
-Multiple names for same category of data.
5
DATA FACTS & STATISTICS
Data Info:
The dataset has 11760 Rows and 17
columns
Of which:
-10 columns are integer/float datatypes
-7 columns are object datatypes.
However, two columns:
Member in family
Yearly avg outstation checkins
columns are object datatype which are to
converted into float datatypes.
6
DATA PRE-PROCESSING
- We have a “Special Character - *” in the “Yearly avg outstation checkins, which
was replaced by the mode value.
- Replaced the attributes which are of similar nature.
Dataset column Repeated attributes Replaced with
Preferred Device ANDROID, Android OS Android OS
Other, Others Others
Preferred Location Type Tour Travel Tour and Travel
Tour and Travel
Member in family Three 3
3
7
MISSING VALUES & OUTLIER
TREATMENT
Missing values can be imputed using statistical methods (like mean, media and mode values) or by using K-Nearest
Neighbors (KNN) algorithm where missing data is estimated based on the values of neighboring data points.
Checking the % of null values in the whole dataset:
Since, the null values overall percentage is less than 5, we can use statistical methods to impute null values in the
variables respectively.
Data Column in which Null values are present Statistical tool used for
imputing Null values
Yearly_avg_view_on_travel_page Median
total_likes_on_outstation_checkin_given Median
Yearly_avg_comment_on_travel_page Median
following_company_page Mode
yearly_avg_Outstation_checkins Mode
preferred_device Mode
preferred_location_type Mode
8
We observed that 2 features contain outliers which are taken care by using the Inter Quartile range.
MODELING APPROACH USED AND
WHY
Modeling involves training a machine learning algorithm to forecast
labels based on input features, adjusting it to suit specific business
requirements, and assessing its performance using separate validation
data.
Since it's a supervised learning task with a binary categorical dependent
variable, I experimented with various algorithms/models to obtain the
most effective trained model for making inferences and predictions.
1. Logistic Regression
2. Random Forest
3. K Nearest Neighbor
4. XGB Classifier
9
MODELING APPROACH USED AND
Metric
WHY Accuracy Score Recall value Precision F1 Score AUC Score
Models used Train Test Train Test Train Test Train Test Train Test
Logistic Regression 0.84 0.86 1 1 0.84 0.86 0.92 0.92 0.65 0.651
Tuned LR 0.86 0.87 0.17 0.17 0.73 0.77 0.27 0.28 0.776 0.776
Random Forest 1 0.98 1 0.88 1 1 1 0.93 1 0.998
Bagged RF 0.98 0.93 0.98 0.96 0.98 0.91 0.98 0.93 0.998 0.971
KNN Classifier 0.92 0.86 0.93 0.89 0.91 0.83 0.92 0.86 0.976 0.906
Tuned KNN 0.92 0.90 0.94 0.94 0.91 0.88 0.92 0.91 0.975 0.943
Gradient Booster 1 0.93 1 0.96 1 0.91 1 0.93 1 0.970
The best model is Random forest, making predictions with 96% accuracy for both mobile and laptop
device.
So, we can implement the Random Forest approach for better predictions and get the best results
which would help the aviation company save money.
10
EDA & MODELING
INSIGHTS
The variables "Total likes on outstation
check-in" and yearly average views on Notably, around 85% of customers Adults constitute 57% of the customer
the travel page demonstrate the have not previously purchased the base, with the remaining falling into
highest correlation with the daily product. other age groups.
average time spent on the page.
Around 26% of users checked in at an
Approximately 90% of customers utilize
A significant majority (84%) of outstation location one week prior,
mobile devices, with tablets being the
individuals are not employed. indicating a reduced inclination
preferred choice.
towards travel.
The variables "Total likes on outstation
Preferences among customers are
More than 8000 customers, the check-in" and yearly average views on
skewed towards beach and financial
majority, do not follow the company's the travel page demonstrate the
destinations, followed by historical sites
page. highest correlation with the daily
and other options.
average time spent on the page.
Random Forest proves to be the best model, achieving 96% accuracy for both mobile and laptop
devices. Implementing the Random Forest approach can lead to better predictions, aiding the
aviation company in cost savings
11
RECOMMENDATIONS
Target Customers Who Haven't Checked In Recently
Campaign Strategy for Engaged Mobile Users
Strategic Targeting of Adult Consumers
Implementing Digital Advertisement for High-Propensity Customer Segments
Here are some ideas tailored for an aviation company based on the given scenarios:
• Family Package Offers
• Beach Destination Discounts
• Social Media Campaigns for User Engagement
• Advance Booking Promotions
• Frequent Flyer Rewards
By implementing these tailored strategies, the aviation company can effectively retain customers,
attract new travelers, and boost overall business performance.
12
13
