Scribd
Upload
42 views17 pages

Presented by

This document presents a machine learning approach to analyze consumer behavior on e-commerce websites. It uses the random forest algorithm to predict consumer purchasing patterns based on product attributes like brand, price, and previous consumer interactions like viewing, adding to cart, or purchasing. The methodology involves collecting online shopping data, feature selection with random forest, data preprocessing, model training and testing, and result analysis. The random forest model achieves good prediction accuracy for consumer behavior categories. The findings can help e-commerce sites optimize their marketing strategies and improve customer experience.

Uploaded by

Shaik shahidcs044
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
42 views17 pages

Presented by

This document presents a machine learning approach to analyze consumer behavior on e-commerce websites. It uses the random forest algorithm to predict consumer purchasing patterns based on product attributes like brand, price, and previous consumer interactions like viewing, adding to cart, or purchasing. The methodology involves collecting online shopping data, feature selection with random forest, data preprocessing, model training and testing, and result analysis. The random forest model achieves good prediction accuracy for consumer behavior categories. The findings can help e-commerce sites optimize their marketing strategies and improve customer experience.

Uploaded by

Shaik shahidcs044
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 17

Analysis of Consumer Behavior in E-Commerce website

using a Machine Learning based approach

Presented By
DEEKSHITH
SHAIK MOHD SHAHID
FIRDOSEUNNISA
NITHYA SRINIVAS
Department Of Computer Science & Engineering
Karavali Institute of Technology
INDEX
Introduction
Objectives
Random forest algorithm
Methodology flow diagram
Result analysis
Conclusion
Reference

2
INTRODUCTION
 People are eager to buy products from
online sites like Amazon, ebay, Flipkart etc.
Online sites also provide facility for
customers to write review on products they
buy.
These reviews help consumers and vendors
for making decision on marketing strategies,
and the improvement of products and
services.
3
Objective
 Customer purchasing behavior is an
important and interesting issue in the
field of digital marketing or e-
commerce service.
 Our proposed system can easily
predict the demand of a product before
adding the product on e-commerce
website.

4
Random Forest is a popular machine learning
Random Forest Algorithm

algorithm that belongs to the supervised


learning technique.
It can be used for both Classification and
Regression problems in ML.
Random Forest is a classifier that contains a
number of decision trees on the given dataset
and takes the average to improve the predictive
accuracy of that dataset.
The greater number of trees in the forest leads
to higher accuracy and prevents the problem of
5
overfitting.
Random Forest Algorithm

6
It takes less training time as compared
Why Random Forest Algorithm

to other algorithms.
It predicts output with high accuracy,
even for the large dataset it runs
efficiently.
It can also maintain accuracy when a
large proportion of data is missing.

7
Methodology Flow Diagram
Methodology(Cond.)
Collection of a huge amount of data from different
online sources.
In features selection , Use of Random Forest
algorithm to select the best suited features for our
proposed method.
preprocessed data using Python and different python
libraries to understand the demand of different
products those were sold online.
visualizing the dataset in such a way, we got a clear
conception about the relation between consumer
behavior and different online e-commerce services.
Methodology(Cond.)
pie chart and bar chart to understand the
consumer behavior graphically that can be used to
make people understand easily.
we trained our preprocessed dataset with some
classification machine learning algorithm. 70%
data were used to train our model and the
remaining 30% data were used to test the validity
of our machine learning algorithm
we created a trained model those were used to
predict the consumers behavior
Result Analysis
Analyzing the consumer behavior for
purchasing products we have used here some
Machine Learning Algorithms and shows 3
cases for purchasing observed here- only view,
add to cart or purchase. Resulted output:-
1- indicate that the product is only view by the
consumer.
2- indicates that the product is added to cart .
3- indicates that the product is purchased by the
consumer.
Result Analysis(contd.)

Python as an experimental tool,


implementing algorithms, deriving prediction
results, and conducting comparative analysis.
Result Analysis(Contd.)
Consumers Purchasing Behavior
Category Brand Price($) Predicted
Output
Computer. notebook HP 1(view)
Smartphone Samsung 3(purchase)
Audio.Headphone Xiaomi 2(cart)

USB _Pendrive SanDisk 3(purchase)


Television Sony 1(view)
kitchen.refrigerators LG 2(cart)
Water_Heater Prestige 1(view)

Rice Cooker Panasonic 3(purchase)


CONCLUSION
The main objective of this research work
is to assist the owners of the E-commerce
websites
We found that Random Forest algorithm
provides the better prediction than any
other algorithms.
Such factors identified by the owners of e-
commerce organizations will help them
take necessary steps to provide a better
service to the consumers and add more
lifetime value to their business.
REFERENCE

[1]Z. L. Y. L. Jiangtao Qiu, "Predicting customer


purchase behavior in the e-commerce context,"
Springer, 2015.
[2] U. Z. A. A. L. T. Ben Yedder, "Modeling prediction in
recommender," in IEEE, 2017.
[3] J. H. C. Z. Y. C. Yanen Li, "Improving one-class
collaborative filtering by incorporating rich user
information," in ACM International Conference on
Information and Knowledge Management, 2010.
.
REFERENCE

THANK
YOU

You might also like