Velmart- Creating smart cart using Apriori and FP growth algorithm in an e-

commerce website
Anand Agarwal| Prof. Shobha Rekh | SCOPE
Introduction Results
This document is meant to delineate the features of VELMART, so as to serve as a
TestID Test Name Test Data ExpectedOutput Actual Output Status (Pass/ Fail)
guide to the developers on one hand and a software validation document for the
prospective client on the other. The Online Shopping System for the shop website
and application is intended to provide complete solutions for vendors as well as
customers through a single get way using the internet. Test Case Login Using invalid credentialsEmail- abc@gmail Invalid Email InvalidEmail Fail
1
Password- 123

Motivation
In today’s rapidly evolving e-commerce landscape, businesses are constantly seeking Test Case Login Using valid credentials Login Successful Login Successful Pass
2
innovative ways to enhance customer experiences and streamline their shopping Email- abc@gmail.comPassword- 123

processes. One such solution that has gained considerable traction is the implementation
of smart carts.
SCOPE of the Project Test Case
3
Updating user
name
New Name- xyz Name Updated
Successfully
Name Updated
Successfully
Pass

The scope of this project is to develop and implement Velmart, a smart cart
system, utilizing the Apriori and FP-Growth algorithms within an e-commerce Test Case Update Password Using DifferentPassword- Passwordsnot match Passwords not Fail
4 match
website. The focus is on integrating the smart cart functionality, implementing a Old Password- 123
New Password- 12345
recommendation engine, and ensuring seamless integration with the existing Confirm New Password-1234
website infrastructure. The scope also includes user experience enhancements,
thorough testing and quality assurance, and documentation for successful
deployment of the system.
Test Case Change Password Using DifferentPassword- Password changed Password changed Pass
5 successfully successfully
Old Password- 123

Methodology New Password- 12345

Confirm New Password-12345

Data Collection and Preprocessing: Gather transactional data from the e-

commerce website, including customer purchase history and product information.
Preprocess the data by removing duplicates, handling missing values, and
transforming it into a suitable format for algorithmic processing. Test Case Add product tocart Adding a product to cart Product added Product added Pass
6 successfully successfully

Apriori Algorithm: a. Generate Frequent Itemsets: Apply the Apriori algorithm to

identify frequent itemsets in the transaction data. Start with individual items and Test Case Search Search item in search bar Item displayed Item displayed Pass
7
progressively generate larger itemsets by pruning infrequent candidates based on
support thresholds. b. Generate Association Rules: From the frequent itemsets,
generate association rules by considering different confidence thresholds.
Evaluate the strength of the rules based on support and confidence measures.

FP-Growth Algorithm: a. Build the FP-Tree: Construct the FP-Tree data structure
using the transaction data. This tree represents the frequency and hierarchy of
itemsets. b. Generate Frequent Itemsets: Traverse the FP-Tree to identify frequent
itemsets by employing a recursive process. Prune infrequent itemsets based on
minimum support thresholds. c. Generate Association Rules: From the frequent
itemsets, generate association rules by considering different confidence
thresholds. Evaluate the rules based on support and confidence measures.

Integration with Smart Cart: a. Incorporate Algorithmic Results: Integrate the

generated frequent itemsets and association rules into the smart cart system. This
involves linking recommended products or related items to the cart interface
based on the user's current selection or transaction history. b. Personalization and
Real-Time Updates: Implement mechanisms to provide personalized
recommendations to users in real-time, taking into account their current cart
contents and browsing behavior. Update recommendations dynamically as the
user interacts with the system.

Testing and Evaluation: a. Validate Algorithm Outputs: Test the Apriori and FP-
Growth algorithms by providing sample data and verifying if the generated
frequent itemsets and association rules are accurate and comply with the defined
Conclusion
thresholds. b. Performance Testing: Evaluate the computational efficiency and We created a website for basically those who can’t or don’t want to visit our store
scalability of the algorithms by testing them with large datasets and monitoring because of distance or lack of time, so this provides users with all the products
their runtime and memory usage. c. User Experience Testing: Assess the which are available on store to them on the website online, so as they can buy
effectiveness and usefulness of the recommendations provided by the algorithms products from then and there. This provides us a free hand also over the store as
within the smart cart system. Solicit feedback from users and iterate on the we can directly access the website and can update products and manage
recommendation generation process based on their preferences. everything corresponding to the user, products and everything related to the
website. So it is an online website (VELMART) for shopping products like grocery
etc easily and efficiently.

References
[1] He JunHua. Design and Implementation for an Ecommerce Development. [2]
Rashad Yazdanifard1 , Noor Al-Huda Edres2 and Arash Pour Seyedi3 . Security and
Privacy Issues as a Potential Risk for Further E- commerce Development(Assoc.
Prof. Dr. Rashad Yazdanifard) [3] Wang, C., & Zhang, P. The Evolution of Social
Commerce: The People, Management, Technology, and Information Dimensions
Saad Subair. The Evolution of Software Process Models: From the Waterfall Model
to the Unified Modeling Language (UML). [4] Stanley Frederick W.T. Lim, Xin Jin,
Jagjit Singh Srai. Consumer-driven e-commerce: A literature review, design
framework, and research agenda on last-mile logistics models (Li et al., 2011)