Table of Contents

1. Introduction .................................................................................................................. 4

2. Problem Statement ....................................................................................................... 4

3. Objectives .................................................................................................................... 5

4. Methodology ................................................................................................................ 5

a. Requirement identification........................................................................................ 6

i. Study of Existing System ...................................................................................... 6

ii. Literature Review .............................................................................................. 7

iii. Requirement Analysis........................................................................................ 8

b. Feasibility Study ....................................................................................................... 8

i. Technical Feasibility ............................................................................................. 8

ii. Operational Feasibility ...................................................................................... 9

iii. Economic Feasibility ......................................................................................... 9

c. High Level Design of System ................................................................................... 9

i. System Flow Chart ................................................................................................ 9

ii. Description of Algorithms ............................................................................... 10

5. Gantt Chart ................................................................................................................. 12

6. Expected Outcome ..................................................................................................... 13

7. References .................................................................................................................. 14
 Table of Figures

Figure 4:1 Waterfall model .................................................................................................. 5

Figure 4:2 System Flow Chart ............................................................................................. 9

 Table of Tables
Table 5:1Gantt Chart.......................................................................................................... 12
1. Introduction
In the current digital era, social media platforms have become an essential part of daily
communication, content sharing, and community building. Platforms such as Instagram,
Facebook, and Twitter have transformed how people interact and share information across
the globe. With increasing internet penetration and smartphone usage, the demand for
intuitive and feature-rich social networking applications continues to grow rapidly.

SocialConnect is a proposed web-based social media platform inspired by Instagram,

designed to allow users to share photos and videos, follow other user, engage through likes
and comments, and discover content through a personalized feed. This project aims to
develop a fully functional, responsive, and scalable web application that demonstrates
modern web development practices and software engineering principles.

The platform will be built using contemporary technologies including a [Link] frontend,
a [Link]/[Link] backend, and a MongoDB database, following RESTful API
architecture. This project serves as a comprehensive demonstration of full-stack web
development skills acquired throughout the academic program.

2. Problem Statement
Existing social media platforms present several significant challenges, including growing
concerns over user privacy, extensive data monetization, and the influence of opaque,
algorithm-driven content feeds. These systems often limit users’ control over what content
they see and how their own content is shared, reducing transparency and trust. In addition,
the presence of excessive advertisements creates a cluttered and distracting user experience.

Furthermore, most platforms offer limited customization options, making it difficult for
users and niche communities to tailor the platform to their specific needs. The lack of open,
flexible, and user-centric design restricts innovation and user engagement. As a result, there
is a clear need for a social networking solution that prioritizes privacy, transparency, user
control, and a clean, customizable experience.
3. Objectives
The primary objective of this project is to design and develop a scalable, secure, and user-
friendly social media web application with modern features and best development
practices.

 To provide core social networking features such as user profiles, posts, likes,
comments, follows, and messaging.
 To ensure the application is scalable, maintainable, and user-friendly.

4. Methodology
This project follows the Waterfall Model, a linear and sequential software development
approach where each phase must be completed before moving to the next.

Figure 4:1 Waterfall model

 Requirement Analysis:

Identify and define the overall system requirements and user needs.

 System Design:

Plan the system architecture, database structure, and user interface.

 Implementation:

Develop the application by building the frontend, backend, and integrating all
components.

 Testing:
 Verify the system functionality and fix any errors or issues.

 Deployment:

Launch the application in a live environment for users.

 Maintenance:

Monitor, update, and improve the system after deployment.

a. Requirement identification
i. Study of Existing System

Existing social media platforms such as Instagram, Facebook, and Twitter are
widely used for communication, content sharing, and online interaction. These
platforms allow users to create profiles, share photos and videos, follow other users,
like and comment on posts, and communicate through messaging systems. They are
designed to handle millions of users and provide fast, scalable, and feature-rich
experiences.

Most of these systems use advanced technologies such as cloud storage,

recommendation algorithms, and real-time communication to enhance user
engagement. Features like news feeds, stories, reels, and notifications are powered
by algorithms that analyze user behavior and preferences to display personalized
content.

However, despite their advanced capabilities, these platforms have several

limitations. One major concern is privacy, as user data is often collected, analyzed,
and monetized for advertising purposes. Users have limited control over how their
data is used and shared. Another issue is the lack of transparency in content
algorithms, where users are not fully aware of how content is prioritized or filtered
in their feeds.

Additionally, the heavy presence of advertisements can disrupt the user experience,
making platforms feel cluttered and less user-focused. Customization options are
also limited, preventing users and communities from modifying the platform
according to their specific needs. Moreover, these systems are closed-source, which
restricts learning and experimentation for students and developers.
Overall, while existing systems are powerful and widely adopted, their limitations
in privacy, transparency, customization, and user control highlight the need for a
more user-centric and open alternative.

ii. Literature Review

Social networking systems have significantly transformed online communication

by enabling users to create profiles, share content, and interact globally. According
to Boyd and Ellison, social networking sites are web-based services that allow
individuals to construct public or semi-public profiles and connect with other users
within a system [1]. These platforms form the foundation of modern social media
applications.

Kaplan and Haenlein explain that modern social media is built on Web 2.0
technologies, which enable user-generated content and interactive communication
[2]. This evolution has led to highly engaging platforms such as Instagram and
Facebook, which support multimedia sharing, real-time interaction, and algorithm
based content delivery.

However, privacy remains a major concern in social networking systems. Acquisti

et al. highlight that users often unknowingly share personal data, which is collected
and monetized by platforms for advertising purposes, raising serious privacy and
security issues [3].

Furthermore, Pariser introduces the concept of the “filter bubble,” where

recommendation algorithms limit users’ exposure to diverse content by showing
only personalized information based on past behavior [4]. Recent research
highlights increasing concerns regarding privacy in social networks. The IEEE
study on large-scale social data emphasizes that user information is often collected
and analyzed for commercial purposes, raising serious privacy risks [5]. Similarly,
blockchain-based and deep-learning approaches have been proposed to enhance
privacy preservation in social networks [6]. This reduces transparency and can
restrict access to varied perspectives.

From the reviewed literature, it is clear that while existing social media platforms
offer advanced features and scalability, they still face challenges related to privacy,
 transparency, and user control. These gaps provide opportunities for developing
more user-centric and transparent systems.

iii. Requirement Analysis

 Functional Requirements
The system will provide core social media features to support user
interaction and content sharing. Users will be able to register, log in, manage
their profiles, and securely access the platform. They can upload photos and
videos with captions, view an infinite scrolling feed, and interact with posts
through likes, comments. The system will also support social features such
as following other users and discovering other user through search and
content through hashtags. In addition, users will be able to send real-time
messages and receive notifications for interactions.
 Non-Functional Requirements
The system is designed to be fast, secure, and scalable. It should load pages
quickly to ensure a smooth user experience. Security will be ensured
through encrypted communication and secure authentication methods. The
application will be built in a way that allows future expansion and handling
of more users. It will also be responsive, meaning it will work properly on
mobile, tablet, and desktop devices. Finally, the system will aim for high
availability to ensure users can access it with minimal downtime.

b. Feasibility Study
i. Technical Feasibility

The project is technically feasible as it will be developed using the MERN stack,
which is widely used, well-documented, and supported by a large open-source
community. The technologies required for development such as [Link] for
frontend, [Link] and [Link] for backend, and MongoDB for database
management are easily accessible and commonly used in modern web applications.
Additional tools like JWT for authentication, [Link] for real-time
communication, and Cloudinary for media storage further enhance the system
capabilities. Since all these technologies are freely available and familiar to the
development team, the project can be implemented successfully within the
academic environment.
 ii. Operational Feasibility

The system is operationally feasible because it is designed to be user-friendly and

similar to existing social media platforms, making it easy for users to understand
and use. Since it is a web-based application, users can access it directly through a
browser without installing any software. The interface will be simple and intuitive,
reducing the learning effort required.

iii. Economic Feasibility

The project is economically feasible because it is developed using open-source

technologies that do not require licensing costs. Most of the tools and platforms
used, such as MongoDB Atlas (free tier), GitHub offer free plans suitable for
development and academic purposes. Therefore, the project can be completed with
minimal to no financial investment.

c. High Level Design of System

i. System Flow Chart

Figure 4:2 System Flow Chart

The flow starts when a user opens the web application in a browser. The system
first checks if the user is logged in. If the login details are wrong, access is denied;
if they are correct, the user is allowed to enter the system. After successful login,
the user can use the main features like the feed, profile, explore page, and messaging
section. These features allow the user to interact with the application. All the user
data and activities are stored in a database (MongoDB) to keep everything saved
and updated. Finally, when the user finishes using the app, they log out to safely
end the session.

ii. Description of Algorithms

SocialConnect is built on three foundational algorithms that power its

recommendation, search, and social discovery features. Each algorithm is selected
based on proven techniques in information retrieval and graph theory, widely used
in real-world platforms such as Google Search, LinkedIn, Netflix, and Facebook.

 TF-IDF with Cosine Similarity -Post Recommendation (Explore Page)

Purpose:
This algorithm recommends posts to users based on their interests derived
from previously liked content.
Overview:
The system uses a content-based recommendation approach where each post
is analyzed using textual features such as captions and hashtags. The TF-
IDF technique converts text into weighted numerical vectors, capturing the
importance of words in a post relative to the entire dataset. Cosine Similarity
is then used to measure how closely a post matches a user’s preference
profile [7].
Working Process:
 Each post is treated as a document.
 TF-IDF transforms text into vector representations.
 A user profile vector is created by averaging vectors of liked
posts.
 Cosine Similarity compares user preferences with unseen posts.
 Top-ranked posts are recommended.
 Key Idea: Posts with similar semantic meaning to previously liked content
are prioritized.

 BM25 Ranking — Smart Search System

Purpose:

Provides highly relevant ranked results for posts, users, and hashtags based
on user search queries.

Overview:

BM25 (Best Matching 25) is an advanced probabilistic retrieval model that

improves traditional keyword matching. It is widely used in modern search
engines such as Elasticsearch and Apache Solr. It ranks documents by
considering term frequency, document length, and inverse document
frequency [8].

Working Process:

 User enters a search query.

 Each document (post/user/hashtag) is scored using BM25.
 Frequent but meaningful terms are weighted higher.
 Long documents are normalized to avoid bias.
 Results are sorted by relevance score.

Key Idea: Balances relevance and fairness by preventing keyword spam and
document length bias.

 Graph-Based User Suggestion -People You May Know

Purpose:

Suggests new users to follow by analyzing social connections within the

network.

Overview:

The social network is modeled as a graph where users are nodes and follow
relationships are edges. The algorithm identifies second-degree connections
(friends of friends) and ranks them based on mutual connections [9].
 Working Process:

 Construct a directed graph of users.

 Extract first-degree connections (users already followed).
 Find second-degree connections through neighbors.
 Count mutual connections between users.
 Rank suggestions by highest shared connections.

Key Idea: Users with more mutual friends are more likely to be relevant
connections.

5. Gantt Chart
The project is planned over 16 weeks. The following table outlines the key milestones and
their scheduled timeframes:

Table 5:1Gantt Chart

Activity W1 W2 W3 W4 W5 W6 W7 W8 W9 W10 W11 W12 W13 W14 W15 W16

[Link],Identification ■ ■

& Literature Review

2. Feasibility study ■ ■

[Link] ■ ■ ■

(Architecture, DB Schema, UI
Wireframes)

4. Frontend Development ■ ■ ■ ■

([Link] UI Components)

5. Backend Development ■ ■ ■ ■ ■

(API, Auth, Database)

6. Media Upload & Cloud ■ ■

Storage Integration

7. Real-Time Features ■ ■ ■
8. Testing ■ ■ ■

9. Bug Fixes & Performance ■ ■

Optimization

10. Deployment & ■ ■ ■

Documentation

11. Final Report Writing & ■ ■

Presentation Prep

6. Expected Outcome
After the successful completion of this project, a fully functional and responsive social
media web application will be developed that can be accessed through modern web
browsers.

Technically, the system will provide secure user authentication, real-time messaging, and
notification features. It will also include RESTful APIs for communication between
frontend and backend, along with cloud-based media storage and a responsive user
interface suitable for all devices. The database will be properly structured to ensure efficient
data handling.

From an academic perspective, this project will demonstrate practical knowledge of full-
stack development using the MERN stack and the application of software engineering
principles such as design, development, testing, and deployment. It will also provide
valuable hands-on experience in building a real-world scalable web application.
7. References

[1] D. M. B. a. N. B. Ellison, "Social network sites: Definition, history, and

scholarship," Journal of Computer-Mediated Communication, vol. 13, no. 1, p.
210–230, 2007.

[2] A. M. K. a. M. Haenlein, "Users of the world, unite! The challenges and

opportunities of Social Media," Business Horizons, vol. 53, no. 1, p. 59–68, 2010.

[3] L. B. a. G. L. A. Acquisti, "Privacy and human behavior in the age of information,"

vol. 347, no. 6221, p. 509–514, 2015.

[4] E. Pariser, "The Filter Bubble: What the Internet Is Hiding from You," USA:
Penguin Press, New York, 2011.

[5] I. A. Editorial, "Privacy preservation for large-scale user data in social networks,"
IEEE Access, vol. 10, p. 4374–4379, 2022.

[6] S. A. F. e. al, "A deep decentralized privacy-preservation framework for online

social networks," Blockchain: Research and Applications, 2024/2025.

[7] G. S. a. C. Buckley, "Term-weighting approaches in automatic text retrieval,"

Information Processing & Management, vol. 24, no. 5, p. 513–523, 1988.

[8] S. R. a. H. Zaragoza, "The Probabilistic Relevance Framework: BM25 and

Beyond," Foundations and Trends in Information Retrieval, vol. 3, no. 4, p. 333–
389, 2009.

[9] L. A. A. a. E. Adar, "Friends and neighbors on the web," Social Networks, vol. 25,
no. 3, p. 211–230, 2003.