A
SYNOPSIS
On
FactFlow AI : Daily Insight For Competitive Exam Aspirant
Abstract :-
Fact Flow AI : Daily Insight For Competitive Exam Aspirant is an AI-powered news platform that
simplifies complex information into easy-to-read daily updates. The system automatically scans multiple
news sources, removes irrelevant content, and uses advanced AI to summarize the most important
stories. It is designed to provide users with exactly 15 essential news highlights every day, saving them
time and effort. This project offers a clean and efficient way for students and general readers to stay
informed without being overwhelmed by digital noise.
1. Introduction (Overview and project background):-
Fact Flow AI : Daily Insight For Competitive Exam Aspirant is an automated, AI-driven news intelligence
platform designed to simplify how people consume information. In today’s digital world, readers are
constantly overwhelmed by an endless stream of news, much of which is repetitive, irrelevant, or
distracting. This "information overload" makes it difficult for both students and general readers to
identify the most significant events of the day. The background of this project lies in the need for a
focused and high-quality news experience. Traditional news websites often prioritize clicks and speed
over depth and relevance. FactFlow: Daily Insight solves this by using a sophisticated 5-stage AI
pipeline to harvest, filter, and rank news stories. By utilizing the Gemini AI model, the system cuts
through the digital noise to deliver exactly 15 essential, summarized news highlights every day. The
project aims to provide a reliable "Vault" of knowledge that saves time while ensuring the reader is well-
informed on the most impactful global and national developments.
2. Literature Survey/Review :-
[1] Y. -T. Xie and C. -H. Tseng, "Memer: News Meme Generator Using Gpt-3: Attention Of
Young Generations To Current Affairs," 2023 IEEE 3rd International Conference on Social
Sciences and Intelligence Management (SSIM), Taichung, Taiwan, 2023.
Young people are paying less attention to current affairs and using social media most of the time in the
information explosion. Recently, memes have been widely used by the young for entertainment. For
current affairs, we investigated whether memes could be used for current affairs using an automatic
meme generator. The meme generator was integrated into a social system on websites to increase the
younger generation's attention to current affairs
Vision of department: To lead in technical, quality education, innovation, research for development of sustainable & inclusive technology for the
society.
Mission of department: 1. To create ambience of academic excellence through state of art infrastructure. 2. To create student-centric pedagogy that
will lead to employability.3. To create a software engineering professional with knowledge of multidisciplinary fields, can provide innovative products
& service to society.4. To train and motivate the students for lifelong learning, employability, and entrepreneurship
[2] H. Joshi, O. Ambalkar, J. Gundawar, R. Bagade, P. Pandit and A. Yenkikar, "PrepMate An
AI-Powered RAG System with News Summarization and Integrated Chatbot for Indian
Languages," 2025 International Conference on Computational Innovations and Sustainable
Technologies (ICCIST), Bangalore, India, 2025
The proposed AI-powered News Aggregator Platform is designed to simplify and optimize current
affairs preparation for UPSC/MPSC aspirants. By leveraging web scraping with BeautifulSoup and
ChromaDB as a vector database, the system efficiently fetches, stores, and retrieves news articles from
Times of India and BBC News. It integrates four specialized chatbots: a general chatbot for query
resolution, a document-based chatbot for extracting insights from uploaded study materials, a scraped
news chatbot for contextual responses from stored articles, and a fine-tuned chatbot trained on
UPSC/MPSC-style QA to deliver structured answers.
[3] E. Plaku, K. Jahaj, A. Cela and N. Civici, "A Machine Learning Framework for Automated
News Article Title Classification in Albanian," 2024 International Conference on INnovations in
Intelligent SysTems and Applications (INISTA), Craiova, Romania, 2024.
Automated news article classification is a method of categorizing textual data into predefined classes.
Addressing this problem finds applications in diverse domains, including information retrieval, topic
modeling, sentiment analysis and content recommendation systems. In Albanian, though there is a rapid
increase of digital content, there is limited availability of text corpora, presenting significant obstacles for
advancement of natural language processing research and applications. The contribution of this paper is
twofold. First, we introduce a dataset consisting of 9600 news article titles spanning across various
categories. Second, we utilize this dataset to assess the effectiveness of several machine learning
algorithms for topic classification. Experimental results demonstrate the efficacy of recurrent neural
networks in comparison to simpler classifiers and ensemble methods.
[4] K. Panetta, "Lessons Learned From Our Friend, Dr. Arthur Winston: A man of action
[Amperes: Current Affairs From Around the World]," in IEEE Women in Engineering Magazine,
vol. 19, no. 1, pp. 44-46, June 2025.
Provides society information that may include news, reviews or technical notes that should be of
interest to practitioners and researchers presented research on efficient regular expression evaluation,
which is widely used in text filtering and pattern matching. Their findings are important for systems that
need to remove irrelevant or noisy textual data quickly and accurately. In the context of FactFlow, such
filtering techniques are useful in the early stage of the pipeline for eliminating duplicate, clickbait, or
non-essential content before AI summarization begins.
[5] K. N, R. N, A. A. V, S. G, T. J and V. S, "Fine Tuning Large Language Models for Accurate
Fake News Detection," 2025 International Conference on Signal Processing, Computation,
Electronics, Power and Telecommunication (IConSCEPT), Karaikal, India, 2025.
In the current digital age, false information spreads quickly via social media and other online channels,
making it challenging to tell real news from fake. This study introduces a framework for detecting fake
news that uses Llama 3 for explanation-based classification in conjunction with conventional machine
learning models. Using TF-IDF vectorization and SMOTE for improved balance and feature
Vision of department: To lead in technical, quality education, innovation, research for development of sustainable & inclusive technology for the
society.
Mission of department: 1. To create ambience of academic excellence through state of art infrastructure. 2. To create student-centric pedagogy that
will lead to employability.3. To create a software engineering professional with knowledge of multidisciplinary fields, can provide innovative products
& service to society.4. To train and motivate the students for lifelong learning, employability, and entrepreneurship
representation, we experimented with LSTM, SVM, Naive Bayes, Decision Trees, CNN, and Logistic
Regression. Because of its capacity for sequential learning, Decision Tree demonstrated impressive
results, and Llama 3 produced explanations in natural language to elucidate predictions. A speech-to-text
module was also developed, enabling users to speak news statements that are converted into text and
analyzed for [Link] addition to identifying false information, the framework offers context-aware
reasoning, which enhances usability and trust in practical applications.
3. Problem Statement & Objectives :-
The digital news landscape is overwhelmed by information overload, where valuable news is buried
under repetitive, sensational, and irrelevant [Link] students and professionals, identifying
meaningful global events is time-consuming and mentally exhausting.
Objectives :
1. To build an AI-driven pipeline that filters raw news data and retains only high-priority,
relevant information.
2. To use the Gemini AI engine to generate concise, fact-based summaries of top news for quick
daily updates.
5. Requirements Analysis
5.1 Hardware Requirements :
Processor: Intel Core i5 or AMD Ryzen 5 for development; Dual-core CPU for cloud hosting.
Memory (RAM): 8 GB minimum for development environments and 2 GB for production
servers. Storage: 20 GB of available SSD space for fast database operations and application logs.
Connectivity: High-speed internet for real-time RSS harvesting and Gemini AI API communication.
Client Device: Any modern smartphone, tablet, or PC equipped with a standard web browser.
5.2 Software Requirements :
Operating System: Windows 10/11, macOS, or Linux; Ubuntu 22.04 LTS is recommended for
production.
Development Stack: Python 3.10 or higher using the Django 5.0 framework for backend architecture.
Database: SQLite for local testing and PostgreSQL for reliable production data management.
AI & Frontend: Google Gemini API for intelligence; HTML5, CSS3, and Vanilla JavaScript for the
UI.
Tools & Server: Nginx web server, Git for version control, VS Code editor.
Vision of department: To lead in technical, quality education, innovation, research for development of sustainable & inclusive technology for the
society.
Mission of department: 1. To create ambience of academic excellence through state of art infrastructure. 2. To create student-centric pedagogy that
will lead to employability.3. To create a software engineering professional with knowledge of multidisciplinary fields, can provide innovative products
& service to society.4. To train and motivate the students for lifelong learning, employability, and entrepreneurship
6. Expected Outcomes & Applications :
The primary outcome is a functional and autonomous AI news system that transforms massive raw data
into a curated list of significant events. Using the Gemini AI engine, the platform will deliver high
ranking accuracy and objective summaries while eliminating digital clutter. The system will provide
exactly fifteen high-quality highlights daily, successfully demonstrating how artificial intelligence can
solve the problem of information overload for modern readers.
Application :
FactFlow: Daily Insight serves as a critical tool for competitive exam candidates, such as UPSC and SSC
aspirants, by automating current affairs preparation. It also offers busy professionals a way to stay
nationally and globally informed in under five minutes through clear, summarized insights.
Additionally,researchers and content creators can utilize the platform for a filtered, noise-free
feed of daily developments, fostering a more productive and focused reading habit for all users.
7. Conclusion :
FactFlow: Daily Insight is an AI-powered news platform built using Django and the Gemini AI model
to convert raw news into clear, meaningful daily updates. Its five-stage pipeline filters out irrelevant,
fake, and clickbait content, delivering only important and factual news. The platform is especially
useful for students and the general public, as it saves time, reduces information overload, and provides a
smarter, more reliable way to stay informed.
8. References (IEEE format) :
[1] Y. -T. Xie and C. -H. Tseng, "Memer: News Meme Generator Using Gpt-3: Attention Of Young
Generations To Current Affairs," 2023 IEEE 3rd International Conference on Social Sciences and
Intelligence Management (SSIM), Taichung, Taiwan, 2023.
[2] H. Joshi, O. Ambalkar, J. Gundawar, R. Bagade, P. Pandit and A. Yenkikar, "PrepMate An AI-
Powered RAG System with News Summarization and Integrated Chatbot for Indian Languages," 2025
International Conference on Computational Innovations and Sustainable Technologies (ICCIST),
Bangalore, India, 2025
[3] E. Plaku, K. Jahaj, A. Cela and N. Civici, "A Machine Learning Framework for Automated News
Article Title Classification in Albanian," 2024 International Conference on INnovations in Intelligent
Vision of department: To lead in technical, quality education, innovation, research for development of sustainable & inclusive technology for the
society.
Mission of department: 1. To create ambience of academic excellence through state of art infrastructure. 2. To create student-centric pedagogy that
will lead to employability.3. To create a software engineering professional with knowledge of multidisciplinary fields, can provide innovative products
& service to society.4. To train and motivate the students for lifelong learning, employability, and entrepreneurship
SysTems and Applications (INISTA), Craiova, Romania, 2024.
[4] K. Panetta, "Lessons Learned From Our Friend, Dr. Arthur Winston: A man of action [Amperes:
Current Affairs From Around the World]," in IEEE Women in Engineering Magazine, vol. 19, no. 1, pp.
44-46, June 2025.
[5] K. N, R. N, A. A. V, S. G, T. J and V. S, "Fine Tuning Large Language Models for Accurate Fake
News Detection," 2025 International Conference on Signal Processing, Computation, Electronics,
Power and Telecommunication (IConSCEPT), Karaikal, India, 2025.
Sr. No. Name of Students Signature
1 Mangesh Shankar Chavan
2 Atharva Sangram Ghadge
3 Aniket Chandrakant Kumbhar
4 Raj Namdeo Lokhande
5 Vaibhav Ashok Chikane
Batch Coordinator Course Coordinator HOD
Vision of department: To lead in technical, quality education, innovation, research for development of sustainable & inclusive technology for the
society.
Mission of department: 1. To create ambience of academic excellence through state of art infrastructure. 2. To create student-centric pedagogy that
will lead to employability.3. To create a software engineering professional with knowledge of multidisciplinary fields, can provide innovative products
& service to society.4. To train and motivate the students for lifelong learning, employability, and entrepreneurship