Continuous Alternate Assessment –II
Machine Learning
(R204GA33601)
III [Link] II Semester
Regulation:SRIT-R20
Department of Computer Science & Engineering(AI&ML)
Documentation on Mini-Project:
Name of the Project :
Name of the Student :
Roll number :
Last Date to Submit :
Date of Submission :
Date of Issue :
Signature of the Faculty :
Under the Guidance of:
[Link]
Associate Professor(Hod)
(2024-2025)
AI Based Fake News Detector
Overview:
The widespread use of digital platforms has led to the rapid spread of fake
news, causing misinformation and public confusion. Manual detection is
inefficient and not scalable, highlighting the need for automated solutions.
AI-based fake news detection uses machine learning (ML) and natural language
processing (NLP) to analyze and classify news as real or fake. The system is
trained on labeled datasets and involves key steps like data collection, text
preprocessing, feature extraction, model training, and evaluation.
This technology helps reduce misinformation, supports media integrity, and
enables users to access more reliable information online.
Aim:
To develop a system that automatically identifies and classifies news articles or
social media content as real or fake using artificial intelligence techniques such
as natural language processing (NLP) and machine learning (ML), thereby
helping to prevent the spread of misinformation and improve the reliability of
online information.
Key Features:
• Automated Text Analysis
• Machine Learning Classification
• Feature Extraction
• Real-Time Detection
• Scalability
(2024-2025)
Purpose:
The purpose of AI-based fake news detection is to automatically identify and
classify misleading or false information using machine learning and natural
language processing. This helps reduce the spread of misinformation, promotes
reliable information sharing, and supports informed decision-making among the
public and media platforms.
(2024-2025)
Procedure to Develop Al-Based Fake
News Detector:
Define the Problem:
Clearly state the objective: classify news articles as real or fake.
Collect and Prepare the Dataset:
Gather labeled datasets (e.g., FakeNewsNet, LIAR dataset, Kaggle
datasets).
Ensure the dataset includes both fake and real news.
(2024-2025)
Data Preprocessing:
Clean the text (remove punctuation, numbers, stopwords, etc.).
Normalize text (lowercasing, stemming/lemmatization).
Handle missing or duplicate data.
Model Evaluation:
Evaluate using metrics like:
Accuracy
Precision
Recall
F1-Score
Confusion Matrix
Optimization and Tuning
(2024-2025)
Deployment:
Build a user interface or web app using Flask, Django, or Streamlit.
Deploy the model for real-time fake news detection.
(2024-2025)
Monitoring and Maintenance
Continuously update the model with new data to adapt to evolving language
and news trends.
Conclusion:
AI-based fake news detection provides an effective solution to combat the
rapid spread of misinformation in the digital age.
By leveraging machine learning and natural language processing, the system
can analyze and classify news content with high accuracy. This not only aids in
promoting reliable information but also supports social media platforms, news
agencies, and the general public in identifying and preventing the influence of
false content. Continuous improvement and real-time deployment of such
models can play a vital role in ensuring a more informed and trustworthy
information ecosystem.
(2024-2025)