An Approach for Detecting False News Using Machine Learning Techniques
Author : Vijay P, Vishal Surya AM and Revathy S
Abstract :
The rapidly changing digital landscape offers numerous advantages, yet is not devoid of drawbacks, with fake news standing out as a significant concern. Fake news detection involves discerning genuine information from deliberately misleading content masquerading as news. This issue holds paramount importance in the modern digital era due to its potential to swiftly disseminate through online platforms, fostering misinformation and polarization. The ease with which false news can be propagated underscores the need for solutions. Here, Machine Learning (ML) emerges as a pivotal tool.
ML encompasses AI's capacity to enable systems to learn and execute diverse tasks. Addressing the complex challenge of fake news detection, various ML algorithms, including supervised, unsupervised, and reinforcement techniques, come into play. Although articles sharing similar word counts might diverge vastly in meaning, this study delves into distinct textual attributes capable of distinguishing authentic content from fabricated ones. By harnessing these attributes, a fusion of supervised ML models-like Logistic Regression, Support Vector Classifier, Passive Aggressive Classifier, Random Forest Classifier, Decision Tree Classifier, and K Nearest Neighbour Classifier-are trained and assessed on extensive real-world datasets.
This approach involves training on labeled instances of both genuine and false news, empowering the models to classify new examples accurately. The proposed endeavor involves curating a comprehensive dataset encompassing both fake and authentic news articles. By meticulously employing data pre-processing techniques and algorithms, the objective is to establish a robust model capable of categorizing articles as genuine or fake based on their textual content, encompassing words and phrases. Through this, the study seeks to achieve high accuracy in classification, thereby contributing to the battle against fake news dissemination.
Keywords :
Machine learning, supervised learning, logistic regression, support vector classifier, passive aggressive classifier, random forest, decision tree
