Integrated Framework for Fake News Detection Using
Deep Learning Functional Neural Networks
Guided by.,
P. SUJITHA(612720104046)
R.MONISHA(612720104028) B.SOWMEYA .ME
T.SOWMIYA(612720104044) A\P-CSE
(DEPARTMENT OF CSE)
ABSTRACT
Newspapers are the main source of news for people around the world. However, there was a surprising
rise in social media spread in later periods due to the significant increase in technology and updates.
Most of the existing methods made use of natural language processing (NLP), machine learning(ML).We
will do this project in Deep learning(DL).
This method's efforts have been made to automate the process of fake news detection.
This existing machine learning method has less classification accuracy, and multiple combination
Machine learning methods have not consumed memory usage; it provides a higher false rate.
This technique significantly achieved higher accuracy compared to existing methods for early detection
of fake news.
AREA AND EXPLANATION
The advent of the World Wide Web and the rapid adoption of social media platforms paved the way for
information dissemination that has never been witnessed in the human history before.
Fake news is defined as a deviation from other observations, and the suspicions obtained are that
different institutions generate the observations.
Regardless of the development of machine learning and deep learning technology, it has received a
great deal of attention from research groups in different industries.
The overall probability of the alignment is called the normalization in the complex adjustment to
obtain the distribution's adjustment value.
CONT.,
Logistic Model
is a generalized linear model that is used for binomial regression in which the predictor variables
can be either numerical or categorical.
Neural Networks
The neural network is a technique that imitates the functionality of the
human brain using a set of interconnected vertices.
Fuzzy Logic
Fuzzy Logics a multi-valued logic which is derived from fuzzy set theory to deal with the
reasoning that is approximate rather than precise.
Similar to Boolean Logic, Fuzzy Logic has three basic operations: intersection, union,
and complement.
CONT.,
Natural Language Processing is a method of cutting, extracting, and transforming languages
utilized in the system to create new data so that we can get insightful knowledge.
Analyze feature extraction methods with different supervised classifiers to identify which
classifier yields best results.
Develop an unsupervised learning model making use of clustering algorithms for Fake
news detection.
Logistic regression is also called as model with minimal neural network.
Optimize the unsupervised Topic Modeling method using a suitable optimizer for rapid fake
news detection
EXISTING METHOD
This existing work considers some forms of news in social networks, integrates the text content,
images, propagation, and user information of publishing news to improve the performance of fake news
detection.
Next, to obtain the hidden information and internal relationship between features, Bidirectional GRU
layer and Self-Attention layer are added into the auto encoder, and then reconstruct residual to detect
fake news.
This paper treats fake news as an abnormal data in social networks and proposes a method based on
auto encoder for unsupervised fake news detection with various features, namely UFNDA.
CONT.,
After training UFNDA with real news data, the UFNDA has the ability to detect fake news.
By testing the test set, to obtain the reconstructed residual to distinguish real news and fake
news.
Previous efforts proposed single models as an effective way to solve the fake news problem on
digital platforms.
Do not find efforts that explore automatic fake news detection focusing on investigating
whether there is a set of features that remain useful to build models with low performance.
DEEP LEARNING
Deep learning is a sub field of machine learning that very closely tries to mimic human brain’s
working using neurons.
It works technically in the same way as machine learning does ,but with different capabilities and
approaches.
Deep learning models are capable enough to focus on the accurate features
themselves by requiring a little guidance from the programmer.
Deep learning is implemented with help of Neural Network.
PROPOSED METHOD
The proposed Modeling (Latent Dirichlet Allocation-LDA) and Deep leaning (Recurrent Convolutional
Neural Network) method analyses the fake or real news from a given dataset.
This word analysis process uses a deep neural network, and it extracts the word vector of each news data
and title.
There are different techniques namely Latent Semantic Analysis , Probabilistic Latent Semantic Analysis
and Latent Dirichlet Allocation .
It presents a pre-processing task to remove noise and inconsistent data obtained from various sources using
a Count Victories distribution.
LDA is a type of generative probabilistic model used to construct data points and estimate probabilities.
LDA creates probabilities for the words that are used to form themes, which are then categorized into texts.
All the input weights are evaluated, and the error estimate updates the hidden neuron.
REFERENCE
1. MINJUNG PARK AND SANGMI CHAI Ewha School of Business, Ewha Womans University,
Seodaemun-gu, Seoul 03760, South Korea Corresponding author: Sangmi Chai ,”Constructing a User
–Centered Fack News Detection Model by Using Classification Algorithms In Machine Learning
Techniques”,Vol-11,pp.12 July,2023.
2. DAOKANG WANG,WUBO ZHANG ,WHNHUAN WU,AND XIAOLEI GUO Electrical and
Information Engineering Department ,Hubei University of Automotive
Technology ,Shiyan442002,China Corresponding author:Wubo Zhang,”Soft-Label for Multi-Domain
Fack News Detection”,Vol-11,PP.11 Sep ,2023.
3. AHMED HASHIM JAWAD ALMARSAHY,MOHAMMAD-REZA FEZI-DERAKHSHI,AND
PEDRAM SALEHPOUR ComInSys laboratory,Department of Computer Engineering,University of
Tabriz,Tabriz 51666,Iron “Enhancing Eack News Detection by Multi-Feature Classification”,Vol-
11,PP. 5 Dec,2023.
THANK YOU
