Scribd
Upload
217 views21 pages

Cyberbullying Detection Based On Semantic-Enhanced Marginalized Denoising Auto-Encoder

This document presents a system for detecting cyberbullying using a semantic-enhanced marginalized denoising auto-encoder (smSDA) model. The smSDA model incorporates semantic information into representation learning to better capture bullying signals. The system allows users to register, post messages to a social media environment, analyzes messages for sentiment, predicts whether they contain bullying using the smSDA model, and filters any identified bullying messages. Future work could extend the system to also predict and prevent bullying images and videos.

Uploaded by

mohammed yaseen
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
217 views21 pages

Cyberbullying Detection Based On Semantic-Enhanced Marginalized Denoising Auto-Encoder

This document presents a system for detecting cyberbullying using a semantic-enhanced marginalized denoising auto-encoder (smSDA) model. The smSDA model incorporates semantic information into representation learning to better capture bullying signals. The system allows users to register, post messages to a social media environment, analyzes messages for sentiment, predicts whether they contain bullying using the smSDA model, and filters any identified bullying messages. Future work could extend the system to also predict and prevent bullying images and videos.

Uploaded by

mohammed yaseen
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 21

CYBERBULLYING DETECTION BASED ON

SEMANTIC-ENHANCED MARGINALIZED
DENOISING AUTO-ENCODER

PRESENTED BY UNDER THE GUIDANCE OF


A B H I N AV K K ( 1 H M 1 5 C S 0 0 1 ) D R . S H I V M U RT H Y R . C
ARUNPRAKASH PM (HM15CS005) BE, M-TECH, PH.D

M O H A M M E D YA S E E N P ( 1 H M 1 5 C S 0 1 5 ) HOD. DEPT OF CS&E


ABSTRACT
 Cyberbullying has emerged as a serious problem afflicting children, adolescents and young
adults.
 Machine learning techniques make automatic detection of bullying messages in social media
possible, and this could help to construct a healthy and safe social media environment.
 Our method named Semantic-Enhanced Marginalized Denoising Auto-Encoder (smSDA) is
developed via semantic extension of the popular deep learning model stacked denoising auto-
encoder.
INTRODUCTION
Cyberbullying can be defined as aggressive, intentional actions performed by an
individual or a group of people via digital communication methods such as sending
messages and posting comments against a victim.

 Different from traditional bullying that usually occurs at school during face-to-face
communication, cyberbullying on social media can take place anywhere at any time.
SYSTEM SPECIFICATION
HARDWARE REQUIREMENTS

Processor : Intel Pentium 4 or more

Memory : 1GB RAM or more

Harddisk : 40GB or more.


Cont…
SOFTWARE REQUIREMENTS

Operating System : Windows XP or Windows 7

IDE : Eclipse Juno 4.2 with Tomcat Server

Visual Interface : Java Server Pages, Servlets

Database : MySQL 5.0


LITERATURE SURVEY
[1] A. M. Kaplan and M. Haenlein, “Users of the world, unite! the challenges and
opportunities of social media,” Business horizons, vol. 53, no. 1, pp. 59–68, 2010.

[2] R. M. Kowalski, G. W. Giumetti, A. N. Schroeder, and M. R. Lattanner, “Bullying in


the digital age: A critical review and meta analysis of cyberbullying research among
youth.” 2014.

[5] S. R. Jimerson, S. M. Swearer, and D. L. Espelage, Handbook of bullying in schools:


An international perspective. Routledge/Taylor & Francis Group, 2010.
SYSTEM DESIGN
System Architecture
EXISTING SYSTEM AND ITS DISADVANTAGES
NLP solutions including BoW(Bags of Words)models, LSA(Latent Semantic Analysis)
for representation learning to capture bullying signals in social media.

As an introductory work, they did not develop specialized models for cyberbullying
detection.
PROPOSED SYSTEM
Our proposed Semantic-enhanced Marginalized Stacked Denoising Autoencoder is able
to learn robust features from BoW representation in an efficient and effective way.

Semantic information is incorporated into the reconstruction process via the designing
of semantic dropout noises and imposing sparsity constraints on mapping matrix.
MODULES
REGISTRATION:

oUsing this module user can register to the social media environment that we created to
simulate text based cyber bullying detection so that user will be getting a clear picture about
how it works.

oUser want to enter personal information like user name, password, email id, Mobile
number in the expected format for successful registration.
Cont…
MESSAGE POST :

oWith the help of this module user can post messages.

oThe posted message has been stored in the database for further processing of message
like sentimental analysis and data prediction.
Cont…
SENTIMENTAL ANALYSIS :

oWith the help of this module posted message can be retrieved from database and analyzed
and predict the sentiment of posted message.
Cont…
DATA PREDICTION AND FILTRATION :

oIn this module the output from the sentimental analysis is predicted weather the
message contains any bad words or not

o If the result is positive, the message is filtered so that cyber bullying can be controlled
to an extend.
FUTURE ENHANCEMENT
Semantic-Enhanced Marginalized Denoising Auto-Encoder (smSDA) is developed to
filter bad words in social network messages which can be extended to another level
where bad images and bad video can also be predict and prevented before it reach to
destination user.
CONCLUSION
This paper addresses the text-based cyberbullying detection problem, where robust
and discriminative representations of messages are critical for an effective detection
system. By designing semantic dropout noise and enforcing sparsity, we have developed
semantic-enhanced marginalized denoising auto encoder as a specialized representation
learning model for cyberbullying detection.
Thank You
SNAPSHOTS

Registration
Cont…

Chatbox
Cont..

Message Filter Server


Cont…

Filtered Messages
Cont…

Bullying Graph

You might also like