AbstractIn this work we tackle one of the problems identified above, that of cyberbullying, which is continuously increasing in the social Web and is becoming a major threat to teenagers and adolescents 3. In fact, 55% of teens using social media have witnessed outright bullying via that medium.1in the digital age there has been increasing in cyber bullying against special community including the women and minorities.Our project focuses on these two issues to detect bullying through test and sentiment analysis in tweets of twitter.We use a dictionary of weighted offensive words along with the presence of pronouns or collective nouns to calculate the Offensiveness percentage in each tweet. We also make use of a Sentiment Analysis to detect sentiment of the tweets.
1.IntroductionPeople share their life over social media and interconnect on daily basis this is the extend of reach in technology today. This has resulted in a vast quantity of data to be generated. In this particular lots of data social media plays an important role specially sites like Facebook, twitter, blogs and so on.
With such large amounts of data, scalability and processing such data with complex algorithms are challenging task.The field of data mining presents a task of lots of unique information. It also presents need of inquisitive ways of gathering and processing of the data. In a way it provides new insights ways into the data from the social network sites.The information gathered from this sites can be of various ways like personal information, their locations or preferences and analysis of text based on sentiment or words. Analysis of sentiment on various subject shows the judgement of people, this is specially done in micro blogging as in twitter. Using such sentiment and analysis of tweets with natural language text processing we can figure out if the tweet in offensive or not.
This is particularly helpful in finding out cyberbullying.This idea of analysis of sentiments and tweets can be formed as forensic subject now taken into forensic linguist as a subject. 2.Research ScenarioTwitter has big potential for data mining as its users produce Big Data that can be processed.
In addition, there are requirements for architecture development that can scale to continuous new-streamed tweets and also ability to integrate with advanced machine learning algorithms. Knowing what users think or how they feel about products is valuable proposition for companies. Sentiment analysis are part of data mining, which monitors public perceptions about various topics. It can analyse what people think about business products and quality, brands, pricing strategies or worldwide trends. Moreover, it can identify business opportunities and thus become an effective factor for companies to innovate their services.
4 Twitter as micro blogging platform backed with its active users create opportunities for data mining and more particular sentiment analysis based on tweets. Twitter users often express their opinions about various topics within their posted tweets. And so by applying text-processing data mining technique can serve companies as feedback or for brand management. On the other hand, since Twitter generates massive volumes of data every day, sentiment analysis can help with marketing related campaigns to research public opinions about newly released product, for example blockbuster movie and analyze sentiment about users satisfaction, whether they felt positive or negative about movie. According to 4 consumers are willing to pay from 20% to 99% more for movie rated with 5/5 stars. This research discusses that positive comments or reviews on product are great influencers and will indicate success among consumers. 3.Problem UnderstandingI.
Synopsis:the world has seen many revolutions that were made possible by Social Media. It is an extremely influential innovation of our time, and is a great way to expand the boundaries of one’s experiences and become socially active. However, social media is a double-sided weapon.it can be either good for business and bad for cyber bullyingAs the size of Twitter© data is increasing, so are undesirable behaviors of its users. A lot of antisocial behavior is observed on social media, including cyber-stalking, cyber-bullying, and cyber-harassment .
one of the major undesirable behavior is cyberbullying, which could lead to catastrophic consequences. Moreover, this is not limited to children and young adults; anybody can be a victim. Hence, it is critical to efficiently detect cyberbullying behavior by analyzing tweets, in real time if possible. Prevalent approaches to identifying cyberbullying are mainly stand-alone, and thus, are time-consuming. I.
Understanding Concept of Cyberbullying:Cyberbullying is formally defined as “willful and repeated harm inflicted through the use of computers, cell phones, and other electronic devices” Patchin, J. W., 2014. In short, bullies typically exploit the use of electronic communication for harassing people. This harassment may be motivated by anger, frustration, revenge, or from a basic desire to control others and feel more powerful (Why do kids, n.d.
). Sometimes kids cyberbully others to cope with their own low self-esteem and/or to fit in with their peers (Why Do People, n.d.). Examples of cyberbullying can include rumors sent by e-mail or posted on social media; embarrassing pictures or videos; and intimidating, insulting, and / or harassing messages posted on social networks.
Once such derogatory messages, pictures, or videos are posted, it is very difficult to take these posts off the social media sites. It can happen 24 hours a day and 7 days a week, and it can even reach its victim when they are alone, outside in the school yard, or in the sports field Patchin, J. W.
, 2014. Cyberbullying empowers a bully to humiliate and hurt the victim in online communities without ever getting recognized. Furthermore, the fear of getting punished or being a social pariah stops victims and bystanders from reporting incidents.
This becomes a difficult problem to control. Cyberbullying behavior is not only unacceptable, but can also lead to catastrophic consequences. Studies performed by The Journal of Psychosocial Research on Cyberspace show that “critical impacts occurred in almost all of the respondents’ cases in the form of lower self-esteem, loneliness and disillusionment and distrust of people: The more extreme impacts were self-harm and increased aggression towards friends and family” Šleglova, 2011. It further mentions that some of the victims developed “coping strategies.” Many times, victims try to deal with cyberbullying all by themselves, which leads to a stressful situation. Additionally, it is tough for parents of the victims to know what is happening with their child online.
In order for support systems to help a victim, they need to identify the cyberbullying or signs of it at the onset. They should not expect the victim to approach them about cyberbullying. This calls for automated cyberbullying watchdog programs that could alert family members regarding cyberbullying.
II. Countermeasures set by twitter:Social networks provide some degree of support for the safe web experience. Tools that help to protect one’s privacy are as follows: Twitter© provides users with the following tools (“Learn How”, 2017). 1.
Allowing users to block, mute, or unfollow unwanted followers. 2. Filters on notifications that allows users to filter out any unwanted replies or mentions from the accounts that the user do not follow.3. Reporting the undesirable behavior to Twitter.4. Warning the user about sensitive content before showing it. It works only for photos and videos.
5. Tagging privacy for photos allows the user to decide who can and cannot tag him/her in photos. III.Challenges and Changes to be madeAlthough tools are provided by contemporary social networking sites and laws are in place to fight cyberbullying in some countries.
the majority of cyberbullying instances go unreported (Peterson, 2013). At the same time, there is no system in place for automatic detection of such behavior. Cyberbullying is one of the widely recognized problems which has a lasting impact on its victims. While healthy social behavior is the solution to this problem, social media platforms need to consider integrating tools and / or mechanisms that can help in the detection and prevention of such incidents. Therefore, to have a safer and more constructive social environment, it is necessary to design a smart network or an online patrol that will prohibit such behavior by monitoring and filtering the obscene, hateful, and improper content from social media posts.