Twitter is working on phenomenal changes to its tweets. This makes it a safe place on social media. Tweeting has become a habit of present day social citizens. It reflects ones ideas, lifestyle and activities. It helps to voice out their perceptions. Recently twitter has been used to promote anti-social threats and terrorism. This has shaken twitter a lot. Twitter has been making policy changes since then. These changes ensure its tweets are safer. It is been building mechanisms to reduce the abuse. It also reduces the use of anti-social phrases. It is easier to abuse someone by a tweet. It becomes prominent due to its wider reach. The tweets are not really screened for its legitimacy. It may harm one’s reputation.
Building up user and community profiles
Twitter has been building up user or social profiles. These profiles check on the language. They also check the style of use in its tweets. It is also researching on building community based profiles. This checks for misspellings, slangs, colloquial terms, etc. This ensures the local reach is safer on twitter. Anti-social elements usually use slangs. They use local terms to preach their propaganda. By building community profiles, twitter can detect the abusiveness in tweets. Once detected it removes those tweets.
Incident data sets
In addition, twitter is focussing on incident datasets. When an incident happens,it gets recorded. And when a similar incident happens again, it can scan the tweets of those users. The scan predicts the abusive languages used. This also involves profiling those users. The data set covers varied synonyms. These are used for antisocial phrases. Groups use alternative terms to escape the screening of tweets. So twitter is focussed on extensive data sets. This helps to avoid such abusive incidents.
Predictive analytical behaviour
Thisdeals with the study ofanalytical behaviour. The study is done on tweets posted by its users. Researchers from Stanford develop algorithms to evaluate user behaviour. They use sample tweets for those algorithms. Twitter is also investigating thealgorithms that predict the user behaviour. The user’s tweets are sampled. And a pattern is generated. This pattern is subjected to a predictive analysis. Thisdetermines if the user can abuse or not. The languages and the style of tweeting are vital for the analysis.
Features to limit abuse
Twitter has been adding features to its service. These features limit such abusive languages. It takes into account of the context of the profile. It also considers its age. It determines the abusive content in the tweets. It draws similarity of tweets. This determines the potential for abuse. These features block the visibility of such tweets. Followers are also tracked. Their tweets are screened for abusiveness.
Integrate transliteration services
Some groups use transliteration services for abuse. They use the abusive language in a local slang. Then they transliterate it. This escapes the patterns of detection. But it is still prone to abuse on who is being tweeted upon. Twitter should also integrate the transliteration services in its tweeting protection techniques.
Source: Techcrunch
