Twitter Tweets subjected to Credibility Analysis

By Harshad Patankar

Most of us use the popular short social media app Twitter which provides broadcasting 140 characters to your followers. The social media app changed its structure and excluded images and links from the 140 character limit. Twitter’s use by Business Intelligence has increased and these platforms are creating huge data to be analysed. Keeping this in mind Twitter has also created Twitter Engage.

Use of twitter platform and need of credibility

Increase in Twitter’s popularity, has led to its use by various celebrities and government bodies to reach customers in real time. The government bodies like Bombay Stock Exchange (BSE), Bengaluru Police and Government departments have created ‘Twitter Seva’ to enable bill payments and lodging of various complaints over Twitter.

As more and more services are coming up on Twitter there is a need to increase the credibility of Twitter handles based on the specificity of the tweet.

[su_pullquote]An Indian researcher, Tanushree Mitra put together a system to analyse whether a tweet is credible or not.[/su_pullquote]

A team led by Georgia Tech PhD candidate Tanushree Mitra scanned 66 million tweets linked to nearly 1,400 real-world events to build a language model that identified words and phrases that lead to strong or weak perceived levels of credibility on Twitter. An Indian-origin researcher put together a system to analyse whether a tweets is credible or not.

How to make your tweet credible

Twitter earlier opened up its verified service to public in order to prove one’s real identity to the followers of one’s handle, but there is need of something to value one’s tweet.

Many studies are being conducted on social media credibility in recent years but very little is known about what types of words or phrases create credibility perceptions during rapidly unfolding events,” Mitra  stated.

The team looked at tweets surrounding events in 2014 and 2015, including the emergence of Ebola in West Africa, the Charlie Hebdo attack in Paris and the death of Eric Garner in New York City. The team asked people to judge the posts on their credibility — from “certainly accurate” to “certainly inaccurate” — and then fed the words into a model that split them into 15 different linguistic categories.

Researchers in the field of Social Media: Eric Gilbert, an assistant professor and Tanushree Mitra, a PhD candidate in the Georgia Tech’s School of Interactive Computing.

The Georgia Tech computer then examined the words to judge if the tweets were credible or not. It matched the people’s opinions about 68 per cent of the time. That was significantly higher than the random baseline of 25 per cent.
The Tweets with booster words, such as ‘undeniable,’ and positive emotion terms, such as ‘eager’ and ‘terrific,’ were viewed as highly credible. Mitra also said that the words indicating positive sentiment but mocking the impracticality of the event, such as ‘ha,’ ‘grins’ or ‘joking,’ were seen as less credible and so were hedge words, including ‘certain level’ and ‘suspects.

The system analyses also indicated that the tweets with longer messages were more credible as they provided more information or reasoning. On the other hand, a higher number of re-tweets, scoring lower on credibility, represented an attempt to elicit collective reasoning during times of crisis or uncertainty.

We can expect the intelligence in a compact app that displays the perceived trustworthiness of an event as it unfolds on social media.

So get ready to check the credibility of your Tweet !

Harshad Patankar is the Chief Blogger at BumbleBeeHub Blog.
This article was originally published on the BumbleBeeHub
Featured Image Source: Pixabay
[su_note note_color=”#d2eaf6″]Fresh insights delivered to your phone each morning. Download our Android App today![/su_note]