Nothing Special   »   [go: up one dir, main page]
Scribd Logo
Upload

Welcome to Scribd!

  • 35 views

    Retweeting of Tweets: Investigation of Correlating Variables & Its Implications

    Uploaded by

    Pooja Dhaka
    The study aims at to explore how retweet works to understand how information is disseminated in the Twitter network. The correlation between retweet and variables are investigated on the twitter platform in a sample size of 500 tweets which have more than 5000 retweets and implications are calculated using Pearson correlation method.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd

    Retweeting of Tweets: Investigation of Correlating Variables & Its Implications

    Uploaded by

    Pooja Dhaka
    35 views25 pages
    The study aims at to explore how retweet works to understand how information is disseminated in the Twitter network. The correlation between retweet and variables are investigated on the twitter platform in a sample size of 500 tweets which have more than 5000 retweets and implications are calculated using Pearson correlation method.

    Original Title

    Btp Presentation Final

    Copyright

    © © All Rights Reserved

    Available Formats

    PPTX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    The study aims at to explore how retweet works to understand how information is disseminated in the Twitter network. The correlation between retweet and variables are investigated on the twitter platform in a sample size of 500 tweets which have more than 5000 retweets and implications are calculated using Pearson correlation method.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    35 views25 pages

    Retweeting of Tweets: Investigation of Correlating Variables & Its Implications

    Uploaded by

    Pooja Dhaka
    The study aims at to explore how retweet works to understand how information is disseminated in the Twitter network. The correlation between retweet and variables are investigated on the twitter platform in a sample size of 500 tweets which have more than 5000 retweets and implications are calculated using Pearson correlation method.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    of 25

    Retweeting of tweets: investigation of

    correlating variables & its implications Pooja Dhaka | 120205033


    Gu id e :Prof . Ke y u r So r a thia
    Twitter

    In Twitter, every second, on average, around 6,000 tweets are tweeted, which corresponds to
    over 350,000 tweets sent per minute, 500 million tweets per day and around 200 billion tweets
    per year
    Twitter

    Tweets that are twitter messages are restricted to 140 characters and which are usually much
    focused. A particular tweet can contains URL, hashtag, mentions, replies, number of favorites,
    and number of retweets.

    Popular categories which are identified on twitter are Business, Film & TV, Music, Online
    Media, Hosting, News, Blogging, Commerce, Politics, Comedian, Sports, Author, Journalist,
    Entrepreneur, Government, Model, Magazine, Science, Health, Medicine, Film Maker,
    Gaming, Comics, Crime, people and community in one’s locality
    Retweets

    A Retweet is a re-posting of someone else's Tweet.


    Retweet is a particular case of mentioning the content and has become the key mechanism for
    information diffusion in Twitter. Retweeting plays an important role in visibility and validity of
    information across users.
    Most retweeted

    Obama’s victory tweet was the most retweeted tweet ever with more than 800K retweets in
    2012.

    Ellen brakes the record (by far) during the Oscars ceremony, with a world-renowned
    tweet. The now World’s Most Retweeted Tweet surpassed the millionth mark within hours,
    now residing well over 3M. It’s certainly a kind of selfie we will not be forgetting anytime soon
    in 2014.
    Application of retweet
    The power of twitter network is utilized by Catalan government in November 2014 to detect
    high polarization of twitter network and the political leaning of users in political discussions.
    During the 2008 election in Germany including the attention gained by the political parties in
    political opinions by the users and analyzing whether tweets can predict the elections.
    Users use twitter to update traffic conditions such as accidents, etc. and spread information via
    retweeting.
    Companies post tweets on twitter to publicize advertising campaigns to promote their
    businesses and they can make very effective real-time decisions.
    Research gap

    It is important to explore how retweet works to understand how information is disseminated in


    the Twitter network. Critical examination of literature suggests that the study of exploration on
    retweets is not expanded and there are potential areas which are yet to be explored.

    The research study aims to investigate the correlation between different variables with the
    retweet and implications are concluded that influences the retweetability of tweets on Twitter
    platform.
    Variables
    Followers are people who receive a user’s tweet whom they follow. The tweets of the users
    will be visible on their home timeline.

    Favorite is when a particular tweet is marked as a favorite and original tweeter will be informed
    that someone liked their post.

    Friends are when the follower is also being followed back by the users.

    Status are the posts posted on twitter timeline by the users and visible to their followers.
    Variables

    Favorites are the likes done by a particular profile user on different posts posted by other
    users.

    Listed is when people are mentioned in others user’s tweets.

    Media is used in different type formats such as photos, short videos, animated gifs and
    sometimes URLs are also included in tweets.
    Timeline
    Methodology

    Collected the tweets which are retweeted more than 5000 times .
    An intensive characterization of a Twitter dataset of which includes all users (500) whose
    tweets have been retweeted more than 5000 times and the messages are posted up to march
    2016.
    Identified factors which influence the retweetability of a particular tweet: the retweets count,
    the favorites count of a particular count, the usage of media, the users background and their
    profile is also studied in reference to that particular tweet and their relationships with the
    retweet count. The users’ profiles are segregated and the factors are plotted in relation to
    retweet count.
    Data
    TWEET USER • Friends count 1 RETWEET
    • Created at • Name • Listed count • Created at

    • Retweet count • Screen name • Favourites count • Text

    • Favorites count • Location • Statues count • User

    • Media • url • Account created at • User’s profile


    • Photo • Description • Timezone
    • Video
    • urls • Followers count • Profile backgorund
    Data
    TWEET : ISIS is making big threats today - no respect for U.S.A. or our \"leader\" - If I win it will be a very different story, with
    very fast results,

    Created at Tue Feb 09 21:40:19 +0000 2016,

    Retweet count 6930, Favorites count 17151

    USER

    • Name Donald J. Trump, Screen name realDonaldTrump

    • Location New York, NY

    • Followers count 6111190 Friends count 49

    • Listed count 29437, Favourites count 61

    • Statues count 30718, Account created at Wed Mar 18 13:46:38 +0000 2009

    • Description #MakeAmericaGreatAgain #Trump2016 #Facebook: http:\\/\\/www.facebook.com\\/DonaldTrump #Instagram:


    https:\\/\\/www.instagram.com\\/realdonaldtrump\\/
    .
    Data analysis
    The users were characterized into normal users, pages, celebrities and personalities.
    Normal users are those users whose followers count is less than 10,000.
    Celebrities are those users who are popular worldwide entertainers. Also celebrities were
    divided into singers (Celebrity1), actors (Celebrity2), youtubers (Celebrity3) and rappers
    (Celebrity4).
    Personalities comprises of businessman, president, social reformers etc. and are popular
    worldwide.
    The pages were further divided into quotable pages and non-quotable pages (specific topic).
    Second set of results consists the subsequent regression analysis via Pearson’s correlation
    between the variables and the retweets count which predict the retweeting intentions of the
    users. More specifically, the relationships between the retweet count with the user’s followers
    count, favorite count, friends count, statuses count, favorites count and media.
    Results
    Retweet v/s follower count

    Celebrity4 (Rappers) Normal users


    Retweet v/s follower count

    Pages Normal users


    Retweet v/s follower count

    Celebrity1 Celebrity2
    Retweet v/s follower count

    Celebrity3 Celebrity4
    Retweet v/s follower count

    Personalities
    Retweet v/s status count

    Celebrity1 Celebrity4
    Conclusion

    Retweets has a great impact in terms of visibility and validity and analyzing retweets in terms
    of the above parameters through which people design their tweets in a way which gains
    maximum retweets and which further disseminates information in a wide groups and gains
    maximum visibility of information. To understand why certain tweets, get more retweet, we
    investigated some variables that have potential relationship with retweeting of tweets. In the
    research, some variables were identified which affect the retweetability. We drew the
    relationship between retweets and variables which affect retweetability through Pearson
    correlation.

    Generally, it was found that amongst all variables favorite count and followers count affect
    retweetability as there was a significant relationship built of these variables with retweeting.
    Some variables also showed negative relationship with retweeting which might be due to the
    limitation in the sample size of tweets. Overall, there was relationships were drawn and the
    variables were identified which affects retweeting.
    Thank you

    You might also like