Results 1 -
6 of
6
Credibility in context: An analysis of feature distributions
- in twitter. ASE/IEEE International Conference on Social Computing, SocialCom
, 2012
"... Abstract—Twitter is a major forum for rapid dissemination of user-provided content in real time. As such, a large proportion of the information it contains is not particularly relevant to many users and in fact is perceived as unwanted ’noise ’ by many. There has been increased research interest in ..."
Abstract
-
Cited by 11 (4 self)
- Add to MetaCart
(Show Context)
Abstract—Twitter is a major forum for rapid dissemination of user-provided content in real time. As such, a large proportion of the information it contains is not particularly relevant to many users and in fact is perceived as unwanted ’noise ’ by many. There has been increased research interest in predicting whether tweets are relevant, newsworthy or credible, using a variety of models and methods. In this paper, we focus on an analysis that highlights the utility of the individual features in Twitter such as hashtags, retweets and mentions for predicting credibility. We first describe a context-based evaluation of the utility of a set of features for predicting manually provided credibility assessments on a corpus of microblog tweets. This is followed by an evaluation of the distribution/presence of each feature across 8 diverse crawls of tweet data. Last, an analysis of feature distribution across dyadic pairs of tweets and retweet chains of various lengths is described. Our results show that the best indicators of credibility include URLs, mentions, retweets and tweet length and that features occur more prominently in data describing emergency and unrest situations. I.
Towards Customizing Credibility in Different Contexts: Languages, Topics and Locations- A Twitter Case Study
"... Even though online social networks (ONS) have been increasingly used as a source of news and information, the credibility of that easily-available information may be questionable. In this paper, we review selected literature related to information credibility focusing more on micro-blogging credibil ..."
Abstract
- Add to MetaCart
(Show Context)
Even though online social networks (ONS) have been increasingly used as a source of news and information, the credibility of that easily-available information may be questionable. In this paper, we review selected literature related to information credibility focusing more on micro-blogging credibility as a part of on-going research. From this review it can be concluded that credibility is situational and contextual; it varies from one context to another. We propose to examine messages related to confirmed news topics and false rumors topics and then study the effect of three dimensions: language, topic content, and location on the detection features utilized by the existing work. Firstly, we will check if the frequency distribution of selected features remains similar or changes according these dimensions. We then examine different available classifying techniques and study the effect of each context on the credibility classifying results. Another goal is to report on user studies that explore how people from different environments perceive and judge information credibility on Twitter. We believe the analysis of our results will improve existing credibility measurement by helping us in selecting the most suitable features for information credibility classification in each context.
Leveraging social relevance: Using social networks to enhance literature access and microblog search
, 2013
"... ..."
Risk and Trust Credibility in Context: An Analysis of Feature Distributions in Twitter
"... Abstract—Twitter is a major forum for rapid dissemination of user-provided content in real time. As such, a large proportion of the information it contains is not particularly relevant to many users and in fact is perceived as unwanted ’noise ’ by many. There has been increased research interest in ..."
Abstract
- Add to MetaCart
(Show Context)
Abstract—Twitter is a major forum for rapid dissemination of user-provided content in real time. As such, a large proportion of the information it contains is not particularly relevant to many users and in fact is perceived as unwanted ’noise ’ by many. There has been increased research interest in predicting whether tweets are relevant, newsworthy or credible, using a variety of models and methods. In this paper, we focus on an analysis that highlights the utility of the individual features in Twitter such as hashtags, retweets and mentions for predicting credibility. We first describe a context-based evaluation of the utility of a set of features for predicting manually provided credibility assessments on a corpus of microblog tweets. This is followed by an evaluation of the distribution/presence of each feature across 8 diverse crawls of tweet data. Last, an analysis of feature distribution across dyadic pairs of tweets and retweet chains of various lengths is described. Our results show that the best indicators of credibility include URLs, mentions, retweets and tweet length and that features occur more prominently in data describing emergency and unrest situations. I.
“Picture the scene...”; Visually Summarising Social Media Events
"... Due to the advent of social media and web 2.0, we are faced with a deluge of information; recently, research efforts have focused on filtering out noisy, irrelevant information items from social media streams and in particular have attempted to automatically identify and summarise events. However, d ..."
Abstract
- Add to MetaCart
(Show Context)
Due to the advent of social media and web 2.0, we are faced with a deluge of information; recently, research efforts have focused on filtering out noisy, irrelevant information items from social media streams and in particular have attempted to automatically identify and summarise events. However, due to the heterogeneous na-ture of such social media streams, these efforts have not reached fruition. In this paper, we investigate how images can be used as a source for summarising events. Existing approaches have con-sidered only textual summaries which are often poorly written, in a different language and slow to digest. Alternatively, images are “worth 1,000 words ” and are able to quickly & easily convey an idea or scene. Since images in social media can also be noisy, irrelevant & repetitive, we propose new techniques for their au-tomatic selection, ranking and presentation. We evaluate our ap-proach on a recently created social media event data set containing 365k tweets and 50 events, for which we extend by collecting 625k related images. By conducting two crowdsourced evaluations, we firstly show how our approach overcomes the problems of automat-ically collecting relevant and diverse images from noisy microblog data, before highlighting the advantages of multimedia summarisa-tion over text based approaches.
TREC Microblog 2012 Track: Real-Time Algorithm for Microblog Ranking Systems
"... As a matter of fact Twitter is becoming the new big data container, due to the deep increase of amount of users and its growing popularity. Moreover the huge amount of user profiles and rough text data, are providing continuosly new research challenges. This paper reports our contribution and result ..."
Abstract
- Add to MetaCart
As a matter of fact Twitter is becoming the new big data container, due to the deep increase of amount of users and its growing popularity. Moreover the huge amount of user profiles and rough text data, are providing continuosly new research challenges. This paper reports our contribution and results to the Trec 2012 Microblog Track. In this particular, challenge each par-ticipant is required to conduct a ”real-time ” retrieval task, which given a query topic seeks for the most recent and relevant tweets. We devised an effective real time ranking algorithm, avoiding heavy computational requirements. Our contribution is multifold: (1) adapting an existing ranking method BM25 to the microblogging purpose (2) enhancing traditional content-based features with knowledge extracted from Wikipedia, (3) employing Pseudo Relevance Feedback techniques for query expansion (4) performing text analysis such as ad-hoc text normalization and POS Tagging to limit noise data and better represent useful information. 1.
