Posted Content•
Stance Classification in Rumours as a Sequential Task Exploiting the Tree Structure of Social Media Conversations
TL;DR: This work is the first to model Twitter conversations as a tree structure in this manner, introducing a novel way of tackling NLP tasks on Twitter conversations.
Abstract: Rumour stance classification, the task that determines if each tweet in a collection discussing a rumour is supporting, denying, questioning or simply commenting on the rumour, has been attracting substantial interest. Here we introduce a novel approach that makes use of the sequence of transitions observed in tree-structured conversation threads in Twitter. The conversation threads are formed by harvesting users' replies to one another, which results in a nested tree-like structure. Previous work addressing the stance classification task has treated each tweet as a separate unit. Here we analyse tweets by virtue of their position in a sequence and test two sequential classifiers, Linear-Chain CRF and Tree CRF, each of which makes different assumptions about the conversational structure. We experiment with eight Twitter datasets, collected during breaking news, and show that exploiting the sequential structure of Twitter conversations achieves significant improvements over the non-sequential methods. Our work is the first to model Twitter conversations as a tree structure in this manner, introducing a novel way of tackling NLP tasks on Twitter conversations.
Citations
More filters
01 Jul 2018
TL;DR: This work proposes two recursive neural models based on a bottom-up and a top-down tree-structured neural networks for rumor representation learning and classification, which naturally conform to the propagation layout of tweets.
Abstract: Automatic rumor detection is technically very challenging. In this work, we try to learn discriminative features from tweets content by following their non-sequential propagation structure and generate more powerful representations for identifying different type of rumors. We propose two recursive neural models based on a bottom-up and a top-down tree-structured neural networks for rumor representation learning and classification, which naturally conform to the propagation layout of tweets. Results on two public Twitter datasets demonstrate that our recursive neural models 1) achieve much better performance than state-of-the-art approaches; 2) demonstrate superior capacity on detecting rumors at very early stage.
370 citations
Cites background from "Stance Classification in Rumours as..."
..., the root), is usually responsive to its immediate ancestor (Lukasik et al., 2016; Zubiaga et al., 2016a), suggesting obvious local characteristic of the interaction....
[...]
...…a supervised classification problem, which learns a classifier f from labeled claims, that is f : Ci → Yi, where Yi takes one of the four finer-grained classes: non-rumor, false rumor, true rumor, and unverified rumor that are introduced in the literature (Ma et al., 2017; Zubiaga et al., 2016b)....
[...]
...Meanwhile, a reply, rather than directly responding to the source tweet (i.e., the root), is usually responsive to its immediate ancestor (Lukasik et al., 2016; Zubiaga et al., 2016a), suggesting obvious local characteristic of the interaction....
[...]
...Analysis shows that people tend to stop spreading a rumor if it is known as false (Zubiaga et al., 2016b)....
[...]
13 Sep 2017
TL;DR: A novel approach using Conditional Random Fields that learns from the sequential dynamics of social media posts with the current state-of-the-art rumour detection system, as well as other baselines, and results provide evidence for the generalisability of the classifier.
Abstract: Tools that are able to detect unverified information posted on social media during a news event can help to avoid the spread of rumours that turn out to be false. In this paper we compare a novel approach using Conditional Random Fields that learns from the sequential dynamics of social media posts with the current state-of-the-art rumour detection system, as well as other baselines. In contrast to existing work, our classifier does not need to observe tweets querying the stance of a post to deem it a rumour but, instead, exploits context learned during the event. Our classifier has improved precision and recall over the state-of-the-art classifier that relies on querying tweets, as well as outperforming our best baseline. Moreover, the results provide evidence for the generalisability of our classifier.
172 citations
Cites background from "Stance Classification in Rumours as..."
...be as input to classifiers that determine stance of tweets towards rumours [16,38] or classifiers that determine the veracity of rumours [9]....
[...]
23 Apr 2018
TL;DR: Extensive experiments on real-world datasets gathered from Twitter and news portals demonstrate that the proposed framework improves both rumor detection and stance classification tasks consistently with the help of the strong inter-task connections, achieving much better performance than state-of-the-art methods.
Abstract: In recent years, an unhealthy phenomenon characterized as the massive spread of fake news or unverified information (i.e., rumors) has become increasingly a daunting issue in human society. The rumors commonly originate from social media outlets, primarily microblogging platforms, being viral afterwards by the wild, willful propagation via a large number of participants. It is observed that rumorous posts often trigger versatile, mostly controversial stances among participating users. Thus, determining the stances on the posts in question can be pertinent to the successful detection of rumors, and vice versa. Existing studies, however, mainly regard rumor detection and stance classification as separate tasks. In this paper, we argue that they should be treated as a joint, collaborative effort, considering the strong connections between the veracity of claim and the stances expressed in responsive posts. Enlightened by the multi-task learning scheme, we propose a joint framework that unifies the two highly pertinent tasks, i.e., rumor detection and stance classification. Based on deep neural networks, we train both tasks jointly using weight sharing to extract the common and task-invariant features while each task can still learn its task-specific features. Extensive experiments on real-world datasets gathered from Twitter and news portals demonstrate that our proposed framework improves both rumor detection and stance classification tasks consistently with the help of the strong inter-task connections, achieving much better performance than state-of-the-art methods.
170 citations
Cites background or methods from "Stance Classification in Rumours as..."
...performance beyond the majority class [48]....
[...]
...We followed the common practice of prior works [30, 48] that employed this dataset to convert the original labels into SDQC set based on a set of rules proposed in [30]....
[...]
...[48] exploited the conversational structure among microblog texts for classifying tweet stance....
[...]
...[48] built a tree-CRF classifier that learns the dynamics of stance in tree-structured conversations such as Twitter replies, instead of classifying tweets in isolation....
[...]
TL;DR: A survey of stance detection in social media posts and (online) regular texts is presented and it is hoped that this newly emerging topic will act as a significant resource for interested researchers and practitioners.
Abstract: Automatic elicitation of semantic information from natural language texts is an important research problem with many practical application areas. Especially after the recent proliferation of online content through channels such as social media sites, news portals, and forums; solutions to problems such as sentiment analysis, sarcasm/controversy/veracity/rumour/fake news detection, and argument mining gained increasing impact and significance, revealed with large volumes of related scientific publications. In this article, we tackle an important problem from the same family and present a survey of stance detection in social media posts and (online) regular texts. Although stance detection is defined in different ways in different application settings, the most common definition is “automatic classification of the stance of the producer of a piece of text, towards a target, into one of these three classes: {Favor, Against, Neither}.” Our survey includes definitions of related problems and concepts, classifications of the proposed approaches so far, descriptions of the relevant datasets and tools, and related outstanding issues. Stance detection is a recent natural language processing topic with diverse application areas, and our survey article on this newly emerging topic will act as a significant resource for interested researchers and practitioners.
131 citations
Cites methods from "Stance Classification in Rumours as..."
...…the approaches for stance detection, those for rumour stance detection are usually supervised machine learning approaches with different feature sets [Lukasik et al. 2019; Pamungkas et al. 2019; Zubiaga et al. 2018a, 2016, 2018b] in addition to semi-supervised approaches [Giasemidis et al. 2018]....
[...]
TL;DR: It is shown that sequential classifiers that exploit the use of discourse properties in social media conversations while using only local features, outperform non-sequential classifiers and that LSTM using a reduced set of features can outperform the other sequentialclassifiers.
Abstract: Rumour stance classification, defined as classifying the stance of specific social media posts into one of supporting, denying, querying or commenting on an earlier post, is becoming of increasing interest to researchers. While most previous work has focused on using individual tweets as classifier inputs, here we report on the performance of sequential classifiers that exploit the discourse features inherent in social media interactions or ‘conversational threads’. Testing the effectiveness of four sequential classifiers – Hawkes Processes, Linear-Chain Conditional Random Fields (Linear CRF), Tree-Structured Conditional Random Fields (Tree CRF) and Long Short Term Memory networks (LSTM) – on eight datasets associated with breaking news stories, and looking at different types of local and contextual features, our work sheds new light on the development of accurate stance classifiers. We show that sequential classifiers that exploit the use of discourse properties in social media conversations while using only local features, outperform non-sequential classifiers. Furthermore, we show that LSTM using a reduced set of features can outperform the other sequential classifiers; this performance is consistent across datasets and across types of stances. To conclude, our work also analyses the different features under study, identifying those that best help characterise and distinguish between stances, such as supporting tweets being more likely to be accompanied by evidence than denying tweets. We also set forth a number of directions for future research.
130 citations
Cites background or methods from "Stance Classification in Rumours as..."
...In work that is closer to our objectives, stance classification has also been used to help determine the veracity of information in micro-posts [16], often referred to as rumour stance classification [30, 18, 12, 19]....
[...]
...For example, in preliminary work we showed that a sequential classifier modelling the temporal sequence of tweets outperforms standard classifiers [18, 19]....
[...]
...Here we extend the experimentation presented in our previous work using Conditional Random Fields for rumour stance classification [19] in a number of directions: (1) we perform a comparison of a broader range of classifiers, including state-of-the-art rumour stance classifiers such as Hawkes Processes introduced by Lukasik et al....
[...]
References
More filters
Proceedings Article•
05 Dec 2013TL;DR: This paper presents a simple method for finding phrases in text, and shows that learning good vector representations for millions of phrases is possible and describes a simple alternative to the hierarchical softmax called negative sampling.
Abstract: The recently introduced continuous Skip-gram model is an efficient method for learning high-quality distributed vector representations that capture a large number of precise syntactic and semantic word relationships. In this paper we present several extensions that improve both the quality of the vectors and the training speed. By subsampling of the frequent words we obtain significant speedup and also learn more regular word representations. We also describe a simple alternative to the hierarchical softmax called negative sampling.
An inherent limitation of word representations is their indifference to word order and their inability to represent idiomatic phrases. For example, the meanings of "Canada" and "Air" cannot be easily combined to obtain "Air Canada". Motivated by this example, we present a simple method for finding phrases in text, and show that learning good vector representations for millions of phrases is possible.
24,012 citations
"Stance Classification in Rumours as..." refers methods in this paper
...The Word2Vec model for each of the eight folds is trained from the collection of tweets pertaining to the seven events in the training set, so that the event (and the vocabulary) in the test set is unknown....
[...]
...• Word Embeddings: a vector with 300 dimensions averaging vector representations of the words in the tweet using Word2Vec (Mikolov et al., 2013)....
[...]
...• Word Embeddings: a vector with 300 dimensions averaging vector representations of the words in the tweet using Word2Vec (Mikolov et al., 2013)....
[...]
Proceedings Article•
28 Jun 2001TL;DR: This work presents iterative parameter estimation algorithms for conditional random fields and compares the performance of the resulting models to HMMs and MEMMs on synthetic and natural-language data.
Abstract: We present conditional random fields , a framework for building probabilistic models to segment and label sequence data. Conditional random fields offer several advantages over hidden Markov models and stochastic grammars for such tasks, including the ability to relax strong independence assumptions made in those models. Conditional random fields also avoid a fundamental limitation of maximum entropy Markov models (MEMMs) and other discriminative Markov models based on directed graphical models, which can be biased towards states with few successor states. We present iterative parameter estimation algorithms for conditional random fields and compare the performance of the resulting models to HMMs and MEMMs on synthetic and natural-language data.
13,190 citations
"Stance Classification in Rumours as..." refers methods in this paper
...Hence, having a data sequence X as input, CRF outputs a sequence of labels Y (Lafferty et al., 2001), where the output of each element yi will not only depend on its features, but also on the probabilities of other labels surrounding it....
[...]
Posted Content•
TL;DR: In this paper, the Skip-gram model is used to learn high-quality distributed vector representations that capture a large number of precise syntactic and semantic word relationships and improve both the quality of the vectors and the training speed.
Abstract: The recently introduced continuous Skip-gram model is an efficient method for learning high-quality distributed vector representations that capture a large number of precise syntactic and semantic word relationships. In this paper we present several extensions that improve both the quality of the vectors and the training speed. By subsampling of the frequent words we obtain significant speedup and also learn more regular word representations. We also describe a simple alternative to the hierarchical softmax called negative sampling. An inherent limitation of word representations is their indifference to word order and their inability to represent idiomatic phrases. For example, the meanings of "Canada" and "Air" cannot be easily combined to obtain "Air Canada". Motivated by this example, we present a simple method for finding phrases in text, and show that learning good vector representations for millions of phrases is possible.
11,343 citations
26 Apr 2010
TL;DR: In this paper, the authors have crawled the entire Twittersphere and found a non-power-law follower distribution, a short effective diameter, and low reciprocity, which all mark a deviation from known characteristics of human social networks.
Abstract: Twitter, a microblogging service less than three years old, commands more than 41 million users as of July 2009 and is growing fast. Twitter users tweet about any topic within the 140-character limit and follow others to receive their tweets. The goal of this paper is to study the topological characteristics of Twitter and its power as a new medium of information sharing.We have crawled the entire Twitter site and obtained 41.7 million user profiles, 1.47 billion social relations, 4,262 trending topics, and 106 million tweets. In its follower-following topology analysis we have found a non-power-law follower distribution, a short effective diameter, and low reciprocity, which all mark a deviation from known characteristics of human social networks [28]. In order to identify influentials on Twitter, we have ranked users by the number of followers and by PageRank and found two rankings to be similar. Ranking by retweets differs from the previous two rankings, indicating a gap in influence inferred from the number of followers and that from the popularity of one's tweets. We have analyzed the tweets of top trending topics and reported on their temporal behavior and user participation. We have classified the trending topics based on the active period and the tweets and show that the majority (over 85%) of topics are headline news or persistent news in nature. A closer look at retweets reveals that any retweeted tweet is to reach an average of 1,000 users no matter what the number of followers is of the original tweet. Once retweeted, a tweet gets retweeted almost instantly on next hops, signifying fast diffusion of information after the 1st retweet.To the best of our knowledge this work is the first quantitative study on the entire Twittersphere and information diffusion on it.
6,108 citations