arXiv subjects

Kyumin Lee

Publications and source records attributed to Kyumin Lee.

36 records · Page 2Linked to original sources

arXiv2021

Hierarchical Multi-head Attentive Network for Evidence-aware Fake News Detection

The widespread of fake news and misinformation in various domains ranging from politics, economics to public health has posed an urgent need to automatically fact-check information. A recent trend in fake news detection is to utilize evidence from external sources. However, existing evidence-aware fake news detection methods focused on either only word-level attention or evidence-level attention, which may result in suboptimal performance. In this paper, we propose a Hierarchical Multi-head Attentive Network to fact-check textual claims. Our model jointly combines multi-head word-level attention and multi-head document-level attention, which aid explanation in both word-level and evidence-level. Experiments on two real-word datasets show that our model outperforms seven state-of-the-art baselines. Improvements over baselines are from 6\% to 18\%. Our source code and datasets are released at \texttt{\url{https://github.com/nguyenvo09/EACL2021}}.

cs.AI↗

arXiv2020

HABERTOR: An Efficient and Effective Deep Hatespeech Detector

We present our HABERTOR model for detecting hatespeech in large scale user-generated content. Inspired by the recent success of the BERT model, we propose several modifications to BERT to enhance the performance on the downstream hatespeech classification task. HABERTOR inherits BERT's architecture, but is different in four aspects: (i) it generates its own vocabularies and is pre-trained from the scratch using the largest scale hatespeech dataset; (ii) it consists of Quaternion-based factorized components, resulting in a much smaller number of parameters, faster training and inferencing, as well as less memory usage; (iii) it uses our proposed multi-source ensemble heads with a pooling layer for separate input sources, to further enhance its effectiveness; and (iv) it uses a regularized adversarial training with our proposed fine-grained and adaptive noise magnitude to enhance its robustness. Through experiments on the large-scale real-world hatespeech dataset with 1.4M annotated comments, we show that HABERTOR works better than 15 state-of-the-art hatespeech detection methods, including fine-tuning Language Models. In particular, comparing with BERT, our HABERTOR is 4~5 times faster in the training/inferencing phase, uses less than 1/3 of the memory, and has better performance, even though we pre-train it by using less than 1% of the number of words. Our generalizability analysis shows that HABERTOR transfers well to other unseen hatespeech datasets and is a more efficient and effective alternative to BERT for the hatespeech classification.

cs.CL↗

arXiv2020

Hierarchical Evidence Set Modeling for Automated Fact Extraction and Verification

Automated fact extraction and verification is a challenging task that involves finding relevant evidence sentences from a reliable corpus to verify the truthfulness of a claim. Existing models either (i) concatenate all the evidence sentences, leading to the inclusion of redundant and noisy information; or (ii) process each claim-evidence sentence pair separately and aggregate all of them later, missing the early combination of related sentences for more accurate claim verification. Unlike the prior works, in this paper, we propose Hierarchical Evidence Set Modeling (HESM), a framework to extract evidence sets (each of which may contain multiple evidence sentences), and verify a claim to be supported, refuted or not enough info, by encoding and attending the claim and evidence sets at different levels of hierarchy. Our experimental results show that HESM outperforms 7 state-of-the-art methods for fact extraction and claim verification. Our source code is available at https://github.com/ShyamSubramanian/HESM.

cs.CL↗

arXiv2020

Where Are the Facts? Searching for Fact-checked Information to Alleviate the Spread of Fake News

Although many fact-checking systems have been developed in academia and industry, fake news is still proliferating on social media. These systems mostly focus on fact-checking but usually neglect online users who are the main drivers of the spread of misinformation. How can we use fact-checked information to improve users' consciousness of fake news to which they are exposed? How can we stop users from spreading fake news? To tackle these questions, we propose a novel framework to search for fact-checking articles, which address the content of an original tweet (that may contain misinformation) posted by online users. The search can directly warn fake news posters and online users (e.g. the posters' followers) about misinformation, discourage them from spreading fake news, and scale up verified content on social media. Our framework uses both text and images to search for fact-checking articles, and achieves promising results on real-world datasets. Our code and datasets are released at https://github.com/nguyenvo09/EMNLP2020.

cs.IR↗

arXiv2020

Quaternion-Based Self-Attentive Long Short-Term User Preference Encoding for Recommendation

Quaternion space has brought several benefits over the traditional Euclidean space: Quaternions (i) consist of a real and three imaginary components, encouraging richer representations; (ii) utilize Hamilton product which better encodes the inter-latent interactions across multiple Quaternion components; and (iii) result in a model with smaller degrees of freedom and less prone to overfitting. Unfortunately, most of the current recommender systems rely on real-valued representations in Euclidean space to model either user's long-term or short-term interests. In this paper, we fully utilize Quaternion space to model both user's long-term and short-term preferences. We first propose a QUaternion-based self-Attentive Long term user Encoding (QUALE) to study the user's long-term intents. Then, we propose a QUaternion-based self-Attentive Short term user Encoding (QUASE) to learn the user's short-term interests. To enhance our models' capability, we propose to fuse QUALE and QUASE into one model, namely QUALSE, by using a Quaternion-based gating mechanism. We further develop Quaternion-based Adversarial learning along with the Bayesian Personalized Ranking (QABPR) to improve our model's robustness. Extensive experiments on six real-world datasets show that our fused QUALSE model outperformed 11 state-of-the-art baselines, improving 8.43% at HIT@1 and 10.27% at NDCG@1 on average compared with the best baseline.

cs.IR↗

arXiv2020

Attributed Multi-Relational Attention Network for Fact-checking URL Recommendation

To combat fake news, researchers mostly focused on detecting fake news and journalists built and maintained fact-checking sites (e.g., Snopes.com and Politifact.com). However, fake news dissemination has been greatly promoted via social media sites, and these fact-checking sites have not been fully utilized. To overcome these problems and complement existing methods against fake news, in this paper we propose a deep-learning based fact-checking URL recommender system to mitigate impact of fake news in social media sites such as Twitter and Facebook. In particular, our proposed framework consists of a multi-relational attentive module and a heterogeneous graph attention network to learn complex/semantic relationship between user-URL pairs, user-user pairs, and URL-URL pairs. Extensive experiments on a real-world dataset show that our proposed framework outperforms eight state-of-the-art recommendation models, achieving at least 3~5.3% improvement.

cs.IR↗

arXiv2019

The Rise of Guardians: Fact-checking URL Recommendation to Combat Fake News

A large body of research work and efforts have been focused on detecting fake news and building online fact-check systems in order to debunk fake news as soon as possible. Despite the existence of these systems, fake news is still wildly shared by online users. It indicates that these systems may not be fully utilized. After detecting fake news, what is the next step to stop people from sharing it? How can we improve the utilization of these fact-check systems? To fill this gap, in this paper, we (i) collect and analyze online users called guardians, who correct misinformation and fake news in online discussions by referring fact-checking URLs; and (ii) propose a novel fact-checking URL recommendation model to encourage the guardians to engage more in fact-checking activities. We found that the guardians usually took less than one day to reply to claims in online conversations and took another day to spread verified information to hundreds of millions of followers. Our proposed recommendation model outperformed four state-of-the-art models by 11%~33%. Our source code and dataset are available at https://github.com/nguyenvo09/CombatingFakeNews.

cs.IR↗

arXiv2019

Learning from Fact-checkers: Analysis and Generation of Fact-checking Language

In fighting against fake news, many fact-checking systems comprised of human-based fact-checking sites (e.g., snopes.com and politifact.com) and automatic detection systems have been developed in recent years. However, online users still keep sharing fake news even when it has been debunked. It means that early fake news detection may be insufficient and we need another complementary approach to mitigate the spread of misinformation. In this paper, we introduce a novel application of text generation for combating fake news. In particular, we (1) leverage online users named \emph{fact-checkers}, who cite fact-checking sites as credible evidences to fact-check information in public discourse; (2) analyze linguistic characteristics of fact-checking tweets; and (3) propose and build a deep learning framework to generate responses with fact-checking intention to increase the fact-checkers' engagement in fact-checking activities. Our analysis reveals that the fact-checkers tend to refute misinformation and use formal language (e.g. few swear words and Internet slangs). Our framework successfully generates relevant responses, and outperforms competing models by achieving up to 30\% improvements. Our qualitative study also confirms that the superiority of our generated responses compared with responses generated from the existing models.

cs.CL↗

arXiv2019

Adversarial Mahalanobis Distance-based Attentive Song Recommender for Automatic Playlist Continuation

In this paper, we aim to solve the automatic playlist continuation (APC) problem by modeling complex interactions among users, playlists, and songs using only their interaction data. Prior methods mainly rely on dot product to account for similarities, which is not ideal as dot product is not metric learning, so it does not convey the important inequality property. Based on this observation, we propose three novel deep learning approaches that utilize Mahalanobis distance. Our first approach uses user-playlist-song interactions, and combines Mahalanobis distance scores between (i) a target user and a target song, and (ii) between a target playlist and the target song to account for both the user's preference and the playlist's theme. Our second approach measures song-song similarities by considering Mahalanobis distance scores between the target song and each member song (i.e., existing song) in the target playlist. The contribution of each distance score is measured by our proposed memory metric-based attention mechanism. In the third approach, we fuse the two previous models into a unified model to further enhance their performance. In addition, we adopt and customize Adversarial Personalized Ranking (APR) for our three approaches to further improve their robustness and predictive capabilities. Through extensive experiments, we show that our proposed models outperform eight state-of-the-art models in two large-scale real-world datasets.

cs.IR↗

arXiv2019

Signed Distance-based Deep Memory Recommender

Personalized recommendation algorithms learn a user's preference for an item by measuring a distance/similarity between them. However, some of the existing recommendation models (e.g., matrix factorization) assume a linear relationship between the user and item. This approach limits the capacity of recommender systems, since the interactions between users and items in real-world applications are much more complex than the linear relationship. To overcome this limitation, in this paper, we design and propose a deep learning framework called Signed Distance-based Deep Memory Recommender, which captures non-linear relationships between users and items explicitly and implicitly, and work well in both general recommendation task and shopping basket-based recommendation task. Through an extensive empirical study on six real-world datasets in the two recommendation tasks, our proposed approach achieved significant improvement over ten state-of-the-art recommendation models.

cs.IR↗

arXiv2018

Regularizing Matrix Factorization with User and Item Embeddings for Recommendation

Following recent successes in exploiting both latent factor and word embedding models in recommendation, we propose a novel Regularized Multi-Embedding (RME) based recommendation model that simultaneously encapsulates the following ideas via decomposition: (1) which items a user likes, (2) which two users co-like the same items, (3) which two items users often co-liked, and (4) which two items users often co-disliked. In experimental validation, the RME outperforms competing state-of-the-art models in both explicit and implicit feedback datasets, significantly improving Recall@5 by 5.9~7.0%, NDCG@20 by 4.3~5.6%, and MAP@10 by 7.9~8.9%. In addition, under the cold-start scenario for users with the lowest number of interactions, against the competing models, the RME outperforms NDCG@5 by 20.2% and 29.4% in MovieLens-10M and MovieLens-20M datasets, respectively. Our datasets and source code are available at: https://github.com/thanhdtran/RME.git.

cs.IR↗

arXiv2018

MRAttractor: Detecting Communities from Large-Scale Graphs

Detecting groups of users, who have similar opinions, interests, or social behavior, has become an important task for many applications. A recent study showed that dynamic distance based Attractor, a community detection algorithm, outperformed other community detection algorithms such as Spectral clustering, Louvain and Infomap, achieving higher Normalized Mutual Information (NMI) and Adjusted Rand Index (ARI). However, Attractor often takes long time to detect communities, requiring many iterations. To overcome the drawback and handle large-scale graphs, in this paper we propose MRAttractor, an advanced version of Attractor to be runnable on a MapReduce framework. In particular, we (i) apply a sliding window technique to reduce the running time, keeping the same community detection quality; (ii) design and implement the Attractor algorithm for a MapReduce framework; and (iii) evaluate MRAttractor's performance on synthetic and real-world datasets. Experimental results show that our algorithm significantly reduced running time and was able to handle large-scale graphs.

cs.DC↗

arXiv2017

Identifying On-time Reward Delivery Projects with Estimating Delivery Duration on Kickstarter

In Crowdfunding platforms, people turn their prototype ideas into real products by raising money from the crowd, or invest in someone else's projects. In reward-based crowdfunding platforms such as Kickstarter and Indiegogo, selecting accurate reward delivery duration becomes crucial for creators, backers, and platform providers to keep the trust between the creators and the backers, and the trust between the platform providers and users. According to Kickstarter, 35% backers did not receive rewards on time. Unfortunately, little is known about on-time and late reward delivery projects, and there is no prior work to estimate reward delivery duration. To fill the gap, in this paper, we (i) extract novel features that reveal latent difficulty levels of project rewards; (ii) build predictive models to identify whether a creator will deliver all rewards in a project on time or not; and (iii) build a regression model to estimate accurate reward delivery duration (i.e., how long it will take to produce and deliver all the rewards). Experimental results show that our models achieve good performance -- 82.5% accuracy, 78.1 RMSE, and 0.108 NRMSE at the first 5% of the longest reward delivery duration.

cs.CY↗

arXiv2017

Characteristics of On-time and Late Reward Delivery Projects

The crowdfunding market size has increased exponentially, reaching tens of billions of dollars and showing the popularity of crowdfunding. However, according to Kickstarter, 35% backers did not receive rewards on time. To maintain the trust between creators and backers, and sustain the crowdfunding business growth, it is crucial to understand how on-time and late reward delivery projects are different. In this paper, we analyze characteristics of on-time and late reward delivery projects, especially, focusing on project descriptions, creator profiles, and activeness and linguistic patterns of creators and backers. Our analysis reveals that the less complicated a project is and more actively a creator responds to backers, the higher an on-time reward delivery probability is. It shows there are significant differences between on-time and late reward delivery projects.

cs.CY↗

arXiv2016

How to Succeed in Crowdfunding: a Long-Term Study in Kickstarter

Crowdfunding platforms have become important sites where people can create projects to seek funds toward turning their ideas into products, and back someone else's projects. As news media have reported successfully funded projects (e.g., Pebble Time, Coolest Cooler), more people have joined crowdfunding platforms and launched projects. But in spite of rapid growth of the number of users and projects, a project success rate at large has been decreasing because of launching projects without enough preparation and experience. Little is known about what reactions project creators made (e.g., giving up or making the failed projects better) when projects failed, and what types of successful projects we can find. To solve these problems, in this manuscript we (i) collect the largest datasets from Kickstarter, consisting of all project profiles, corresponding user profiles, projects' temporal data and users' social media information; (ii) analyze characteristics of successful projects, behaviors of users and understand dynamics of the crowdfunding platform; (iii) propose novel statistical approaches to predict whether a project will be successful and a range of expected pledged money of the project; (iv) develop predictive models and evaluate performance of the models; (v) analyze what reactions project creators had when project failed, and if they did not give up, how they made the failed projects successful; and (vi) cluster successful projects by their evolutional patterns of pledged money toward understanding what efforts project creators should make in order to get more pledged money. Our experimental results show that the predictive models can effectively predict project success and a range of expected pledged money.

cs.SI↗

arXiv2016

Understanding Citizen Reactions and Ebola-Related Information Propagation on Social Media

In severe outbreaks such as Ebola, bird flu and SARS, people share news, and their thoughts and responses regarding the outbreaks on social media. Understanding how people perceive the severe outbreaks, what their responses are, and what factors affect these responses become important. In this paper, we conduct a comprehensive study of understanding and mining the spread of Ebola-related information on social media. In particular, we (i) conduct a large-scale data-driven analysis of geotagged social media messages to understand citizen reactions regarding Ebola; (ii) build information propagation models which measure locality of information; and (iii) analyze spatial, temporal and social properties of Ebola-related information. Our work provides new insights into Ebola outbreak by understanding citizen reactions and topic-based information propagation, as well as providing a foundation for analysis and response of future public health crises.

cs.SI↗

arXiv2014

Who Will Retweet This? Automatically Identifying and Engaging Strangers on Twitter to Spread Information

There has been much effort on studying how social media sites, such as Twitter, help propagate information in different situations, including spreading alerts and SOS messages in an emergency. However, existing work has not addressed how to actively identify and engage the right strangers at the right time on social media to help effectively propagate intended information within a desired time frame. To address this problem, we have developed two models: (i) a feature-based model that leverages peoples' exhibited social behavior, including the content of their tweets and social interactions, to characterize their willingness and readiness to propagate information on Twitter via the act of retweeting; and (ii) a wait-time model based on a user's previous retweeting wait times to predict her next retweeting time when asked. Based on these two models, we build a recommender system that predicts the likelihood of a stranger to retweet information when asked, within a specific time window, and recommends the top-N qualified strangers to engage with. Our experiments, including live studies in the real world, demonstrate the effectiveness of our work.

cs.SI↗

arXiv2014

The Dark Side of Micro-Task Marketplaces: Characterizing Fiverr and Automatically Detecting Crowdturfing

As human computation on crowdsourcing systems has become popular and powerful for performing tasks, malicious users have started misusing these systems by posting malicious tasks, propagating manipulated contents, and targeting popular web services such as online social networks and search engines. Recently, these malicious users moved to Fiverr, a fast-growing micro-task marketplace, where workers can post crowdturfing tasks (i.e., astroturfing campaigns run by crowd workers) and malicious customers can purchase those tasks for only $5. In this paper, we present a comprehensive analysis of Fiverr. First, we identify the most popular types of crowdturfing tasks found in this marketplace and conduct case studies for these crowdturfing tasks. Then, we build crowdturfing task detection classifiers to filter these tasks and prevent them from becoming active in the marketplace. Our experimental results show that the proposed classification approach effectively detects crowdturfing tasks, achieving 97.35% accuracy. Finally, we analyze the real world impact of crowdturfing tasks by purchasing active Fiverr tasks and quantifying their impact on a target site. As part of this analysis, we show that current security systems inadequately detect crowdsourced manipulation, which confirms the necessity of our proposed crowdturfing task detection approach.

cs.SI↗