arXiv subjects

Roy Ka-Wei Lee

Publications and source records attributed to Roy Ka-Wei Lee.

105 records · Page 6Linked to original sources

arXiv2021

Syntax Matters! Syntax-Controlled in Text Style Transfer

Existing text style transfer (TST) methods rely on style classifiers to disentangle the text's content and style attributes for text style transfer. While the style classifier plays a critical role in existing TST methods, there is no known investigation on its effect on the TST methods. In this paper, we conduct an empirical study on the limitations of the style classifiers used in existing TST methods. We demonstrate that the existing style classifiers cannot learn sentence syntax effectively and ultimately worsen existing TST models' performance. To address this issue, we propose a novel Syntax-Aware Controllable Generation (SACG) model, which includes a syntax-aware style classifier that ensures learned style latent representations effectively capture the syntax information for TST. Through extensive experiments on two popular TST tasks, we show that our proposed method significantly outperforms the state-of-the-art methods. Our case studies have also demonstrated SACG's ability to generate fluent target-style sentences that preserved the original content.

cs.CL↗

arXiv2021

Disentangling Hate in Online Memes

Hateful and offensive content detection has been extensively explored in a single modality such as text. However, such toxic information could also be communicated via multimodal content such as online memes. Therefore, detecting multimodal hateful content has recently garnered much attention in academic and industry research communities. This paper aims to contribute to this emerging research topic by proposing DisMultiHate, which is a novel framework that performed the classification of multimodal hateful content. Specifically, DisMultiHate is designed to disentangle target entities in multimodal memes to improve hateful content classification and explainability. We conduct extensive experiments on two publicly available hateful and offensive memes datasets. Our experiment results show that DisMultiHate is able to outperform state-of-the-art unimodal and multimodal baselines in the hateful meme classification task. Empirical case studies were also conducted to demonstrate DisMultiHate's ability to disentangle target entities in memes and ultimately showcase DisMultiHate's explainability of the multimodal hateful content classification task.

cs.IR↗

arXiv2021

DeepStyle: User Style Embedding for Authorship Attribution of Short Texts

Authorship attribution (AA), which is the task of finding the owner of a given text, is an important and widely studied research topic with many applications. Recent works have shown that deep learning methods could achieve significant accuracy improvement for the AA task. Nevertheless, most of these proposed methods represent user posts using a single type of feature (e.g., word bi-grams) and adopt a text classification approach to address the task. Furthermore, these methods offer very limited explainability of the AA results. In this paper, we address these limitations by proposing DeepStyle, a novel embedding-based framework that learns the representations of users' salient writing styles. We conduct extensive experiments on two real-world datasets from Twitter and Weibo. Our experiment results show that DeepStyle outperforms the state-of-the-art baselines on the AA task.

cs.CL↗

arXiv2021

DeepHate: Hate Speech Detection via Multi-Faceted Text Representations

Online hate speech is an important issue that breaks the cohesiveness of online social communities and even raises public safety concerns in our societies. Motivated by this rising issue, researchers have developed many traditional machine learning and deep learning methods to detect hate speech in online social platforms automatically. However, most of these methods have only considered single type textual feature, e.g., term frequency, or using word embeddings. Such approaches neglect the other rich textual information that could be utilized to improve hate speech detection. In this paper, we propose DeepHate, a novel deep learning model that combines multi-faceted text representations such as word embeddings, sentiments, and topical information, to detect hate speech in online social platforms. We conduct extensive experiments and evaluate DeepHate on three large publicly available real-world datasets. Our experiment results show that DeepHate outperforms the state-of-the-art baselines on the hate speech detection task. We also perform case studies to provide insights into the salient features that best aid in detecting hate speech in online social platforms.

cs.CL↗

arXiv2021

AngryBERT: Joint Learning Target and Emotion for Hate Speech Detection

Automated hate speech detection in social media is a challenging task that has recently gained significant traction in the data mining and Natural Language Processing community. However, most of the existing methods adopt a supervised approach that depended heavily on the annotated hate speech datasets, which are imbalanced and often lack training samples for hateful content. This paper addresses the research gaps by proposing a novel multitask learning-based model, AngryBERT, which jointly learns hate speech detection with sentiment classification and target identification as secondary relevant tasks. We conduct extensive experiments to augment three commonly-used hate speech detection datasets. Our experiment results show that AngryBERT outperforms state-of-the-art single-task-learning and multitask learning baselines. We conduct ablation studies and case studies to empirically examine the strengths and characteristics of our AngryBERT model and show that the secondary tasks are able to improve hate speech detection.

cs.CL↗

arXiv2020

On Analyzing Antisocial Behaviors Amid COVID-19 Pandemic

The COVID-19 pandemic has developed to be more than a bio-crisis as global news has reported a sharp rise in xenophobia and discrimination in both online and offline communities. Such toxic behaviors take a heavy toll on society, especially during these daunting times. Despite the gravity of the issue, very few studies have studied online antisocial behaviors amid the COVID-19 pandemic. In this paper, we fill the research gap by collecting and annotating a large dataset of over 40 million COVID-19 related tweets. Specially, we propose an annotation framework to annotate the antisocial behavior tweets automatically. We also conduct an empirical analysis of our annotated dataset and found that new abusive lexicons are introduced amid the COVID-19 pandemic. Our study also identified the vulnerable targets of antisocial behaviors and the factors that influence the spreading of online antisocial content.

cs.SI↗

arXiv2020

On Analyzing Annotation Consistency in Online Abusive Behavior Datasets

Online abusive behavior is an important issue that breaks the cohesiveness of online social communities and even raises public safety concerns in our societies. Motivated by this rising issue, researchers have proposed, collected, and annotated online abusive content datasets. These datasets play a critical role in facilitating the research on online hate speech and abusive behaviors. However, the annotation of such datasets is a difficult task; it is often contentious on what should be the true label of a given text as the semantic difference of the labels may be blurred (e.g., abusive and hate) and often subjective. In this study, we proposed an analytical framework to study the annotation consistency in online hate and abusive content datasets. We applied our proposed framework to evaluate the consistency of the annotation in three popular datasets that are widely used in online hate speech and abusive behavior studies. We found that there is still a substantial amount of annotation inconsistency in the existing datasets, particularly when the labels are semantically similar.

cs.SI↗

arXiv2020

Keen2Act: Activity Recommendation in Online Social Collaborative Platforms

Social collaborative platforms such as GitHub and Stack Overflow have been increasingly used to improve work productivity via collaborative efforts. To improve user experiences in these platforms, it is desirable to have a recommender system that can suggest not only items (e.g., a GitHub repository) to a user, but also activities to be performed on the suggested items (e.g., forking a repository). To this end, we propose a new approach dubbed Keen2Act, which decomposes the recommendation problem into two stages: the Keen and Act steps. The Keen step identifies, for a given user, a (sub)set of items in which he/she is likely to be interested. The Act step then recommends to the user which activities to perform on the identified set of items. This decomposition provides a practical approach to tackling complex activity recommendation tasks while producing higher recommendation quality. We evaluate our proposed approach using two real-world datasets and obtain promising results whereby Keen2Act outperforms several baseline models.

cs.IR↗

arXiv2019

Linky: Visualizing User Identity Linkage Results For Multiple Online Social Networks

User identity linkage across online social networks is an emerging research topic that has attracted attention in recent years. Many user identity linkage methods have been proposed so far and most of them utilize user profile, content and network information to determine if two social media accounts belong to the same person. In most cases, user identity linkage methods are evaluated by performing some prediction tasks with the results presented using some overall accuracy measures. However, the methods are rarely compared at the individual user level where a predicted matched (or linked) pair of user identities from different online social networks can be visually compared in terms of user profile (e.g. username), content and network information. Such a comparison is critical to determine the relative strengths and weaknesses of each method. In this work, we present Linky, a visual analytical tool which extracts the results from different user identity linkage methods performed on multiple online social networks and visualizes the user profiles, content and ego networks of the linked user identities. Linky is designed to help researchers to (a) inspect the linked user identities at the individual user level, (b) compare results returned by different user linkage methods, and (c) provide a preliminary empirical understanding on which aspects of the user identities, e.g. profile, content or network, contributed to the user identity linkage results.

cs.SI↗

arXiv2019

Unsupervised User Identity Linkage via Factoid Embedding

User identity linkage (UIL), the problem of matching user account across multiple online social networks (OSNs), is widely studied and important to many real-world applications. Most existing UIL solutions adopt a supervised or semi-supervised approach which generally suffer from scarcity of labeled data. In this paper, we propose Factoid Embedding, a novel framework that adopts an unsupervised approach. It is designed to cope with different profile attributes, content types and network links of different OSNs. The key idea is that each piece of information about a user identity describes the real identity owner, and thus distinguishes the owner from other users. We represent such a piece of information by a factoid and model it as a triplet consisting of user identity, predicate, and an object or another user identity. By embedding these factoids, we learn the user identity latent representations and link two user identities from different OSNs if they are close to each other in the user embedding space. Our Factoid Embedding algorithm is designed such that as we learn the embedding space, each embedded factoid is "translated" into a motion in the user embedding space to bring similar user identities closer, and different user identities further apart. Extensive experiments are conducted to evaluate Factoid Embedding on two real-world OSNs data sets. The experiment results show that Factoid Embedding outperforms the state-of-the-art methods even without training data.

cs.SI↗

arXiv2018

Wisdom in Sum of Parts: Multi-Platform Activity Prediction in Social Collaborative Sites

In this paper, we proposed a novel framework which uses user interests inferred from activities (a.k.a., activity interests) in multiple social collaborative platforms to predict users' platform activities. Included in the framework are two prediction approaches: (i) direct platform activity prediction, which predicts a user's activities in a platform using his or her activity interests from the same platform (e.g., predict if a user answers a given Stack Overflow question using the user's interests inferred from his or her prior answer and favorite activities in Stack Overflow), and (ii) cross-platform activity prediction, which predicts a user's activities in a platform using his or her activity interests from another platform (e.g., predict if a user answers a given Stack Overflow question using the user's interests inferred from his or her fork and watch activities in GitHub). To evaluate our proposed method, we conduct prediction experiments on two widely used social collaborative platforms in the software development community: GitHub and Stack Overflow. Our experiments show that combining both direct and cross-platform activity prediction approaches yield the best accuracies for predicting user activities in GitHub (AUC=0.75) and Stack Overflow (AUC=0.89).

cs.SI↗

arXiv2018

Discovering Hidden Topical Hubs and Authorities in Online Social Networks

Finding influential users in online social networks is an important problem with many possible useful applications. HITS and other link analysis methods, in particular, have been often used to identify hub and authority users in web graphs and online social networks. These works, however, have not considered topical aspect of links in their analysis. A straightforward approach to overcome this limitation is to first apply topic models to learn the user topics before applying the HITS algorithm. In this paper, we instead propose a novel topic model known as Hub and Authority Topic (HAT) model to combine the two process so as to jointly learn the hub, authority and topical interests. We evaluate HAT against several existing state-of-the-art methods in two aspects: (i) modeling of topics, and (ii) link recommendation. We conduct experiments on two real-world datasets from Twitter and Instagram. Our experiment results show that HAT is comparable to state-of-the-art topic models in learning topics and it outperforms the state-of-the-art in link recommendation task.

cs.SI↗

arXiv2017

GitHub and Stack Overflow: Analyzing Developer Interests Across Multiple Social Collaborative Platforms

Increasingly, software developers are using a wide array of social collaborative platforms for software development and learning. In this work, we examined the similarities in developer's interests within and across GitHub and Stack Overflow. Our study finds that developers share common interests in GitHub and Stack Overflow, on average, 39% of the GitHub repositories and Stack Overflow questions that a developer had participated fall in the common interests. Also, developers do share similar interests with other developers who co-participated activities in the two platforms. In particular, developers who co-commit and co-pull-request same GitHub repositories and co-answer same Stack Overflow questions, share more common interests compare to other developers who co-participate in other platform activities.

cs.SI↗

arXiv2017

Friendship Maintenance and Prediction in Multiple Social Networks

Due to the proliferation of online social networks (OSNs), users find themselves participating in multiple OSNs. These users leave their activity traces as they maintain friendships and interact with other users in these OSNs. In this work, we analyze how users maintain friendship in multiple OSNs by studying users who have accounts in both Twitter and Instagram. Specifically, we study the similarity of a user's friendship and the evenness of friendship distribution in multiple OSNs. Our study shows that most users in Twitter and Instagram prefer to maintain different friendships in the two OSNs, keeping only a small clique of common friends in across the OSNs. Based upon our empirical study, we conduct link prediction experiments to predict missing friendship links in multiple OSNs using the neighborhood features, neighborhood friendship maintenance features and cross-link features. Our link prediction experiments shows that un- supervised methods can yield good accuracy in predicting links in one OSN using another OSN data and the link prediction accuracy can be further improved using supervised method with friendship maintenance and others measures as features.

cs.SI↗

arXiv2015

SignEPC : A Digital Signature Scheme for Efficient and Scalable Access Control in EPCglobal Network

The EPCglobal network is a computer network which allows supply chain companies to search for their unknown partners globally and share information stored in product RFID tags with each other. Although there have been quite a number of recent research works done to improve the security of EPCglobal Network, the existing access control solutions are not efficient and scalable. For instance, when a user queries Electronic Product Code Information Service (EPCIS) for EPC event information, the EPCIS would have to query the Electronic Product Code Discovery Service (EPCDS) to check the access rights of the user. This implementation is not efficient and creates a bottleneck at EPCDS. In this paper, we design and propose a digital signature scheme, SignEPC, as a more efficient and scalable access control solution for EPCglobal network. Our paper will also evaluate SignEPC by considering the various possible attacks that could be done on our proposed model

cs.CR↗