With the increase in volume of daily online news items, it is more and more difficult for readers to identify news articles relevant to their interests. Thus, effective recommendation systems are critical for an effective user news consumption experience. Existing news recommendation methods usually rely on the news click history to model user interest. However, there are other signals about user behaviors, such as user commenting activity, which have not been used before. We propose a recommendation algorithm that predicts articles a user may be interested in, given her historical sequential commenting behavior on news articles. We show that following this sequential user behavior the news recommendation problem falls into in the class of session-based recommendation. The techniques in this class seek to model users' sequential and temporal behaviors. While we seek to follow the general directions in this space, we face unique challenges specific to news in modeling temporal dynamics, e.g., users' interests shift over time, users comment irregularly on articles, and articles are perishable items with limited lifespans. We propose a recency-regularized neural attentive framework for session-based news recommendation. The proposed method is able to capture the temporal dynamics of both users and news articles, while maintaining interpretability. We design a lag-aware attention and a recency regularization to model the time effect of news articles and comments. We conduct extensive empirical studies on 3 real-world news datasets to demonstrate the effectiveness of our method.

Imbalanced data classification has always been one of the most challenging problems in data science especially in the cybersecurity field, where we observe an out-of-balance proportion between benign and phishing examples in security datasets. Even though there are many phishing detection methods in literature, most of them neglect the imbalanced nature of phishing email datasets. In this paper, we examine the imbalanced property by varying legitimate to phishing class ratios. We generate new synthetic instances using a generative adversarial network model for long sentences (LeakGAN) to balance out the training process and ameliorate its impact on classification. These synthetic instances are labeled by positive-unlabeled learning and added to the initial imbalanced training set. The resulting dataset is given to the Bidirectional Encoder Representations from Transformers (BERT) model for sequence classification. We compare several state-of-the-art methods from the literature against our approach, which achieves a high performance throughout all the imbalanced ratios reaching an F1-score of 99.6% for the most extreme imbalanced ratio and an F1-score of 99.8% for balanced cases.

Social media is the ultimate challenge for many natural language processing tools. The constant emergence of linguistic constructs challenge even the most sophisticated NLP tools. Predicting word embeddings for out of vocabulary words is one of those challenges. Word embedding models only include terms that occur a sufficient number of times in their training corpora. Word embedding vector models are unable to directly provide any useful information about a word not in their vocabularies. We propose a fast method for predicting vectors for out of vocabulary terms that makes use of the surrounding terms of the unknown term and the hidden context layer of the word2vec model. We propose this method as a strong baseline in the sense that 1) while it does not surpass all state-of-the-art methods, it surpasses several techniques for vector prediction on benchmark tasks, 2) even when it underperforms, the margin is very small retaining competitive performance in downstream tasks, and 3) it is inexpensive to compute, requiring no additional training stage. We also show that our technique can be incorporated into existing methods to achieve a new state-of-the-art on the word vector prediction problem.

We extend evidence-aware claim verification to the context of positive-unlabeled (PU) learning. Existing works assume the truth and the falsity of the claims are known for training and form the task as a supervised learning problem. However, this assumption underestimates the difficulty of collecting false claims; we argue that claim verification is more challenging in the absence of negative labels. We consider a more practical setting, where only a comparatively small number of true claims are labeled and more claims remain unlabeled. Thus, we formulate the claim verification task as a PU learning problem. We decouple learning representation of claim-evidence pair from PU learning and adopt a pre-trained universal language model to encode claim-evidence pairs. We further propose to use the generative adversarial network (GAN) to capture the latent alignment between encoded claim-evidence pair and the truthfulness. We leverage the verification as part of the GAN by extending previous GAN based PU learning. We show that the proposed model achieves the best performance with a small amount of labeled data and is robust to the truthfulness prior estimation. We conduct a thorough analysis of the model selection. The proposed approach performs the best under two practical scenarios: (i) the unlabeled data is more than the labeled data; (ii) and the unlabeled positive data is more than the unlabeled negative data.

Predicting users' opinions in their response to social events has important real-world applications, many of which with political and social impacts.Existing approaches derive a population's opinion on a going event from large scores of user generated content.In certain scenarios, we may not be able to acquire such content and thus cannot infer opinion on those emerging events.To address this problem, we propose to explore opinion on unseen articles based on an user's fingerprinting: the prior reading and commenting history.This work presents a focused study on modeling and leveraging fingerprinting techniques to predict a user's future opinion to an unseen event or topic.We introduce a recurrent neural network based model that integrates fingerprinting.We collect a large dataset that consists of event-comment pairs from six news websites.We evaluate the proposed model on this dataset.The results show substantial performance gains demonstrating the effectiveness of our approach.

Electronic commerce has a dominant role in consumer economics. and popular garnering a lot of research attention. Understanding consumer market dynamics based on product popularity is crucial for business intelligence. This work explores the temporal dynamics in online marketing. We introduce a new popularity index based on Amazon: Product Popularity based on Sales Review Volume (PPSRV). We explore and evaluate sequential deep learning models to obtain time series embedding that can predict the product popularity. We further characterize popularity competition between similar products and extend our model of popularity prediction in a competitive environment. Experimental results on large-scale reviews demonstrate the effectiveness of our approach.

Satirical news detection is important in order to prevent the spread of misinformation over the Internet. Existing approaches to capture news satire use machine learning models such as SVM and hierarchical neural networks along with hand-engineered features, but do not explore sentence and document difference. This paper proposes a robust, hierarchical deep neural network approach for satire detection, which is capable of capturing satire both at the sentence level and at the document level. The architecture incorporates pluggable generic neural networks like CNN, GRU, and LSTM. Experimental results on real world news satire dataset show substantial performance gains demonstrating the effectiveness of our proposed approach. An inspection of the learned models reveals the existence of key sentences that control the presence of satire in news.

Many data-intensive applications collect (structured) data from a variety of sources. A key task in this process is record linkage, which is the problem of determining the records from these sources that refer to the same real-world entities. Traditional approaches use the record representation of entities to accomplish this task. With the nascence of social media, entities on the Web are now accompanied by user generated content. We present a method for record linkage that uses this hitherto untapped source of entity information. We use document-based distances, with an emphasis on word embedding document distances, to determine if two entities match. Our rationale is that user evaluations of entities converge in semantic content, and hence in the word embedded space, as the number of user evaluations grows. We analyze the effectiveness of the proposed method both as a stand-alone method and in combination with record-based record linkage methods. Experimental results using real-world reviews demonstrate the high effectiveness of our approach. To our knowledge, this is the first work exploring the use of user generated content accompanying entities in the record linkage task.

Abstract In this paper, we study the problem of aspect-based sentiment analysis. Our model simultaneously extracts aspect-specific opinion expressions and determines the rating for each aspect in reviews. Previous works have mainly focused on the problem of opinion phrase extraction and aspect rating prediction in a pipelined manner and are not able to capture the dependencies of aspect opinion expression on aspect rating and vice-versa. They are also unable to discover aspect-specific opinion expressions and their associated rating scores. We present a joint modelling approach to extract aspect-specific sentiment expression and aspect rating prediction simultaneously. This paper proposes a novel LDA–CRF hybrid model which employs discriminative conditional random field component for phrase extraction, a regression component for rating prediction and a generative component for grouping aspect–sentiment expressions (aspect-specific opinion expressions) into coherent topics. To show the effectiveness of our approach, we evaluate the performance of the model on both task: (i) aspect-specific opinion expressions and (ii) rating prediction on the dataset of hotel and restaurant reviews from TripAdvisor.com. Experimental results show that both task potentially reinforce each other and joint modeling outperformed state-of-the-art baselines for each individual tasks.

We focus on email-based attacks, a rich field with well-publicized consequences. We show how current Natural Language Generation (NLG) technology allows an attacker to generate masquerade attacks on scale, and study their effectiveness with a within-subjects study. We also gather insights on what parts of an email do users focus on and how users identify attacks in this realm, by planting signals and also by asking them for their reasoning. We find that: (i) 17% of participants could not identify any of the signals that were inserted in emails, and (ii) Participants were unable to perform better than random guessing on these attacks. The insights gathered and the tools and techniques employed could help defenders in: (i) implementing new, customized anti-phishing solutions for Internet users including training next-generation email filters that go beyond vanilla spam filters and capable of addressing masquerade, (ii) more effectively training and upgrading the skills of email users, and (iii) understanding the dynamics of this novel attack and its ability of tricking humans.

Satirical news is considered to be entertainment, but it is potentially deceptive and harmful.Despite the embedded genre in the article, not everyone can recognize the satirical cues and therefore believe the news as true news.We observe that satirical cues are often reflected in certain paragraphs rather than the whole document.Existing works only consider documentlevel features to detect the satire, which could be limited.We consider paragraphlevel linguistic features to unveil the satire by incorporating neural network and attention mechanism.We investigate the difference between paragraph-level features and document-level features, and analyze them on a large satirical news dataset.The evaluation shows that the proposed model detects satirical news effectively and reveals what features are important at which level.

Online reviews play a crucial role in helping consumers evaluate and compare products and services. This critical importance of reviews also incentivizes fraudsters (or spammers) to write fake or spam reviews to secretly promote or demote some target products and services. Existing approaches to detecting spam reviews and reviewers employed review contents, reviewer behaviors, star rating patterns, and reviewer-product networks for detection. In this research, we further discovered that reviewers' posting rates (number of reviews written in a period of time) also follow an interesting distribution pattern, which has not been reported before. That is, their posting rates are bimodal. Multiple spammers also tend to collectively and actively post reviews to the same set of products within a short time frame, which we call co-bursting. Furthermore, we found some other interesting patterns in individual reviewers' temporal dynamics and their co-bursting behaviors with other reviewers. Inspired by these findings, we first propose a two-mode Labeled Hidden Markov Model to model spamming using only individual reviewers' review posting times. We then extend it to the Coupled Hidden Markov Model to capture both reviewer posting behaviors and co-bursting signals. Our experiments show that the proposed model significantly outperforms state-of-the-art baselines in identifying individual spammers. Furthermore, we propose a co-bursting network based on co-bursting relations, which helps detect groups of spammers more effectively than existing approaches.

Sentiment classification becomes more and more important with the rapid growth of user generated content. However, sentiment classification task usually comes with two challenges: first, sentiment classification is highly domain-dependent and training sentiment classifier for every domain is inefficient and often impractical; second, since the quantity of labeled data is important for assessing the quality of classifier, it is hard to evaluate classifiers when labeled data is limited for certain domains. To address the challenges mentioned above, we focus on learning high-level features that are able to generalize across domains, so a global classifier can benefit with a simple combination of documents from multiple domains. In this paper, the proposed model incorporates both sentiment polarity and unlabeled data from multiple domains and learns new feature representations. Our model doesn’t require labels from every domain, which means the learned feature representation can be generalized for sentiment domain adaptation. In addition, the learned feature representation can be used as classifier since our model defines the meaning of feature value and arranges high-level features in a prefixed order, so it is not necessary to train another classifier on top of the new features. Empirical evaluations demonstrate our model outperforms baselines and yields competitive results to other state-of-the-art works on benchmark datasets.

Recently, the problem of opinion spam has been widespread and has attracted a lot of research attention. While the problem has been approached on a variety of dimensions, the temporal dynamics in which opinion spamming operates is unclear. Are there specific spamming policies that spammers employ? What kind of changes happen with respect to the dynamics to the truthful ratings on entities. How do buffered spamming operate for entities that need spamming to retain threshold popularity and reduced spamming for entities making better success? We analyze these questions in the light of time-series analysis on Yelp. Our analyses discover various temporal patterns and their relationships with the rate at which fake reviews are posted. Building on our analyses, we employ vector autoregression to predict the rate of deception across different spamming policies. Next, we explore the effect of filtered reviews on (long-term and imminent) future rating and popularity prediction of entities. Our results discover novel temporal dynamics of spamming which are intuitive, arguable and also render confidence on Yelp's filtering. Lastly, we leverage our discovered temporal patterns in deception detection. Experimental results on large-scale reviews show the effectiveness of our approach that significantly improves the existing approaches.

Identifying aspect-based opinions has been studied extensively in recent years. However, existing work primarily focused on adjective, adverb, and noun expressions. Clearly, verb expressions can imply opinions too. We found that in many domains verb expressions can be even more important to applications because they often describe major issues of products or services. These issues enable brands and businesses to directly improve their products or services. To the best of our knowledge, this problem has not received much attention in the literature. In this paper, we make an attempt to solve this problem. Our proposed method first extracts verb expressions from reviews and then employs Markov Networks to model rich linguistic features and long distance relationships to identify negative issue expressions. Since our training data is obtained from titles of reviews whose labels are automatically inferred from review ratings, our approach is applicable to any domain without manual involvement. Experimental results using real-life review datasets show that our approach outperforms strong baselines.

In this paper, we propose a system, FLORIN, which provides support for near real-time applications on user generated content on daily news. FLORIN continuously crawls news outlets for articles and user comments accompanying them. It attaches the articles and comments to daily event stories. It identifies the opinionated content in user comments and performs named entity recognition on news articles. All these pieces of information are organized hierarchically and exportable to other applications. Multiple applications can be built on this data. We have implemented a sentiment analysis system that runs on top of it.

Aspect extraction is an important task in sentiment analysis.Topic modeling is a popular method for the task.However, unsupervised topic models often generate incoherent aspects.To address the issue, several knowledge-based models have been proposed to incorporate prior knowledge provided by the user to guide modeling.In this paper, we take a major step forward and show that in the big data era, without any user input, it is possible to learn prior knowledge automatically from a large amount of review data available on the Web.Such knowledge can then be used by a topic model to discover more coherent aspects.There are two key challenges: (1) learning quality knowledge from reviews of diverse domains, and (2) making the model fault-tolerant to handle possibly wrong knowledge.A novel approach is proposed to solve these problems.Experimental results using reviews from 36 domains show that the proposed approach achieves significant improvements over state-of-the-art baselines.

In this paper we first exploit cash-tags ("$" followed by stocks' ticker symbols) in Twitter to build a stock network, where nodes are stocks connected by edges when two stocks co-occur frequently in tweets.We then employ a labeled topic model to jointly model both the tweets and the network structure to assign each node and each edge a topic respectively.This Semantic Stock Network (SSN) summarizes discussion topics about stocks and stock relations.We further show that social sentiment about stock (node) topics and stock relationship (edge) topics are predictive of each stock's market.For prediction, we propose to regress the topic-sentiment time-series and the stock's price time series.Experimental results demonstrate that topic sentiments from close neighbors are able to help improve the prediction of a stock markedly.

As social media is becoming an increasingly important source of public information, companies, organizations and individuals are actively using social media platforms to promote their products, services, ideas and ideologies. Unlike promotional campaigns on TV or other traditional mass media platforms, campaigns on social media often appear in stealth modes. Campaign promoters often try to influence people's behaviors/opinions/decisions in a latent manner such that the readers are not aware that the messages they see are strategic campaign posts aimed at persuading them to buy target products/services. Readers take such campaign posts as just organic posts from the general public. It is thus important to discover such campaigns, their promoter accounts and how the campaigns are organized and executed as it can uncover the dynamics of Internet marketing. This discovery is clearly useful for competitors and also the general public. However, so far little work has been done to solve this problem. In this paper, we study this important problem in the context of the Twitter platform. Given a set of tweets streamed from Twitter based on a set of keywords representing a particular topic, the proposed technique aims to identify user accounts that are involved in promotion. We formulate the problem as a relational classification problem and solve it using typed Markov Random Fields (T-MRF), which is proposed as a generalization of the classic Markov Random Fields. Our experiments are carried out using three real-life datasets from the health science domain related to smoking. Such campaigns are interesting to health scientists, government health agencies and related businesses for obvious reasons. Our results show that the proposed method is highly effective.

Topic models have been widely used to discover latent topics in text documents. However, they may produce topics that are not interpretable for an application. Researchers have proposed to incorporate prior domain knowledge into topic models to help produce coherent topics. The knowledge used in existing models is typically domain dependent and assumed to be correct. However, one key weakness of this knowledge-based approach is that it requires the user to know the domain very well and to be able to provide knowledge suitable for the domain, which is not always the case because in most real-life applications, the user wants to find what they do not know. In this paper, we propose a framework to leverage the general knowledge in topic models. Such knowledge is domain independent. Specifically, we use one form of general knowledge, i.e., lexical semantic relations of words such as synonyms, antonyms and adjective attributes, to help produce more coherent topics. However, there is a major obstacle, i.e., a word can have multiple meanings/senses and each meaning often has a different set of synonyms and antonyms. Not every meaning is suitable or correct for a domain. Wrong knowledge can result in poor quality topics. To deal with wrong knowledge, we propose a new model, called GK-LDA, which is able to effectively exploit the knowledge of lexical relations in dictionaries. To the best of our knowledge, GK-LDA is the first such model that can incorporate the domain independent knowledge. Our experiments using online product reviews show that GK-LDA performs significantly better than existing state-of-the-art models.

Aspect extraction is one of the key tasks in sentiment analysis.In recent years, statistical models have been used for the task.However, such models without any domain knowledge often produce aspects that are not interpretable in applications.To tackle the issue, some knowledge-based topic models have been proposed, which allow the user to input some prior domain knowledge to generate coherent aspects.However, existing knowledge-based topic models have several major shortcomings, e.g., little work has been done to incorporate the cannot-link type of knowledge or to automatically adjust the number of topics based on domain knowledge.This paper proposes a more advanced topic model, called MC-LDA (LDA with m-set and c-set), to address these problems, which is based on an Extended generalized Pólya urn (E-GPU) model (which is also proposed in this paper).Experiments on real-life product reviews from a variety of domains show that MC-LDA outperforms the existing state-of-the-art models markedly.

Opinionated social media such as product reviews are now widely used by individuals and organizations for their decision making. However, due to the reason of profit or fame, people try to game the system by opinion spamming (e.g., writing fake reviews) to promote or to demote some target products. In recent years, fake review detection has attracted significant attention from both the business and research communities. However, due to the difficulty of human labeling needed for supervised learning and evaluation, the problem remains to be highly challenging. This work proposes a novel angle to the problem by modeling spamicity as latent. An unsupervised model, called Author Spamicity Model (ASM), is proposed. It works in the Bayesian setting, which facilitates modeling spamicity of authors as latent and allows us to exploit various observed behavioral footprints of reviewers. The intuition is that opinion spammers have different behavioral distributions than non-spammers. This creates a distributional divergence between the latent population distributions of two clusters: spammers and non-spammers. Model inference results in learning the population distributions of the two clusters. Several extensions of ASM are also considered leveraging from different priors. Experiments on a real-life Amazon review dataset demonstrate the effectiveness of the proposed models which significantly outperform the state-of-the-art competitors.

Social media has become a major source of information for many applications. Numerous techniques have been proposed to analyze network structures and text contents. In this paper, we focus on fine-grained mining of contentions in discussion/debate forums. Contentions are perhaps the most important feature of forums that discuss social, political and religious issues. Our goal is to discover contention and agreement indicator expressions, and contention points or topics both at the discussion collection level and also at each individual post level. To the best of our knowledge, limited work has been done on such detailed analysis. This paper proposes three models to solve the problem, which not only model both contention/agreement expressions and discussion topics, but also, more importantly, model the intrinsic nature of discussions/debates, i.e., interactions among discussants or debaters and topic sharing among posts through quoting and replying relations. Evaluation results using real-life discussion/debate posts from several domains demonstrate the effectiveness of the proposed models.

Opinionated social media such as product reviews are now widely used by individuals and organizations for their decision making. However, due to the reason of profit or fame, people try to game the system by opinion spamming (e.g., writing fake reviews) to promote or demote some target products. For reviews to reflect genuine user experiences and opinions, such spam reviews should be detected. Prior works on opinion spam focused on detecting fake reviews and individual fake reviewers. However, a fake reviewer group (a group of reviewers who work collaboratively to write fake reviews) is even more damaging as they can take total control of the sentiment on the target product due to its size. This paper studies spam detection in the collaborative setting, i.e., to discover fake reviewer groups. The proposed method first uses a frequent itemset mining method to find a set of candidate groups. It then uses several behavioral models derived from the collusion phenomenon among fake reviewers and relation models based on the relationships among groups, individual reviewers, and products they reviewed to detect fake reviewer groups. Additionally, we also built a labeled dataset of fake reviewer groups. Although labeling individual fake reviews and reviewers is very hard, to our surprise labeling fake reviewer groups is much easier. We also note that the proposed technique departs from the traditional supervised learning approach for spam detection because of the inherent nature of our problem which makes the classic supervised learning approach less effective. Experimental results show that the proposed method outperforms multiple strong baselines including the state-of-the-art supervised classification, regression, and learning to rank algorithms.