Misinformation in Online Social Networks

A Theory-based Deep-Learning Approach to Detecting Disinformation in Financial Social Media

The spreading of disinformation in social media threatens cybersecurity and undermines market efficiency. Detecting disinformation is challenging due to large volumes of social media content and a rapidly changing environment. This research developed and validated a theory-based, novel deep-learning approach (called TRNN) to disinformation detection. Grounded in social and psychological theories, TRNN uses deep-learning and data-centric augmentation to enhance disinformation detection in financial social media. Temporal and contextual information is encoded as specific knowledge about human-validated disinformation, which was identified from our unique collection of 745,139 financial social media messages about four U.S. high-tech company stocks and their fine-grained trading data. TRNN uses multiple series of long short-term memory (LSTM) recurrent neurons to learn dynamic and hidden patterns to support disinformation detection. Our experimental findings show that TRNN significantly outperformed widely-used machine learning techniques in terms of precision, recall, F-score and accuracy, achieving consistently better classification performance in disinformation detection. A case study of Apple Inc.'s stock price movement demonstrates the potential usability of TRNN for secure knowledge management. The research contributes to developing novel approach and model, producing new information systems artifacts and dataset, and providing empirical findings of detecting online disinformation.

Multiple Benefit Thresholds Problem in Online Social Networks: An Algorithmic Approach

An important problem in the context of viral marketing in social networks is the Influence Threshold (IT) problem, which aims at finding some users (referred to as a seed set) to begin the process of disseminating their product’s information so that the benefit gained exceeds a predetermined threshold. Even though, marketing strategies exhibit different in several realistic scenarios due to market dependence or budget constraints. As a consequence, picking a seed set for a specific threshold is not enough to come up with an effective solution. To address the disadvantages of previous works with a new approach, we study the Multiple Benefit Thresholds (MBT), a generalized version of the IT problem, as a result of this phenomenon. Given a social network that is subjected to information distribution and a set of thresholds, T={T1,T2,…,Tk},Ti>0, the issue aims to seek the seed sets S1,S2,…,Sk with the lowest possible cost so that the benefit achieved from the influence process is at the very least T1,T2,…,Tk, respectively. The main challenges of this problem are a #NP-hard problem and the estimation of the objective function #P-Hard under traditional information propagation models. In addition, adapting the exist algorithms many times to different thresholds can lead to large computational costs. To address the abovementioned challenges, we introduced Efficient Sampling for Selecting Multiple Seed Sets, an efficient technique with theoretical guarantees (ESSM). At the core of our algorithm, we developed a novel algorithmic framework that (1) can use the solution to a smaller threshold to find that of larger ones and (2) can leverage existing samples with the current solution to find that of larger ones. The extensive experiments on several real social networks were conducted in order to show the effectiveness and performance of our algorithm compared with current ones. The results indicated that our algorithm outperformed other state-of-the-art ones in terms of both the total cost and running time.

Deep learning for misinformation detection on online social networks: a survey and new perspectives

Recently, the use of social networks such as Facebook, Twitter, and Sina Weibo has become an inseparable part of our daily lives. It is considered as a convenient platform for users to share personal messages, pictures, and videos. However, while people enjoy social networks, many deceptive activities such as fake news or rumors can mislead users into believing misinformation. Besides, spreading the massive amount of misinformation in social networks has become a global risk. Therefore, misinformation detection (MID) in social networks has gained a great deal of attention and is considered an emerging area of research interest. We find that several studies related to MID have been studied to new research problems and techniques. While important, however, the automated detection of misinformation is difficult to accomplish as it requires the advanced model to understand how related or unrelated the reported information is when compared to real information. The existing studies have mainly focused on three broad categories of misinformation: false information, fake news, and rumor detection. Therefore, related to the previous issues, we present a comprehensive survey of automated misinformation detection on (i) false information, (ii) rumors, (iii) spam, (iv) fake news, and (v) disinformation. We provide a state-of-the-art review on MID where deep learning (DL) is used to automatically process data and create patterns to make decisions not only to extract global features but also to achieve better results. We further show that DL is an effective and scalable technique for the state-of-the-art MID. Finally, we suggest several open issues that currently limit real-world implementation and point to future directions along this dimension.

FNED

The fast spreading of fake news stories on social media can cause inestimable social harm. Developing effective methods to detect them early is of paramount importance. A major challenge of fake news early detection is fully utilizing the limited data observed at the early stage of news propagation and then learning useful patterns from it for identifying fake news. In this article, we propose a novel deep neural network to detect fake news early. It has three novel components: (1) a status-sensitive crowd response feature extractor that extracts both text features and user features from combinations of users’ text response and their corresponding user profiles, (2) a position-aware attention mechanism that highlights important user responses at specific ranking positions, and (3) a multi-region mean-pooling mechanism to perform feature aggregation based on multiple window sizes. Experimental results on two real-world datasets demonstrate that our proposed model can detect fake news with greater than 90% accuracy within 5 minutes after it starts to spread and before it is retweeted 50 times, which is significantly faster than state-of-the-art baselines. Most importantly, our approach requires only 10% labeled fake news samples to achieve this effectiveness under PU-Learning settings.

Understanding Assimilation-contrast Effects in Online Rating Systems

“Unbiasedness,” which is an important property to ensure that users’ ratings indeed reflect their true evaluations of products, is vital both in shaping consumer purchase decisions and providing reliable recommendations in online rating systems. Recent experimental studies showed that distortions from historical ratings would ruin the unbiasedness of subsequent ratings. How to “discover” historical distortions in each single rating (or at the micro-level), and perform the “debiasing operations” are our main objective. Using 42M real customer ratings, we first show that users either “assimilate” or “contrast” to historical ratings under different scenarios, which can be further explained by a well-known psychological argument: the “Assimilate-Contrast” theory. This motivates us to propose the Historical Influence Aware Latent Factor Model (HIALF), the “first” model for real rating systems to capture and mitigate historical distortions in each single rating. HIALF allows us to study the influence patterns of historical ratings from a modelling perspective, which perfectly matches the assimilation and contrast effects observed in experiments. Moreover, HIALF achieves significant improvements in predicting subsequent ratings and characterizing relationships in ratings. It also contributes to better recommendations, wiser consumer purchase decisions, and deeper understanding of historical distortions in both honest rating and misbehaving rating settings.

Social Influence Spectrum at Scale

Given a social network, the Influence Maximization (InfMax) problem seeks a seed set of k people that maximizes the expected influence for a viral marketing campaign. However, a solution for a particular seed size k is often not enough to make an informed choice regarding budget and cost-effectiveness.

In this article, we propose the computation of Influence Spectrum (InfSpec), the maximum influence at each possible seed set size k within a given range [ k lower , k upper ], thus providing optimal decision making for any availability of budget or influence requirements. As none of the existing methods for InfMax are efficient enough for the task in large networks, we propose LISA (sub-Linear Influence Spectrum Approximation), an efficient approximation algorithm for InfSpec (and also InfMax) with the best-known worst-case guarantees for billion-scale networks. LISA returns an (1-1/e -ϵ)-approximate influence spectrum with high probability (1-δ), where ϵ, δ are precision parameters provided by users. Using statistical decision theory, LISA has an asymptotic optimal running time (in addition to optimal approximation guarantee). In practice, LISA surpasses the state-of-the-art InfMax methods, taking less than 15 minutes to process a network of 41.7 million nodes and 1.5 billions edges.