Stand for Something or Fall for Everything: Predict Misinformation Spread with Stance-Aware Graph Neural Networks

2023 | International Conference on Information Systems | Citations: 0

Authors: Chen, Zihan; Sun, Jingyi; Liu, Rong Emily; Mai, Feng

Abstract: Although pervasive spread of misinformation on social media platforms has becom ... Expand

Abstract: Although pervasive spread of misinformation on social media platforms has become a pressing challenge, existing platform interventions have shown limited success in curbing its dissemination. In this study, we propose a stance-aware graph neural network (stance-aware GNN) that leverages users' stances to proactively predict misinformation spread. As different user stances can form unique echo chambers, we customize four information passing paths in stance-aware GNN, while the trainable attention weights provide explainability by highlighting each structure's importance. Evaluated on a realworld dataset, stance-aware GNN outperforms benchmarks by 32.65% and exceeds advanced GNNs without user stance by over 4.69%. Furthermore, the attention weights indicate that users' opposition stances have a higher impact on their neighbors' behaviors than supportive ones, which function as social correction to halt misinformation propagation. Overall, our study provides an effective predictive model for platforms to combat misinformation, and highlights the impact of user stances in the misinformation propagation. Collapse

Topics: misinformation Twitter filter bubble social media social influence

Methods: machine learning RoBERTa experiment deep learning graph neural network

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

2023 | Information Systems Frontiers | Citations: 0

Authors: Chung, Wingyan; Zhang, Yinqiang; Pan, Jia

Abstract: The spreading of disinformation in social media threatens cybersecurity and unde ... Expand

Abstract: 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. Collapse

Topics: disinformation social media Twitter fake news usability

Methods: machine learning experiment design artifact deep learning recurrent neural network

Theories: psychological theory

Integration of Computer Vision with Analogical Reasoning for Characterizing Unknowns

2023 | HICSS | Citations: 0

Authors: Combs, K.; Bihl, T.; Ganapathy, S.

Abstract: Current state-of-the-art artificial intelligence struggles with accurate interp ... Expand

Abstract: Current state-of-the-art artificial intelligence struggles with accurate interpretation of out-of-library (OOL) objects. One method proposed remedy is analogical reasoning (AR), which utilizes abductive reasoning to draw inferences on an unfamiliar scenario given knowledge about a similar familiar scenario. Currently, applications of visual AR gravitate toward analogy-formatted image problems rather than to computer vision data sets. The Image Recognition Through Analogical Reasoning Algorithm (IRTARA) approach described herein shows how AR can be leveraged to improve computer vision in OOL situations. IRTARA produces a word-based term frequency list that characterizes the OOL object of interest. To evaluate the quality of the results of IRTARA, both quantitative and qualitative assessments are used, including a baseline to compare the automated methods with human-generated results. Fifteen OOL objects were tested using IRTARA, which showed consistent results across all three evaluation methods on the objects that performed exceptionally well or poorly overall. Collapse

Topics: floppy disk Wikipedia Python ergonomics usability

Methods: computational algorithm experiment word embedding GloVe survey

Causal Matching for Startups: Methods for Controlling Confounding

2023 | Americas Conference on Information Systems | Citations: 0

Authors: Fernandes Marinho Ferreira, Vitor; Kuruzovich, Jason

Abstract: Causal inference is a critical methodology that allows researchers to empirical ... Expand

Abstract: Causal inference is a critical methodology that allows researchers to empirically test hypotheses and build theories. Ideally, the design method for establishing this causality would be through randomized controlled trials. However, studying the impact of different practices (hereinafter treatments) on initial outcomes when treatments are not applied exogenously is a challenge. Matching methods are an appropriate way to provide these study groups. The objective of this work is to provide an algorithmbased solution to match startups in a way that can mimic a human match. To do this, we use the human matching effort by Yu (2020) as a ground truth to evaluate various natural language processes and compare the performance of each with human raters. By comparing automated approaches with matching, we provide guidance for researchers interested in the causal analysis of startups using textual data and other covariates. Collapse

Topics: entrepreneurship business model strategic management

Methods: causality analysis natural language processing computational algorithm term frequency–inverse document frequency GloVe

Fine-grained classification of drug trafficking based on Instagram hashtags

2023 | Decision Support Systems | Citations: 0

Authors: Hu, Chuanbo; Liu, Bin; Ye, Yanfang; Li, Xin

Abstract: Social networks have become important platforms for the marketing and sale of il ... Expand

Abstract: Social networks have become important platforms for the marketing and sale of illicit drugs. Hashtags make it easier for users to engage in drug trafficking, further increasing the risk of drug abuse. However, there are significant challenges in the detection and management of drug trafficking activities. In addition, the rapid legalization of some drugs has required a fine-grained classification of drugs to distinguish them from those that are illegal. Motivated by these observations, in this paper, our aim is to develop a methodology using the latest advances in AI technology to classify hashtags from posts advertising illicit drugs for sale on social networks. We present a semi-supervised deep learning approach to classify hashtags from posts advertising illicit drugs. An elegant combination of Bidirectional Encoder Representations from Transformers (BERT) with Graph Convolu­ tional Network (GCN) allows us to analyze the characteristics (e.g., shipping region and platform self-regulation) of illegal drug trafficking. Our BERT+GCN model achieved the best performance with more than 75% accuracy compared to the other three baseline models. Then, fine-grained hashtags identified are applied to explore the characteristics of drug trafficking. Finally, we report our results for further exploration of shipping regions and self-regulation of drug trafficking on the platform in two analysis scenarios. Our developed approach has shown its effectiveness in detecting hashtags for different types of drugs from illegal drug sellers. Based on hashtag classification, we also provide two case studies that indicate that (1) there are differences in self-regulation for different types of drugs on social media, (2) there are regional differences in the demand for different types of drugs. Collapse

Topics: Instagram social media Google+ artificial intelligence advertising management

Methods: BERT case study machine learning deep learning semi-supervised learning

Introduction to Accountability, Evaluation and Obscurity of AI Algorithms Minitrack (CC BY-NC-ND 4.0)

2023 | HICSS | Citations: 0

Authors: Juric, R.; Steele, R.

Abstract: This Minitrack has attracted very interesting submissions and we have chosen to ... Expand

Abstract: This Minitrack has attracted very interesting submissions and we have chosen to accept two.It is obvious that we are still puzzled with the evaluation and obscurity of AI algorithms, and research communities across the world are clear that have not conquered AI accountability. The co-chairs have agreed that it is quite difficult to address the visible shortcomings of AI and guarantee that we will have a means of evaluating its algorithms very soon, but we are also very excited, with the submitted papers, because they pave the way towards AI algorithms which address all three: evaluation, obscurity and accountability.The accepted submissions are very different and illustrate various ways of thinking and ideas which are hidden behind our concerns on AI. One paper gives a comprehensive explanation and possibilities of removing human bias when using natural language processing methods and the other is focused on ethical principles in AI which have been overlooked, and potentially harmed marginalized communities.The paper entitled "Comparing Methods for Mitigating Gender Bias in Word Embedding" gives a comprehensive comparison of debiasing methods, built on top of GloVe, in order to determine which method is the best in removing bias. It would be of interest to natural language processing and AI evaluation enthusiasts, because word embedding, which transforms texts into numeric vectors and reproduces similar words with similar vector representations, can carry forward prejudices of societies and thus bring along human biases. The authors have defined HSR-RAN-GloVe method and compared it with different debiasing methods with the aim of determining a) which methods perform the best at removing bias and b) if these methods truly remove bias from word embedding. The results are interesting. There is no method that outperforms throughout their analysis and no method leads to a decrease in performance for either similarity or semantics. However, RAN-GloVe is probably the method that on average gets better results in debiasing word embedding.In addition, it is worthwhile extracting from this paper that, for their SemBias dataset, the RAN-GloVe word embedding has remarkable accuracy in identifying the genderdefinition pairs. It does not confuse the genderdefinition with the gender-stereotype pairs, which can be extremely useful in any real-life applications.The paper entitled "Applying Reflexivity to Artificial Intelligence for Researching Marginalized Communities and Real-World Problems" has been created by US social scientists who claimed that the problems of ethical AI are due to a lack of critical insight into the complex positionality of the researcher, power dynamics between scholars and the communities being studied, and the structural impact on real-world problems when AI systems appear to be accurate but ethically fail. When addressing AI ethics, the authors apply 5 stages of "reflexivity", a process that yields a better understanding of community-specific nuances, areas requiring local expertise, and the potential consequences of scholastic interventions for real-world problems. They claim that as the focus of AI shifts to more practical, real-world interventions, it begins to enter the space that social work has inhabited. Therefore they advocate for AI researchers to be flexible by engaging in a more reflexive and ethical approach with the understanding that their AI model may produce unintended consequences that affect lives and livelihoods of vulnerable/ marginalized communities. We encourage readers to pay attention to the last paragraph of the paper, where the authors say that they "believe that the application of social work ethics and approaches to data science can possibly prevent future mistakes in our research with communities". Collapse

Topics: artificial intelligence

Methods: word embedding GloVe natural language processing literature sample

Multi-Channel Convolutional Neural Network for the Identification of Eyewitness Tweets of Disaster

2023 | Information Systems Frontiers | Citations: 0

Authors: Kumar, Abhinav; Singh, Jyoti Prakash; Rana, Nripendra P.; Dwivedi, Yogesh K.

Abstract: During a disaster, a large number of disaster-related social media posts are wid ... Expand

Abstract: During a disaster, a large number of disaster-related social media posts are widely disseminated. Only a small percentage of disaster-related information is posted by eyewitnesses. The post of a disaster eyewitness offers an accurate depiction of the disaster. Therefore, the information posted by the eyewitness is preferred over the other source of information as it is more effective at helping organize rescue and relief operations and potentially saving lives. In this work, we propose a multi-channel convolutional neural network (MCNN) that uses three different word-embedding vectors together to classify disaster-related tweets into eyewitness, non-eyewitness, and don’t know classes. We compared the performance of the proposed multi-channel convolutional neural network with several attention-based deep-learning models and conventional machine learning-models such as recurrent neural network, gated recurrent unit, Long-Short-Term-Memory, convolutional neural network, logistic regression, support vector machine, and gradient boosting. The proposed multi-channel convolutional neural network achieved an F1-score of 0.84, 0.88, 0.84, and 0.86 with four disaster-related datasets of floods, earthquakes, hurricanes, and wildfires, respectively. The experimental results show that the training MCNN model with different word embedding together performs better than the conventional machine-learning models and several other deep-learning models. Collapse

Topics: Twitter earthquake hurricane crowdsourcing social media

Methods: machine learning deep learning convolutional neural network experiment GloVe

A novel label-based multimodal topic model for social media analysis

2023 | Decision Support Systems | Citations: 0

Authors: Li, Hao; Qian, Yang; Jiang, Yuanchun; Liu, Yezheng; Zhou, Fan

Abstract: Extracting useful knowledge from multimodal data is the core of many multimedia ... Expand

Abstract: Extracting useful knowledge from multimodal data is the core of many multimedia applications, such as recommendation systems, and cross-modal retrieval. In this paper, we propose a label-based multimodal topic (LB-MMT) model to jointly model text and image data tagged with multiple labels. Specifically, we use the labels as supervised information to generate the text and image data. In the LB-MMT model, we assume that the textual words and visual words related to each text and image are drawn from a mixture of latent topics, where each topic is represented as a group of textual words and visual words. Moreover, we introduce multiple topics for each label, to build the top-down relationship from label to text and image. To investigate the effectiveness of the proposed approach, we conduct extensive experiments on a real-world multimodal dataset with labels. The re­ sults show the proposed approach obtains superior performances on topic coherence and label prediction compared with previous competitors. In addition, we show that our model yields interesting insights about multimodal topics. The proposed model provides important practical implications, e.g., designing more attrac­ tive multimodal contents for marketers. Collapse

Topics: Java social media Linux electronic commerce search engine

Methods: topic model word embedding Gibbs sampling BERT computational algorithm

Theories: technology acceptance model

Is a Pretrained Model the Answer to Situational Awareness Detection on Social Media?

2023 | HICSS | Citations: 0

Authors: Lo, S.; Lee, K.; Zhang, Y.

Abstract: Social media can be valuable for extracting information about an event or incid ... Expand

Abstract: Social media can be valuable for extracting information about an event or incident on the ground. However, the vast amount of content shared, and the linguistic variants of languages used on social media make it challenging to identify important situational awareness content to aid in decision-making for first responders. In this study, we assess whether pretrained models can be used to address the aforementioned challenges on social media. Various pretrained models, including static word embedding (such as Word2Vec and GloVe) and contextualized word embedding (such as DistilBERT) are studied in detail. According to our findings, a vanilla DistilBERT pretrained language model is insufficient to identify situation awareness information. Fine-tuning by using datasets of various event types and vocabulary extension is essential to adapt a DistilBERT model for real-world situational awareness detection. Collapse

Topics: Twitter situational awareness social media Wikipedia wiki

Methods: BERT DistilBERT support vector machine machine learning word embedding

Champions for Social Good: How Can We Discover Social Sentiment and Attitude-Driven Patterns in Prosocial Communication?

2023 | Journal of the Association for Information Systems | Citations: 0

Authors: Mukkamala, Raghava; Kauffman, Robert; Henriksen, Helle

Abstract: The UN High Commissioner on Refugees (UNHCR) is pursuing a social media strateg ... Expand

Abstract: The UN High Commissioner on Refugees (UNHCR) is pursuing a social media strategy to inform people about displaced populations and refugee emergencies. It is actively engaging public figures to increase awareness through its prosocial communications and improve social informedness and support for policy changes in its services. We studied the Twitter communications of UNHCR social media champions and investigated their role as high-profile influencers. In this study, we offer a design science research and data analytics framework and propositions based on the social informedness theory we propose in this paper to assess communication about UNHCR's mission. Two variables-refugee-emergency and champion type-relate to the informedness of UNHCR champions' followers. Based on a Twitter sentiment and attitude corpus, we applied a five-step design science analytics framework involving machine learning and natural language processing to test how the emergency type and champion type impact social communication patterns. Positive and neutral sentiment dominated the tweets of the champions and their followers for most refugee-emergency types. High participation-intensity champions emphasized high-intensity emergencies with tweet patterns reflecting dominant positive or neutral sentiment and sharing/liking attitudes. However, we found that sports figures effects were limited in spreading UNHCR's message, despite their millions of followers. We demonstrate the power of data science for prosocial policy based on refugee crisis awareness and instantiate our methods and knowledge contributions in a research framework that derives knowledge, decisions, and actions from behavioral, design, and economics of information systems perspectives. Collapse

Topics: Twitter social media analytical information system big data application programming interface

Methods: design science BERT natural language processing unsupervised learning artificial neural network

Theories: Lewin's field theory