This paper studies an emerging and important problem of identifying misleading COVID-19 short videos where the content is jointly expressed in visual, audio, textual videos. Existing solutions for video detection mainly focus on authenticity or audios against AI algorithms (e.g., deepfake) manipulation, are insufficient to address our most user-generated intentionally edited. Two critical challenges exist solving problem: i) how effectively extract information from distractive manipulated...

The proliferation of social media has promoted the spread misinformation that raises many concerns in our society. This paper focuses on a critical problem explainable COVID-19 detection aims to accurately identify and explain misleading claims media. Motivated by lack relevant knowledge existing solutions, we construct novel crowdsource graph based approach incorporate facts leveraging collaborative efforts expert non-expert crowd workers. Two important challenges exist developing solution:...

This paper focuses on a critical problem of explainable multimodal COVID-19 misinformation detection where the goal is to accurately detect misleading information in news articles and provide reason or evidence that can explain results. Our work motivated by lack judicious study association between different modalities (e.g., text image) content current solutions. In this paper, we present generative approach investigating cross-modal visual textual deeply embedded content. Two challenges...

Enzyme biocatalysis for plastic treatment and recycling is an emerging field of growing interest. However, it challenging time-consuming to identify plastic-degrading enzymes with desirable functionality, given the large number putative enzyme sequences. There a critical need develop effective approach accurately predict activity in degrading different types plastics. In this study, we developed machine-learning-based enzymatic degradation (PED) framework ability degrade plastics interest by...

With the ever-increasing number of road traffic accidents worldwide, safety has become a critical problem in intelligent transportation systems. A key step towards improving is to identify locations where severe happen with high probability so precautions can be applied effectively. We refer this as risky location identification. While previous efforts have been made address similar problems, two important limitations exist: i) data availability: many cities (especially developing countries)...

With the increasing popularity of online social media (e.g., Facebook, Twitter, Reddit), detection misleading content on has become a critical undertaking. This paper focuses an important but largely unsolved problem: detecting fauxtography (i.e., posts with images). We found that existing literature falls short in solving this problem. In particular, current solutions either focus fake images or misinformed texts post. However, they cannot solve our problem because depends not only...

Forecasting traffic accidents at a fine-grained spatial scale is essential to provide effective precautions and improve safety in smart urban sensing applications. Current solutions primarily rely on complete historical accident records and/or accurate real-time sensor data for risk prediction. These are prone various limitations (e.g., facility availability, privacy legal constraints). In this paper, we address those by exploring two types of widely available complementary sources: social...

Drought has become a critical global threat with significant societal impact. Existing drought monitoring solutions primarily focus on assessing severity using quantitative measurements, overlooking the diverse impact of from human-centric perspectives. Motivated by collective intelligence social media and computational power AI, this paper studies novel problem socially informed AI-driven estimation that aims to leverage news information jointly estimate its Two technical challenges exist:...

Despite recent progress in improving the performance of misinformation detection systems, classifying an unseen domain remains elusive challenge. To address this issue, a common approach is to introduce critic and encourage domain-invariant input features. However, early often demonstrates both conditional label shifts against existing data (e.g., class imbalance COVID-19 datasets), rendering such methods less effective for detecting misinformation. In paper, we propose contrastive...

With emerging topics (e.g., COVID-19) on social media as a source for the spreading misinformation, overcoming distributional shifts between original training domain (i.e., domain) and such target domains remains non-trivial task misinformation detection. This presents an elusive challenge early-stage detection, where good amount of data annotations from is not available training. To address scarcity issue, we propose MetaAdapt, meta learning based approach adaptive few-shot MetaAdapt...

Training large deep learning (DL) models with high performance for natural language downstream tasks usually requires rich-labeled data. However, in a real-world application of COVID-19 information service (e.g., misinformation detection, question answering), fundamental challenge is the lack labeled COVID data to enable supervised end-to-end training different tasks, especially at early stage pandemic. To address this challenge, we propose an unsupervised domain adaptation framework using...

While sequential recommender systems achieve significant improvements on capturing user dynamics, we argue that recommenders are vulnerable against substitution-based profile pollution attacks. To demonstrate our hypothesis, propose a adversarial attack algorithm, which modifies the input sequence by selecting certain elements and substituting them with items. In both untargeted targeted scenarios, observe performance deterioration using proposed algorithm. Motivated such observations,...

Cultural heritage sites are precious and fragile resources that hold significant historical, esthetic, social values in our society. However, the increasing frequency severity of natural man-made disasters constantly strike cultural with damages. In this article, we focus on a damage assessment (CHDA) problem where goal is to accurately locate damaged area site using imagery data posted media during disaster event by exploring collective strengths both AI human intelligence from...

In this vision paper, we propose a new concept, "Social Edge Intelligence (SEI)", where the artificial intelligence (AI) and human (HI) are tightly integrated to address set of critical research challenges in edge computing applications. The SEI concept is motivated by two technical trends: 1) recent rapid advancement AI techniques many mobile applications (e.g., sensing, smart homes, intelligent transportation systems), 2) emergence crowdsourcing platforms Amazon MTurk, Waze) that used...

Autonomous unmanned aerial vehicles (UAVs) have become an important tool for efficient disaster response. Despite the virtues of UAVs in response applications, various limitations (e.g., requiring manual input, finite battery life) hinder their mass adoption. In contrast, social sensing is emerging as a new paradigm that utilizes signals provided by "human sensors" to gather awareness events occurring physical world. being inherently broader scope, shortcoming reliability data are...

Fauxtography is a category of multi-modal posts that spreads misleading information on various online social platforms (e.g., Facebook, Twitter, Reddit). A fauxtography post usually consists an image, text description and comments from its readers. In this paper, we focus explainable detection problem where the goal to explain which specific component leads decision. This motivated by limitations current solutions only but ignore important explanation aspect their results. Two critical...

Social sensing is emerging as an effective and pervasive paradigm to collect timely data observations from human sensors. This paper focuses on the problem of COVID-19 misinformation detection social media. Our work motivated by lack COVID-specific knowledge in current solutions, which critical assess truthfulness media claims about disease. In this paper, we leverage intelligence a crowdsourcing platform obtain essential facts for detecting Two challenges exist solving our problem: i) how...