Search for: All records

Feasible and developmentally appropriate sociotechnical approaches for protecting youth from online risks have become a paramount concern among human-computer interaction research communities. Therefore, we conducted 38 interviews with entrepreneurs, IT professionals, clinicians, educators, and researchers who currently work in the space of youth online safety to understand the different sociotechnical approaches they proposed to keep youth safe online, while overcoming key challenges associated with these approaches. We identified three approaches taken among these stakeholders, which included 1) leveraging artificial intelligence (AI)/machine learning to detect risks, 2) building security/safety tools, and 3) developing new forms of parental control software. The trade-offs between privacy and protection, as well as other tensions among different stakeholders (e.g., tensions toward the big-tech companies) arose as major challenges, followed by the subjective nature of risk, lack of necessary but proprietary data, and costs to develop these technical solutions. To overcome the challenges, solutions such as building centralized and multi-disciplinary collaborations, creating sustainable business plans, prioritizing human-centered approaches, and leveraging state-of-art AI were suggested. Our contribution to the body of literature is providing evidence-based implications for the design of sociotechnical solutions to keep youth safe online. 

Travel-time computation with large transportation networks is often computationally intensive for two main reasons: 1) large computer memory is required to handle large networks; and 2) calculating shortest-distance paths over large networks is computing intensive. Therefore, previous research tends to limit their spatial extent to reduce computational intensity or resolve computational intensity with advanced cyberinfrastructure. In this context, this article describes a new Spatial Partitioning Algorithm for Scalable Travel-time Computation (SPASTC) that is designed based on spatial domain decomposition with computer memory limit explicitly considered. SPASTC preserves spatial relationships required for travel-time computation and respects a user-specified memory limit, which allows efficient and large-scale travel-time computation within the given memory limit. We demonstrate SPASTC by computing spatial accessibility to hospital beds across the conterminous United States. Our case study shows that SPASTC achieves significant efficiency and scalability making the travel-time computation tens of times faster. 

Generating 3D graphs of symmetry-group equivariance is of intriguing potential in broad applications from machine vision to molecular discovery. Emerging approaches adopt diffusion generative models (DGMs) with proper re-engineering to capture 3D graph distributions. In this paper, we raise an orthogonal and fundamental question of in what (latent) space we should diffuse 3D graphs. ❶ We motivate the study with theoretical analysis showing that the performance bound of 3D graph diffusion can be improved in a latent space versus the original space, provided that the latent space is of (i) low dimensionality yet (ii) high quality (i.e., low reconstruction error) and DGMs have (iii) symmetry preservation as an inductive bias. ❷ Guided by the theoretical guidelines, we propose to perform 3D graph diffusion in a low-dimensional latent space, which is learned through cascaded 2D–3D graph autoencoders for low-error reconstruction and symmetry-group invariance. The overall pipeline is dubbed latent 3D graph diffusion. ❸ Motivated by applications in molecular discovery, we further extend latent 3D graph diffusion to conditional generation given SE(3)-invariant attributes or equivariant 3D objects. ❹ We also demonstrate empirically that out-of-distribution conditional generation can be further improved by regularizing the latent space via graph self-supervised learning. We validate through comprehensive experiments that our method generates 3D molecules of higher validity / drug-likeliness and comparable or better conformations / energetics, while being an order of magnitude faster in training. Codes are released at https://github.com/Shen-Lab/LDM-3DG. 

Misinformation in online spaces can stoke mistrust of established media, misinform the public and lead to radicalization. Hence, multiple automated algorithms for misinformation detection have been proposed in the recent past. However, the fairness (e.g., performance across left- and right- leaning news articles) of these algorithms has been repeatedly questioned, leading to decreased trust in such systems. This work motivates and grounds the need for an audit of machine learning based misinformation detection algorithms and possible ways to mitigate bias (if found). Using a large (N>100K) corpus of news articles, we report that multiple standard machine learning based misinformation detection approaches are susceptible to bias. Further, we find that an intuitive post-processing approach (Reject Option Classifier) can reduce bias while maintaining high accuracy in the above setting. The results pave the way for accurate yet fair misinformation detection algorithms. 

The prediction of Secondary Organic Aerosol (SOA) in regional scales is traditionally performed by using gas-particle partitioning models. In the presence of inorganic salted wet aerosols, aqueous reactions of semivolatile organic compounds can also significantly contribute to SOA formation. The UNIfied Partitioning-Aerosol phase Reaction (UNIPAR) model utilizes the explicit gas mechanism to better predict SOA formation from multiphase reactions of hydrocarbons. In this work, the UNIPAR model was incorporated with the Comprehensive Air Quality Model with Extensions (CAMx) to predict the ambient concentration of organic matter (OM) in urban atmospheres during the Korean-United States Air Quality (2016 KORUS-AQ) campaign. The SOA mass predicted with the CAMx-UNIPAR model changed with varying levels of humidity and emissions and in turn, has the potential to improve the accuracy of OM simulations. The CAMx-UNIPAR model significantly improved the simulation of SOA formation under the wet condition, which often occurred during the KORUS-AQ campaign, through the consideration of aqueous reactions of reactive organic species and gas-aqueous partitioning. The contribution of aromatic SOA to total OM was significant during the low-level transport/haze period (24-31 May 2016) because aromatic oxygenated products are hydrophilic and reactive in aqueous aerosols. The OM mass predicted with the CAMx-UNIPAR model was compared with that predicted with the CAMx model integrated with the conventional two product model (SOAP). Based on estimated statistical parameters to predict OM mass, the performance of CAMx-UNIPAR was noticeably better than the conventional CAMx model although both SOA models underestimated OM compared to observed values, possibly due to missing precursor hydrocarbons such as sesquiterpenes, alkanes, and intermediate VOCs. The CAMx-UNIPAR model simulation suggested that in the urban areas of South Korea, terpene and anthropogenic emissions significantly contribute to SOA formation while isoprene SOA minimally impacts SOA formation. 

In engineering, students’ completion of prerequisites indicates an understanding of fundamental knowledge. Recent studies have shown a significant relationship between student performance and prior knowledge. Weak knowledge retention from prerequisite coursework can present challenges in progressive learning. This study investigates the relationship between prior knowledge and students’ performance over a few courses of Statics. Statistics has been considered as the subject of interest since it is the introductory engineering course upon which many subsequent engineering courses rely, including many engineering analysis and design courses. The prior knowledge was determined based on the quantitative and qualitative preparedness. A quiz set was designed to assess quantitative preparedness. The qualitative preparedness was assessed using a survey asking students’ subjective opinions about their preparedness at the beginning of the semester. Student performance was later quantified through final course grades. Each set of data were assigned three categories for grouping purposes to reflect preparedness: 1) high preparedness: 85% or higher score, 2) medium preparedness: between 60% and 85%, and 3) weak preparedness: 60% or lower. Pearson correlation coefficient and T-test was conducted on 129 students for linear regression and differences in means. The analysis revealed a non-significant correlation between the qualitative preparedness and final scores (p-value = 0.29). The data revealed that students underestimated their understanding of the prerequisites for the class, since the quantitative preparedness scores were relatively higher than the qualitative preparedness scores. This can be partially understood by the time gap between when prerequisites were taken and when the course under investigation was taken. Students may have felt less confident at first but were able to pick up the required knowledge quickly. A moderately significant correlation between students’ quantitative preparedness and course performance was observed (p -value < 0.05). Students with high preparedness showed > 80% final scores, with a few exceptions; students with weak preparedness also showed relatively high final scores. However, most of the less prepared students made significant efforts to overcome their weaknesses through continuous communication and follow-up with the instructor. Despite these efforts, these students could not obtain higher than 90% as final scores, which indicates that level of preparedness reflects academic excellence. Overall, this study highlights the role of prior knowledge in achieving academic excellence for engineering. The study is useful to Civil Engineering instructors to understand the role of students’ previous knowledge in their understanding of difficult engineering concepts.