More Like this

1. Backbone Index and GNN Models for Skyline Path Query Evaluation over Multi-cost Road Networks

   https://doi.org/10.1145/3660632  

   Gong, Qixu; Chen, Huiying; Cao, Huiping; Liu, Jiefei (April 2024, ACM Transactions on Spatial Algorithms and Systems)

   Skyline path queries (SPQs) extend skyline queries to multi-dimensional networks, such as multi-cost road networks (MCRNs). Such queries return a set of non-dominated paths between two given network nodes. Despite the existence of extensive works on evaluating different SPQ variants, SPQ evaluation is still very inefficient due to the nonexistence of efficient index structures to support such queries. Existing index building approaches for supporting shortest-path query execution, when directly extended to support SPQs, use an unreasonable amount of space and time to build, making them impractical for processing large graphs. In this paper, we propose a novel index structure,backbone index, and a corresponding index construction method that condenses an initial MCRN to multiple smaller summarized graphs with different granularity. We present efficient approaches to find approximate solutions to SPQs by utilizing the backbone index structure. Furthermore, considering making good use of historical query and query results, we propose two models,SkylinePathGraphNeuralNetwork (SP-GNN) andTransfer SP-GNN (TSP-GNN), to support effective SPQ processing. Our extensive experiments on real-world large road networks show that the backbone index can support finding meaningful approximate SPQ solutions efficiently. The backbone index can be constructed in a reasonable time, which dramatically outperforms the construction of other types of indexes for road networks. As far as we know, this is the first compact index structure that can support efficient approximate SPQ evaluation on large MCRNs. The results on the SP-GNN and TSP-GNN models also show that both models can help get approximate SPQ answers efficiently. 

   more » « less
2. Assessing (Social-Ecological) Systems Thinking by Evaluating Cognitive Maps

   https://doi.org/10.3390/su11205753  

   Gray, Steven; Sterling, Eleanor J.; Aminpour, Payam; Goralnik, Lissy; Singer, Alison; Wei, Cynthia; Akabas, Sharon; Jordan, Rebecca C.; Giabbanelli, Philippe J.; Hodbod, Jennifer; et al (October 2019, Sustainability)

   null (Ed.)

   Systems thinking (ST) skills are often the foundation of sustainability science curricula. Though ST skill sets are used as a basic approach to reasoning about complex environmental problems, there are gaps in our understanding regarding the best ways to promote and assess ST learning in classrooms. Since ST learning provides Science, Technology, Engineering, and Mathematics (STEM) students’ important skills and awareness to participate in environmental problem-solving, addressing these gaps is an important STEM learning contribution. We have created guidelines for teaching and measuring ST skills derived from a hybrid of a literature review and through case study data collection. Our approach is based on semi-quantitative cognitive mapping techniques meant to support deep reasoning about the complexities of social–ecological issues. We begin by arguing that ST should be evaluated on a continuum of understanding rather than a binary of correct/incorrect or present/absent. We then suggest four fundamental dimensions of teaching and evaluating ST which include: (1) system structure, (2) system function, (3) identification of leverage points for change, and (4) trade-off analysis. Finally, we use a case study to show how these ideas can be assessed through cognitive maps to help students develop deep system understanding and the capacity to propose innovative solutions to sustainability problems. 

   more » « less
3. A Novel Deep Subspace Learning Framework to Automatically Uncover Assessment-Specific Independent Brain Networks

   https://doi.org/10.1109/CISS59072.2024.10480204  

   Batta, Ishaan; Abrol, Anees; Calhoun, Vince (March 2024, IEEE)

   We present a novel deep learning framework to automatically compute independently salient networks in the brain that characterize the underlying changes in the brain in association with clinically observed assessments. Unsupervised approaches for high-dimensional neuroimaging data focus on computing low-dimensional brain components for subsequent analysis, while supervised learning approaches aim for predictive performance and yielding a single list of associative feature importance, thus making it hard to interpret at the level of brain subsystems. Our approach integrates the goals of decomposition into lower dimensional subspaces and, identifying salient brain subsystems into a single automated framework. We first train a convolutional neural network on structural brain features to predict clinical assessments, followed by a multi-step decomposition in the saliency space to compute salient brain networks that intrinsically characterize the brain changes associated with the assessment. Through a repeated training procedure on an Alzheimer’s disease (AD) dataset, we show that our method effectively computes AD-related salient brain subsystems directly from high-dimensional neuroimaging data, while maintaining predictive performance. Such approaches are crucial for data-driven biomarker development for brain disorders. 

   more » « less
4. Analyzing students’ systems thinking in-situ through screencasts in the context of computational modeling: a case study

   https://doi.org/10.1186/s43031-024-00115-7  

   Bowers, Jonathan; Eidin, Emil (November 2024, Disciplinary and Interdisciplinary Science Education Research)

   Abstract In our interconnected world, Systems Thinking (ST) is increasingly being recognized as a key learning goal for science education to help students make sense of complex phenomena. To support students in mastering ST, educators are advocating for using computational modeling programs. However, studies suggest that students often have challenges with using ST in the context of computational modeling. While previous studies have suggested that students have challenges modeling change over time through collector and flow structures and representing iterative processes through feedback loops, most of these studies investigated student ST through pre and post tests or through interviews. As such there is a gap in the literature regarding how student ST approaches develop and change throughout a computational modeling unit. In this case study, we aimed to determine which aspects of ST students found challenging during a computational modeling unit, how their approaches to ST changed over time, and how the learning environment was supporting students with ST. Building on prior frameworks, we developed a seven-category analysis tool that enabled us to use a mixture of student discourse, writing, and screen actions to categorize seven ST behaviors in real time. Through using this semi-quantitative tool and subsequent narrative analysis, we found evidence for all seven behavior categories, but not all categories were equally represented. Meanwhile our results suggest that opportunities for students to engage in discourse with both their peers and their teacher supported them with ST. Overall, this study demonstrates how student discourse and student writing can be important evidence of ST and serve as a potential factor to evaluate ST application as part of students’ learning progression. The case study also provides evidence for the positive impact that the implementation of a social constructivist approach has in the context of constructing computational system models. 

   more » « less
5. Automatic Short Math Answer Grading via In-context Meta-learning

   Zhang, Mengxue; Baral, Sami; Heffernan, Neil; Lan, Andrew (July 2022, International Conference on Educational Data Mining)

   Automatic short answer grading is an important research di- rection in the exploration of how to use artificial intelligence (AI)-based tools to improve education. Current state-of-the- art approaches use neural language models to create vector- ized representations of students responses, followed by clas- sifers to predict the score. However, these approaches have several key limitations, including i) they use pre-trained lan- guage models that are not well-adapted to educational sub- ject domains and/or student-generated text and ii) they al- most always train one model per question, ignoring the link- age across question and result in a significant model storage problem due to the size of advanced language models. In this paper, we study the problem of automatic short answer grad- ing for students’ responses to math questions and propose a novel framework for this task. First, we use MathBERT, a variant of the popular language model BERT adapted to mathematical content, as our base model and fine-tune it on the downstream task of student response grading. Sec- ond, we use an in-context learning approach that provides scoring examples as input to the language model to provide additional context information and promote generalization to previously unseen questions. We evaluate our framework on a real-world dataset of student responses to open-ended math questions and show that our framework (often signif- icantly) outperform existing approaches, especially for new questions that are not seen during training. 

   more » « less