教学课程资源库 更多>>

关键词： Computer science   Scientific data  

摘要： Robust object detection is crucial for automating underwater marine debris collection. While supervised deep learning achieves state-of-the-art performance in discriminative tasks, replicating this success on underwater data is challenging. The generalization of these methods suffers due to a lack of available annotated data considering different sources of variation in the unstructured underwater environment and imaging conditions. In this paper, we present the Seaclear Marine Debris Dataset, the first publicly available shallow-water marine debris dataset annotated for instance segmentation/object detection. The dataset contains 8610 images collected using ROVs at multiple locations and with different cameras, annotated for 40 object categories, encompassing not only litter but also observed animals, plants, and robot parts. As part of the technical validation, we provide baseline results for object detection using Faster RCNN and YOLOv6 models. Furthermore, we demonstrate the non-triviality of generalizing the trained model performance to unseen sites and cameras due to domain shift. This underscores the value of the presented dataset in further developing robust models for underwater debris detection.

关键词： Computer science   Information technology  

摘要： It is important to understand the mathematical foundations of neural networks and to include robustness in model evaluation. Here, we introduce algorithms based on manifold curvature estimation to assess neural network robustness. These algorithms rely solely on training data and do not require regular or adversarial test data. Initially, a metric is proposed to measure the curvature of discrete data manifolds by introducing weighted angles concept between subspaces. Following this, a robustness measure is introduced that is independent of network architecture or model parameters. Lastly, two additional methods are introduced, utilizing curvature estimation of special manifolds formed by using gradient vectors between output and input network layers, alongside manifold curvature estimation. A comprehensive evaluation is provided on multiple network models using the CIFAR-10 dataset. Manifold geometry-based robustness analysis may lead to the development of not only accurate but also robust neural network models.

关键词： Medical imaging   Institutions   Computer science  

摘要： Radiological imaging is a globally prevalent diagnostic method, yet the free text contained in radiology reports is not frequently used for secondary purposes. Natural Language Processing can provide structured data retrieved from these reports. This paper provides a summary of the current state of research on Large Language Model (LLM) based approaches for information extraction (IE) from radiology reports. We conduct a scoping review that follows the PRISMA-ScR guideline. Queries of five databases were conducted on August 1st 2023. Among the 34 studies that met inclusion criteria, only pre-transformer and encoder-based models are described. External validation shows a general performance decrease, although LLMs might improve generalizability of IE approaches. Reports related to CT and MRI examinations, as well as thoracic reports, prevail. Most common challenges reported are missing validation on external data and augmentation of the described methods. Different reporting granularities affect the comparability and transparency of approaches.

关键词： Cloud computing   Secure storage   Privacy   Computer science   Task analysis   Intserv networks   Games  

摘要： Cloud storage auditing mechanism is used to check whether the data of users stored in the cloud is intact. Most existing auditing schemes for secure cloud storage are designed to check the integrity for specified files based on file identities. In some scenarios, the user would like to check the integrity of the files generated and uploaded to the cloud in a certain time period. Existing cloud storage auditing schemes cannot work well for supporting this practical requirement because the private information will be exposed. To satisfy this requirement, we propose a brand-new paradigm termed as privacy-preserving time-based auditing for secure cloud storage. The proposed paradigm allows the user to check whether the files generated and uploaded in a certain time period are intactly stored in the cloud. When intending to check the integrity of the files uploaded in this time period, the user only provides the challenged time period t to the Third Party Auditor (TPA). The TPA can verify the integrity of all the files based on this time period, but cannot know how many files and which files the user has generated and uploaded to the cloud in this time period. To decrease the complex overhead associated with certificate management, we introduce an identity-based auditing mechanism. We provide a specific security analysis to show the correctness, auditing soundness and privacy preserving of this scheme. The experiments demonstrate the efficiency of the proposed scheme.

关键词： Integrated computing   computational thinking   CS+X   content analysis   curricula analysis   K-12 computer science education   primary school computer science education  

摘要： Integrated computing curricula combine learning objectives in computing with those in another discipline, like literacy, math, or science, to give all students experience with computing, typically before they must decide whether to take standalone CS courses. One goal of integrated computing curricula is to provide an accessible path to an introductory computing course by introducing computing concepts and practices in required courses. This study analyzed integrated computing curricula to determine which CS practices and concepts are taught, how extensively the curricula are taught, and, by extension, how they might prepare students for later computing courses. The authors conducted a content analysis to examine primary and lower secondary (i.e., K-8) curricula that are taught in non-CS classrooms, have explicit CS learning objectives (i.e., CS+X), and that took 5+ hours to complete. Lesson plans, descriptions, and resources were scored based on frameworks developed from the K-12 CS Framework, including programming concepts, non-programming CS concepts, and CS practices. The results found that curricula most extensively taught introductory concepts and practices, such as sequences, and rarely taught more advanced content, such as conditionals. Students who engage with most of these curricula would have no experience working with fundamental concepts, like variables, operators, data collection or storage, or abstraction in the context of a program. While this focus might be appropriate for integrated curricula, it has implications for the prior knowledge that students should be expected to have when starting standalone computing courses.

关键词： Student voice   equity   ethics and social responsibility with technology  

摘要： Culturally responsive computing (CRC), that centers sociopolitical issues and transformational uses of technology, has been described as valuable for increasing engagement with computing, especially for historically underrepresented minoritized students. But what do high school students think? Through a sociocultural lens prioritizing student voices recorded in 56 interviews over a period of 2 years (1–3 years after students’ first experience with computer science (CS) education through Exploring Computer Science or Advanced Placement CS Principles in high school), this study centers the perspectives of 39 primarily low-income, Latine and Black youth from urban California and rural Mississippi public schools to understand what they perceive as the role of technology in our world and what they subsequently desire of their computing education. While none have studied CRC before, the majority responded with CRC ideas about the kind of pedagogy they believe would make for a more meaningful computing learning experience: They see computing as a form of power that impacts both good and bad in the world and want computing educators to prepare them to take on these issues of equity, ethics, social responsibility, and underrepresentation in the field. The students’ perspectives offer important pedagogical insight into how to support deeper engagement with computing in current CS for All initiatives, while also preparing youth for the rapidly evolving and increasingly complex computing landscape that impacts all of our lives.

关键词： Multimodal generative AI   ChatGPT-4 vision   educational assessment   computer science education  

摘要： The recent integration of visual capabilities into Large Language Models (LLMs) has the potential to play a pivotal role in science and technology education, where visual elements such as diagrams, charts, and tables are commonly used to improve the learning experience. This study investigates the performance of ChatGPT-4 Vision, OpenAI’s most advanced visual model at the time the study was conducted, on the Bachelor in Computer Science section of Brazil’s 2021 National Undergraduate Exam (ENADE). By presenting the model with the exam’s open and multiple-choice questions in their original image format and allowing for reassessment in response to differing answer keys, we were able to evaluate the model’s reasoning and self-reflecting capabilities in a large-scale academic assessment involving textual and visual content. ChatGPT-4 Vision significantly outperformed the average exam participant, positioning itself within the top 10 best score percentile. While it excelled in questions that incorporated visual elements, it also encountered challenges with question interpretation, logical reasoning, and visual acuity. A positive correlation between the model’s performance in multiple-choice questions and the performance distribution of the human participants suggests multimodal LLMs can provide a useful tool for question testing and refinement. However, the involvement of an independent expert panel to review cases of disagreement between the model and the answer key revealed some poorly constructed questions containing vague or ambiguous statements, calling attention to the critical need for improved question design in future exams. Our findings suggest that while ChatGPT-4 Vision shows promise in multimodal academic evaluations, human oversight remains crucial for verifying the model’s accuracy and ensuring the fairness of high-stakes educational exams. The paper’s research materials are publicly available at https://***/nabormendonca/gpt-4v-enade-cs-2021.

关键词： BPA   Trophic transfer   Fluctuating temperature   Climate change   Moulting   Growth rate   ENDOCRINE DISRUPTING COMPOUNDS   BISPHENOL-A BPA   POPULATION-GROWTH   FOOD QUALITY   WATER   ZOOPLANKTON   STRESS   BIOACCUMULATION   RESPONSES   TOXICITY  

摘要： Bisphenol A (BPA), a synthetic organic compound widely used in the production of plastics, is recognized as an emerging contaminant because of its toxicity and the potential risks associated with bioaccumulation in organisms. Despite potential environmental hazards, there is a lack of studies examining BPA toxicity mechanisms and its potential impact on various trophic levels, with even fewer exploring whether global stressors such as temperature can affect the toxicity of BPA in organisms. Our aim was to assess the combined impact of BPA and varying temperature regimes on life-history traits in Daphnia magna. Our results revealed a significant impact of BPA on the growth, reproduction, and accumulated moulting of D. magna, with adverse effects primarily associated with the assimilation of BPA in algae rather than the BPA present in the medium, pointing to a trophic transfer mechanism. The interactive effect between BPA and temperature demonstrated a slight stimulatory effect of low BPA level on D. magna growth rate under warming constant conditions, but an inhibitory under warming fluctuating temperatures. Additionally, a BPA threshold was identified, below which growth became temperature-dependent. This study emphasizes the crucial role of considering temperature in predicting how toxins may affect Daphnia within aquatic food webs.

摘要： This paper evaluates the impact of an English and Business Computer Science program in the Democratic Republic of Congo (DRC), the world’s second-largest French-speaking country after France. The program addresses the gap between academic training and market needs by combining English language proficiency with essential business and computer science skills. This combination aims to equip students for success in the business and technology sectors, where English is a key language. The paper details the program’s implementation, outcomes, and evaluation, using surveys of employers and graduates, along with an analysis of student competencies and employability. The findings underscore the importance of English proficiency, along with business and technological skills, in aligning educational outcomes with job market demands, thereby improving student success.

关键词： Computer science   Machine learning   Society  

摘要： The Large Language Models that power chatbots are known to struggle in languages outside of English — this podcast explores how this challenge can be overcome.