检索结果-内蒙古大学图书馆

Bias analysis of AI models for undergraduate student admissions

Neural computing and Applications 2025年第12期37卷 7785-7795页

作者： Van Busum, Kelly Fang, Shiaofen Department of Computer Science and Software Engineering Butler University IndianapolisIN United States Computing and Engineering Indiana University Luddy School of Informatics Indiana University Indianapolis IndianapolisIN United States

Bias detection and mitigation is an active area of research in machine learning. This work extends previous research done by the authors Van Busum and Fang (Proceedings of the 38th ACM/SIGAPP Symposium on Applied computing, 2023) to provide a rigorous and more complete analysis of the bias found in AI predictive models. Admissions data spanning six years was used to create an AI model to determine whether a given student would be directly admitted into the School of Science under various scenarios at a large urban research university. During this time, submission of standardized test scores as part of a student’s application became optional which led to interesting questions about the impact of standardized test scores on admission decisions. We developed and analyzed AI models to understand which variables are important in admissions decisions, and how the decision to exclude test scores affects the demographics of the students who are admitted. We then evaluated the predictive models to detect and analyze biases these models may carry with respect to three variables chosen to represent sensitive populations: gender, race, and whether a student was the first in his/her family to attend college. We also extended our analysis to show that the biases detected were persistent. Finally, we included several fairness metrics in our analysis and discussed the uses and limitations of these metrics. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Maximizing Energy Efficiency of Mobile Biomechanical Decoding with Spatial Variability Mining

Maximizing Energy Efficiency of Mobile Biomechanical Decodin...

引用

2024 IEEE International Conference on Consumer Electronics, ICCE 2024

作者： Gangadharan, Kiirthanaa Daniel, Cody Zhang, Qingxue Purdue University School of Engineering and Technology Electrical and Computing Engineering Indiana United States Purdue University School of Engineering and Technology Electrical and Computing Engineering Biomedical Engineering Indiana United States

ISBN: (纸本)9798350324136

Mobile sensing and data analytics usually take a substantial amount of energy, which limits the durability of the wearable devices. Especially, when deep learning is applied for data mining, the energy need is even more hungry. In this study, we take a special interest in wearable biomechanical data analytics, and propose to leverage data mining to tackle this challenge. More specifically, we leverage deep learning to mine the spatial variability of motion sensors embedded in mobile devices placed on the forearm and upper arm. In addition to compare the sensor locations, we further compare six different channels of each sensor location, thereby making twelve different configurations. Ultimately, we determine the optimal sensor location and the optimal sensor channel, which indicates the minimized data processing need in the real-world applications. The results indicate that the upper arm location and Y-axis of the accelerometer is the optimal configuration. This study will advance the field of maximizing energy efficiency of mobile biomechanical monitors, towards continuous data-driven precision medicine. © 2024 IEEE.

关键词： Big Data Deep Learning Mobile Devices Spatial Variability

来源：评论

学校读者我要写书评

暂无评论

The Relationships between Influencers and Followers Who Troll and Proactive-Reactive Trolling

引用

Proceedings of the Association for Information Science and Technology 2024年第1期61卷 1093-1095页

作者： Sun, Honglei Fichman, Pnina School of Information Management Nanjing University China Luddy School of Informatics Computing and Engineering Indiana University Bloomington United States

Little is known about the differences between trolling by influencers and followers, the difference between proactive and reactive trolling, and the relationship between them. Based on a content analysis of 1,386 comments on 160 posts of 19 influencers during a Sina Weibo trolling event, we found that influencers troll more frequently than their followers and they troll proactively more often than reactively, while followers troll reactively more often than proactively. Influencers generated less but more impactful content compared with their followers. In both proactive and reactive trolling, influencers derailed the discussion while followers provoked through their proactive trolling and insulted through their reactive trolling. These findings extend the research scope of trolling asymmetry and establish theoretical connections between social media roles and proactive-reactive trolling. 87 Annual Meeting of the Association for Information Science & Technology | Oct. 25 – 29, 2024 | Calgary, AB, Canada.

关键词： Proactive-Reactive trolling Sina Weibo Social media influencers

来源：评论

学校读者我要写书评

暂无评论

Gradual Guarantee via Step-Indexed Logical Relations in Agda

Gradual Guarantee via Step-Indexed Logical Relations in Agda

引用

2024 A Second Soul: Celebrating the Many Languages of Programming - Festschrift in Honor of Peter Thiemann's Sixtieth Birthday, PT 2024

作者： Siek, Jeremy G. School of Informatics Computing and Engineering Indiana University BloomingtonIN United States

The gradual guarantee is an important litmus test for gradually typed languages, that is, languages that enable a mixture of static and dynamic typing. The gradual guarantee states that changing the precision of a type annotation does not change the behavior of the program, except perhaps to trigger an error if the type annotation is incorrect. Siek et al. (2015) proved that the Gradually Typed Lambda Calculus (GTLC) satisfies the gradual guarantee using a simulation-based proof and mechanized their proof in Isabelle. In the following decade, researchers have proved the gradual guarantee for more sophisticated calculi, using step-indexed logical relations. However, given the complexity of that style of proof, there has not yet been a mechanized proof of the gradual guarantee using step-indexed logical relations. This paper reports on a mechanized proof of the gradual guarantee for the GTLC carried out in the Agda proof assistant. © Jeremy G. Siek.

关键词： Differentiation (calculus)

来源：评论

学校读者我要写书评

暂无评论

Designing Everyday Family Technologies: Supporting Health and Wellbeing in Family-Food Interactions with Children 25

Designing Everyday Family Technologies: Supporting Health an...

引用

2025 ACM International Conference on Supporting Group Work, GROUP Companion 2025

作者： Wu, Yuxing Luddy School of Informatics Computing and Engineering Indiana University Bloomington BloomingtonIN United States

ISBN: (纸本)9798400711879

Technology is pervasive in everyday family life. However, creating meaningful sociotechnical systems that are useful for (and desired by) families remains a complex and evolving challenge. To design for families in the digital age, HCI research requires holistic understanding and in-depth investigation to unpack social dynamics and situated use beyond the individual. My dissertation applies a family-centered approach to examine how technologies might be better designed to support everyday social, health, and developmental goals and challenges. Focusing on family-food interactions, my work unpacks complexities in navigating various food-related experiences to inform design implications. © 2025 Copyright held by the owner/author(s).

关键词： Food ingredients

来源：评论

学校读者我要写书评

暂无评论

Exploring Users’ Perspectives of Mobile Health Privacy and Autonomy 17th

Exploring Users’ Perspectives of Mobile Health Privacy and ...

引用

17th EAI International Conference on Pervasive computing Technologies for Healthcare, PervasiveHealth 2023

作者： Starks, Thomas Patil, Kshitij Martin-Hammond, Aqueasha School of Informatics Computing and Engineering Indiana University – Indianapolis IndianapolisIN46202 United States

ISBN: (纸本)9783031597169

The increased use of mobile health (mHealth) applications and the corresponding exchange of sensitive data has underscored privacy concerns. Privacy notices are often unengaging or incomprehensible, leading to questions of informed consent and trust. While studies have focused on providing solutions aimed to simplify privacy language and reduce cognitive burden, often overlooked are the behavioral aspects of individual attitudes, norms, and perceived control that lead to dynamic intentions for engagement. In this paper, we use existing behavior models as a lens to understand users’ privacy experiences, behaviors, and perspectives toward mHealth data privacy policies. In 15 semi-structured interviews with adult users of mHealth applications, participants encountered persistent challenges when engaging and articulating the value of privacy. Participants do not understand how privacy notices are designed, which leads to superficial awareness and control that does not actually support their perceptions of autonomy and trust in mHealth. As a result, users felt sub-optimal autonomy when engaging in privacy interactions. We discuss design considerations for autonomy-supporting privacy notices that may help users feel a greater sense of agency when interacting with mHealth applications. © ICST Institute for Computer Sciences, Social Informatics and Telecommunications engineering 2024.

关键词： mHealth

来源：评论

学校读者我要写书评

暂无评论

LLM Reliability and CBR: How Case Based Reasoning Can Improve the Performance of Large Language Models 32

LLM Reliability and CBR: How Case Based Reasoning Can Improv...

引用

Workshops at the 32nd International Conference on Case-Based Reasoning, ICCBR-WS 2024

作者： Wilkerson, Kaitlynne Luddy School of Informatics Computing and Engineering Indiana University BloomingtonIN47408 United States

Large Language Models (LLMs) are known to make factual errors and hallucinate. This project overview discusses current and future research methods of improving the accuracy, interpretability and explainability of LLMs... 详细信息

关键词： Case based reasoning

来源：评论

学校读者我要写书评

暂无评论

Wav2pos: Exploring syntactic analysis from audio for Highland Puebla Nahuatl 4

Wav2pos: Exploring syntactic analysis from audio for Highlan...

引用

4th Workshop on Natural Language Processing for Indigenous Languages of the Americas, AmericasNLP 2024

作者： Pugh, Robert Sreedhar, Varun Tyers, Francis Luddy School of Informatics Computing and Engineering Indiana University Bloomington United States Department of Linguistics Indiana University Bloomington United States

ISBN: (纸本)9798891761087

We describe an approach to part-of-speech tagging from audio with very little human-annotated data, for Highland Puebla Nahuatl, a low-resource language of Mexico.1 While automatic morphosyntactic analysis is typically trained on annotated textual data, large amounts of text is rarely available for low-resource, marginalized, and/or minority languages, and morphosyntactically-annotated data is even harder to come by. Much of the data from these languages may exist in the form of recordings, often only partially-transcribed or analyzed by field linguists working on language documentation projects. Given this relatively low-availability of text in the low-resource language scenario, we explore end-to-end automated morphosyntactic analysis directly from audio. The experiments described in this paper focus on one piece of morphosyntax, part-of-speech tagging, and builds on existing work in a high-resource setting. We use weak supervision to increase training volume, and explore a few techniques for generating word-level predictions from the acoustic features. Our experiments show promising results, despite less than 400 sentences of audio-aligned, manually-labeled text. © 2024 Association for Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Exploring Quantum Processors in Decentralized Edge computing for Improved Data Processing 2

Exploring Quantum Processors in Decentralized Edge Computing...

引用

2nd International Conference on Machine Learning and Autonomous Systems, ICMLAS 2025

作者： Pitkar, Harshad Luddy School of Informatics Computing and Engineering Indiana University BloomingtonIN United States

ISBN: (纸本)9798331505745

The rapid growth of data generated by Internet of Things (IoT) devices necessitates the development of advanced computational frameworks that can efficiently handle real-time data processing. Traditional cloud and edge computing systems, while effective, face significant challenges, including high latency, limited computational resources, and data transfer bottlenecks. To address these issues, this study introduces a hybrid approach that integrates quantum computing into the edge-cloud continuum. The proposed method utilizes Quantum- Neural Networks (QNNs) in conjunction with local edge processing, leveraging quantum autoencoders for feature compression, and incorporating fully decentralized edge computing with quantum processors to enhance data processing capabilities. Additionally, a hybrid model combining QNNs with classical machine learning techniques is implemented at the edge for improved classification accuracy. The proposed system's performance is evaluated through extensive experiments across multiple datasets, comparing the results of quantum-based models with traditional classical and hybrid approaches. The results show a significant improvement in classification accuracy, reduced feature compression time, and enhanced processing efficiency, particularly in resource-constrained environments. The results underscore the potential of quantum machine learning in real-time, large-scale applications, marking an important step toward the practical integration of quantum computing in distributed computing systems. © 2025 IEEE.

关键词： Qubits

来源：评论

学校读者我要写书评

暂无评论

LSTM-QGAN: Scalable NISQ Generative Adversarial Network

LSTM-QGAN: Scalable NISQ Generative Adversarial Network

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Chu, Cheng Hastak, Aishwarya Chen, Fan Luddy School of Informatics Computing and Engineering Indiana University Bloomington United States

ISBN: (纸本)9798350368741

Current quantum generative adversarial networks (QGANs) still struggle with practical-sized data. First, many QGANs use principal component analysis (PCA) for dimension reduction, which, as our studies reveal, can diminish the QGAN's effectiveness. Second, methods that segment inputs into smaller patches processed by multiple generators face scalability issues. In this work, we propose LSTM-QGAN, a QGAN architecture that eliminates PCA preprocessing and integrates quantum long short-term memory (QLSTM) to ensure scalable performance. Our experiments show that LSTM-QGAN significantly enhances both performance and scalability over state-of-the-art QGAN models, with visual data improvements, reduced Fréchet Inception Distance scores, and reductions of 5× in qubit counts, 5× in single-qubit gates, and 12× in two-qubit gates. © 2025 IEEE.

关键词： Long Short-Term Memory NISQ Quantum Generative Adversarial Network

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：