Debiaser for Multiple Variables to enhance fairness in classification tasks     

《Information processing & management》2023,60(2):103226

Nowadays assuring that search and recommendation systems are fair and do not apply discrimination among any kind of population has become of paramount importance. This is also highlighted by some of the sustainable development goals proposed by the United Nations. Those systems typically rely on machine learning algorithms that solve the classification task. Although the problem of fairness has been widely addressed in binary classification, unfortunately, the fairness of multi-class classification problem needs to be further investigated lacking well-established solutions. For the aforementioned reasons, in this paper, we present the Debiaser for Multiple Variables (DEMV), an approach able to mitigate unbalanced groups bias (i.e., bias caused by an unequal distribution of instances in the population) in both binary and multi-class classification problems with multiple sensitive variables. The proposed method is compared, under several conditions, with a set of well-established baselines using different categories of classifiers. At first we conduct a specific study to understand which is the best generation strategies and their impact on DEMV’s ability to improve fairness. Then, we evaluate our method on a heterogeneous set of datasets and we show how it overcomes the established algorithms of the literature in the multi-class classification setting and in the binary classification setting when more than two sensitive variables are involved. Finally, based on the conducted experiments, we discuss strengths and weaknesses of our method and of the other baselines.  相似文献   

Artificial intelligence implication on energy sustainability in Internet of Things: A survey     

《Information processing & management》2023,60(2):103212

The massive number of Internet of Things (IoT) devices connected to the Internet is continuously increasing. The operations of these devices rely on consuming huge amounts of energy. Power limitation is a major issue hindering the operation of IoT applications and services. To improve operational visibility, Low-power devices which constitute IoT networks, drive the need for sustainable sources of energy to carry out their tasks for a prolonged period of time. Moreover, the means to ensure energy sustainability and QoS must consider the stochastic nature of the energy supplies and dynamic IoT environments. Artificial Intelligence (AI) enhanced protocols and algorithms are capable of predicting and forecasting demand as well as providing leverage at different stages of energy use to supply. AI will improve the efficiency of energy infrastructure and decrease waste in distributed energy systems, ensuring their long-term viability. In this paper, we conduct a survey to explore enhanced AI-based solutions to achieve energy sustainability in IoT applications. AI is relevant through the integration of various Machine Learning (ML) and Swarm Intelligence (SI) techniques in the design of existing protocols. ML mechanisms used in the literature include variously supervised and unsupervised learning methods as well as reinforcement learning (RL) solutions. The survey constitutes a complete guideline for readers who wish to get acquainted with recent development and research advances in AI-based energy sustainability in IoT Networks. The survey also explores the different open issues and challenges.  相似文献   

Multi-task learning based high-value patent and standard-essential patent identification model     

《Information processing & management》2023,60(3):103327

The high-value patent identification (HVPI) and the standard-essential patent identification (SEPI) are two important issues in the fields of intellectual property and the standardization, respectively. Almost all the HVPI and the SEPI are based on the single-task learning. In this paper, we unify the HVPI and the SEPI in a multi-task learning framework in consideration of the mutual reinforcement of the two tasks. In our model, we extract the patent structured features and embed the patent textual features using the pre-training model. Given these features, we explore a multi-task learning based identification model to identify the high-value patents and the standard-essential patents. We evaluate our model by comparing with two state-of-the-art models on the 5 balanced datasets and 2 imbalanced datasets. The results show our multi-task learning based model outperforms significantly these single-tasking learning based models in the measurements: precision, recall, F1 and accuracy. On the balanced datasets, the average increments of measurements are 1.3%, 1.29%, 1.28% and 1.28% respectively. On the imbalanced datasets, the average increments of measurements are 2.24%, 1.62%, 1.75% and 0.66% respectively.  相似文献   

Exploring an intervention’s impact on everyday school activities     

Marianne Björn  Idor Svensson 《Support for Learning》2021,36(1):5-19

The focus of this paper is on a group of pupils with reading and writing difficulties who have been participating in an intervention study using assistive technology. That intervention study contained supervised training sessions with reading and writing tasks using an iPad with special supportive applications. The current study is a qualitative investigation of whether there has been any transfer from the intervention, to the pupils’ everyday school activities. Interviews with pupils and their teachers and observations during classroom lectures have been used to collect data. The results show that the pupils were positive to the assistive technology (the applications on the iPads), they found the apps easy to learn how to use and they appreciated the benefits they could give. Even so, only a few of the pupils had found use for and continued to use the tools after the intervention period finished. Possible reasons are that when the novelty wore off, students reverted to their usual study habits and that older students with many teachers and different classrooms were less able to adapt to using the apps. To improve transfer, it is suggested to introduce assistive technology earlier to students, in the younger grades, before study habits have been formed and to inform teachers about the use of AT in the classroom, including what is available and how it can benefit students.  相似文献   

Assisting,Instructing, Assessing: 21st Century Student Centered Librarianship     

Lauren Newton  Daniel E. Feinberg 《The Reference Librarian》2020,61(1):25-41

ABSTRACT

For librarians at the University of North Florida, there was a need to move beyond information literacy instruction to one-on-one and small group research consultations to aid in student success. By staffing the research desk with staff and students, librarians were able to open their calendars to allow more time for in-person, phone, and online consultations to aid in meeting the research goals of students at the institution. After assessing the research consultation program for two years, there has been a positive correlation between research consultation usage, satisfaction in completing assignments, and student success measures throughout the university.  相似文献   

2019年篮联世界杯广州赛区小组赛对中国队的启示     

崔曼峰  赵毅砥 《广州体育学院学报》2020,(1):82-85

2019年国际篮联篮球世界杯在中国圆满举行,广州作为分赛场,承接了西班牙、波多黎各、突尼斯以及伊朗四支代表队的小组比赛6场。通过对广州赛区四支队伍在小组赛中的表现技术进行分析,并对比中国队在本次赛事中的表现,发现中国男篮在备赛与参赛阶段,存在大赛目标不清晰,心理训练与细节训练不足,教练团队执教水平不稳定,篮球基本功不扎实等一系列问题,反映出我国篮球的发展正处于困境。系统地分析广州小组赛中四支队伍的赛况,并与中国队做出对比,有针对性地提出建议,为中国篮球的发展提供参考。  相似文献   

可宥性事由：体育竞赛“越轨”行为出罪的又一考量因素     

刘峰江 《河北体育学院学报》2020,34(6):9-16

可宥性事由源于英美刑事诉讼中的可抗辩事由，与正当化事由联袂构成行为出罪理论的基石。与正当化事由不同，可宥性事由已经僭越了法定的行为正当化条件，行为的违法性首先被肯定，只是在责任归属上寻求宽恕性处理。引入可宥性事由，可丰富体育竞赛行为入罪和出罪的相关内容，便于司法统一判定。分析认为，体育竞赛行为可宥性事由成立，要满足体育赛事本身具有合法性和正规性、主体仅针对运动员、行为局限于体育竞赛行为、违法目的“单纯”性、行为以“反规则”为前提等条件。入罪和出罪的类型化有利于定罪和量刑的规范化，分别对竞技伤害行为、滥用兴奋剂行为、假球等竞赛舞弊行为适用可宥性事由予以出罪抑或减轻处罚的情形进行具体分析。最后，指出体育竞赛运动员可宥性事由行为入罪不能忽视运动员自由保障的理念，且尽量避免选择性刑事司法。  相似文献   

基于负样本多通道优化SSD网络的钢铁厂提取     

卢凯旋  李国清  陈正超  昝露洋  李柏鹏  高建威 《中国科学院大学学报》2020,37(3):352-359

准确提取钢铁厂对去产能监测和环境保护具有重要意义。传统的人工目视解译方法效率低、成本高，无法满足开展大区域钢铁厂监测的需求。以深度学习目标检测网络SSD为基础，构建面向遥感影像钢铁厂提取的深度学习目标检测网络，提出maxout模块，将负样本通路优化为多分支结构，突出难分负样本特征并提升网络对无用特征的抵制效果。利用国产GF-1数据对京津冀地区的钢铁厂进行快速自动提取实验。与人工解译的钢铁厂点位数据的对比表明，该目标检测方法的提取精度达到80%以上。  相似文献   

动态雾计算网络中基于在线学习的任务卸载算法     

谭友钰  陈蕾  周明拓  王昆仑  杨旸  张武雄 《中国科学院大学学报》2020,37(5):688-698

任务卸载是雾计算的主要技术之一，即计算能力不足的节点将任务卸载给具有富余资源的节点帮助计算。以优化任务平均卸载时延和提升卸载服务成功率为目标，利用多臂老虎机理论为动态雾计算网络提出一种基于在线学习的任务卸载算法，可实时做出最优卸载决策。将该算法扩展到非稳定网络状态，使之可以动态追踪网络中节点的资源与环境变化，实时调整卸载决策。详细分析所提出算法的性能、复杂度和存储占用情况。仿真结果表明，这两种算法可达到的长期平均任务卸载时延均十分接近理想算法下的最优时延，卸载服务成功率也得到显著提升。此外，所提算法在非稳定的网络状态下能够追踪到计算资源与环境的变化。  相似文献   

我国大学生高水平排球队空间分布特征及影响因素研究*     

马茂森    李毅钧  张〓兵  杨玲 《广州体育学院学报》2020,(5):78-84

利用核密度分析、巴雷托截取法和多元回归分析等方法，对我国大学生高水平排球队空间分布特征和影响因素进行分析。主要结论有：大学生高水平排球队的数量呈波动上升趋势；在规模结构上，大学生高水平排球队数量规模的省际分布呈现倒“金字塔”结构，而配比规模的省际分布呈现“金字塔”结构，且极不平衡；在空间格局上，具有显著的区域差异，表现出东部区域较多；在集聚特征上，大学生高水平排球队主要分布在直辖市和省会城市，主要集聚在京津翼地区和长江流域，其他地区比较分散；在空间竞技实力上，具有显著的区域不平衡性，呈现“南弱北强”的特征；影响大学生高水平排球队空间分布的主要因素有高校招生条件、本科院校数量、经济发展实力和体育发展水平。  相似文献