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   

Directed collaboration patterns in funded teams: A perspective of knowledge flow     

《Information processing & management》2023,60(2):103237

Collaborations in funded teams are essential for understanding funded research and funding policies, although of high interest, are still not fully understood. This study aims to investigate directed collaboration patterns from the perspective of the knowledge flow, which is measured based on the academic age. To this end, we proposed a project-based team identification approach, which gives particular attention to funded teams. The method is applicable to other funding systems. Based on identified scientific teams, we detected recurring and significant subgraph patterns, known as network motifs, and under-represented patterns, known as anti-motifs. We found commonly occurred motifs and anti-motifs are remarkably characterized by different structures matching certain functions in knowledge exchanges. Collaboration patterns represented by motifs favor hierarchical structures, supporting intensive interactions across academic generations. Anti-motifs are more likely to show chain-like structures, hindering potentially various knowledge activities, and are thus seldom found in real collaboration networks. These findings provide new insights into the understanding of funded collaborations and also the funding system. Meanwhile, our findings are helpful for researchers, the public and policymakers to gain knowledge on research(ers) evolution, particularly in terms of primordial collaboration patterns.  相似文献   

共享经济可持续性评价：社会感知与客观绩效     

董晓松  霍依凡  赵星 《科研管理》2023,44(1):79-88

共享经济行业可持续性评价的社会感知和客观绩效不一致将造成认知错位、投资误判和政策失效等结果。本文在线挖掘2017—2019年我国主要社交媒体上29 771条文本信息，并收集行业统计数据，综合测度共享经济行业的可持续性并比较其差异。研究发现：社会公众并不认为共享经济行业具有显著的总体可持续性，而且行业实际绩效展现的可持续性与社会大众认知截然不同。进一步发现：受限于主观认知能力，社会公众认为共享经济可持续性主要来自可观察到的行业外部经济和外部生态等维度，但客观绩效表明共享经济行业内部的经济和外部的技术优势才是其可持续性的重要支撑。此外，二者评价的差异在住宿与医疗等产能共享内部过程不易被大众观察的行业中最为突出。  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   

A novel personality detection method based on high-dimensional psycholinguistic features and improved distributed Gray Wolf Optimizer for feature selection     

《Information processing & management》2023,60(2):103217

Existing personality detection methods based on user-generated text have two major limitations. First, they rely too much on pre-trained language models to ignore the sentiment information in psycholinguistic features. Secondly, they have no consensus on the psycholinguistic feature selection, resulting in the insufficient analysis of sentiment information. To tackle these issues, we propose a novel personality detection method based on high-dimensional psycholinguistic features and improved distributed Gray Wolf Optimizer (GWO) for feature selection (IDGWOFS). Specifically, we introduced the Gaussian Chaos Map-based initialization and neighbor search strategy into the original GWO to improve the performance of feature selection. To eliminate the bias generated when using mutual information to select features, we adopt symmetric uncertainty (SU) instead of mutual information as the evaluation for correlation and redundancy to construct the fitness function, which can balance the correlation between features–labels and the redundancy between features–features. Finally, we improve the common Spark-based parallelization design of GWO by parallelizing only the fitness computation steps to improve the efficiency of IDGWOFS. The experiments indicate that our proposed method obtains average accuracy improvements of 3.81% and 2.19%, and average F1 improvements of 5.17% and 5.8% on Essays and Kaggle MBTI dataset, respectively. Furthermore, IDGWOFS has good convergence and scalability.  相似文献   

Joint work is not shifted proximally after a long run in rearfoot strike runners     

Jake A. Melaro  Allison H. Gruber 《Journal of sports sciences》2021,39(1):78-83

ABSTRACT

Distal-to-proximal redistribution of joint work occurs following exhaustive running in recreational but not competitive runners but the influence of a submaximal run on joint work is unknown. The purpose of this study was to assess if a long submaximal run produces a distal-to-proximal redistribution of positive joint work in well-trained runners. Thirteen rearfoot striking male runners (weekly distance: 72.6 ± 21.2 km) completed five running trials while three-dimensional kinematic and ground reaction force data were collected before and after a long submaximal treadmill run (19 ± 6 km). Joint kinetics were calculated from these data and percent contributions of joint work relative to total lower limb joint work were computed. Moderate reductions in absolute negative ankle work (p = 0.045, Cohen’s d = 0.31), peak plantarflexor torque (p = 0.004, d = 0.34) and, peak negative ankle power (p = 0.005, d = 0.32) were observed following the long run. Positive ankle, knee and hip joint work were unchanged (p < 0.05) following the long run. These findings suggest no proximal shift in positive joint work in well-trained runners after a prolonged run. Runner population, running pace, distance, and relative intensity should be considered when examining changes in joint work following prolonged running.  相似文献   

Continuous walking and time- and intensity-matched interval walking: Cardiometabolic demand and post-exercise enjoyment in insufficiently active,healthy adults     

Kate Francis  Tom Williamson  Paul Kelly 《Journal of sports sciences》2021,39(1):23-30

ABSTRACT

We compared cardiometabolic demand and post-exercise enjoyment between continuous walking (CW) and time- and intensity-matched interval walking (IW) in insufficiently active adults. Sixteen individuals (13 females and three males, age 25.3 ± 11.1 years) completed one CW and one IW session lasting 30 min in a randomised-counterbalanced design. For CW, participants walked at a mean intensity of 65–70% predicted maximum heart rate (HR_max). For IW, participants alternated between 3 min at 80% HR_max and 2 min at 50% HR_max. Expired gas was measured throughout each protocol. Participants rated post-exercise enjoyment following each protocol. Mean HR and V˙O₂ showed small positive differences in IW vs. CW (2, 95%CL 0, 4 beat.min^?1; d = 0.23, 95%CL 0.06, 0.41 and 1.4, 95%CL 1.2 ml.kg^?1.min^?1, d = 0.36, 95%CL 0.05, 0.65, respectively). There was a medium positive difference in overall kcal expenditure in IW vs. CW (25, 95%CL 7 kcal, d = 0.58, 95%CL 0.33, 0.82). Post-exercise enjoyment was moderately greater following IW vs. CW (9.1, 95%CL 1.4, 16.8 AU, d = 0.62, 95%CL 0.06, 0.90), with 75% of participants reporting IW as more enjoyable. Interval walking elicits meaningfully greater energy expenditure and is more enjoyable than CW in insufficiently active, healthy adults.  相似文献   

The zone of proximal creativity: What dynamic assessment of divergent thinking reveals about students’ latent class membership     

《Contemporary educational psychology》2021

Divergent Thinking is a domain-general mental attribute closely associated with creativity that can be quantified through the use of text-mining algorithms. Past research has shown that students’ Divergent Thinking is malleable in response to relatively simple contextual prompts. In addition, there is substantial variance in the degree to which individual students’ Divergent Thinking is malleable, suggesting the presence of a student-specific zone-of-proximal-development in relation to creativity. Here, we adopted a dynamic assessment paradigm that included multiple conditions under which student Divergent Thinking was measured and fit a latent profile analysis model to that dynamic assessment data. We found that, although on average the Originality of student responses can be augmented through a prompt to generate surprising or unusual ideas, three latent classes emerged that differed significantly on their patterns of augmentation. These three latent classes were termed: (a) Conventional Thinkers (7.80% of the sample), whose response to the Divergent Thinking task were highly constrained and unoriginal across all conditions (b) Prompted Shifters (66.56%), whose Originality significantly increased across conditions, and (c) Idea Generators (25.64%), whose responses were highly original across all conditions. These latent profiles were validated in regard to personality characteristics and domain-specific creative activities, with Idea Generators reporting significantly more Openness and Intellect, less Industriousness, and more creative activities across the domains of Literature, Music, Sports, Visual Art, Science, and Cooking than did the other latent classes.  相似文献   

An analysis of the reach and effectiveness of distance learning in India during school closures due to COVID-19     

《International Journal of Educational Development》2021

In India, more than 276 million children and youth were out of school for extended periods since March 2020 due to school closures in response to COVID-19. A key challenge has been how to measure the impact of responses to continuity of learning both to ensure more effective responses in the event of further disruptions, but also to help the education community conceptualize more creative and effective approaches to learning, through blended and flexible approaches. This study reflects on the findings from a UNICEF survey targeting parents and adolescents across 6 states in India, and identifies lessons learned for addressing learning inequities during future school closures. We focus on measuring three key variables – access to technology, their utilization, and perceived learning for different profiles of children. As students began learning from home, technology access rates in households were initially used to determine the estimated maximum reach of different distance learning modalities during school closures. Beyond access, we find significant variations in adolescents’ use of technology for learning purposes and their perceptions of learning, linked to the type of remote learning modality, gender, location and type of school. We discuss the implications for government strategies and policies to ensure better utilization of technologies which are available in households and to address equity gaps in learning opportunities.  相似文献