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   

全渠道零售场景下配送车辆路径问题     

施天娇  杨斌  朱小林 《上海海事大学学报》2020,41(2):51-57

为研究全渠道零售的配送系统对企业物流成本的影响,综合考虑时间窗、车型选择等因素,以运输成本最低为目标,建立高度协同的全渠道零售场景下的物流配送模型。该模型采用自适应遗传算法求解,利用MATLAB进行算例分析。结果显示:全渠道零售模式下的车辆调度方案能迅速响应消费者需求,与一般配送模式相比可以减少物流成本,优化车辆配置。与粒子群优化算法相比,自适应遗传算法的运行时间更短、优化结果更好,这验证了模型和算法的可行性和有效性。  相似文献   

线状需求下两阶段设施选址问题     

王紫萌  周建勤 《上海海事大学学报》2020,41(3):60-65

鉴于大型交通线路建设项目的物资需求随时间变化(在初期阶段需求较低;在后续阶段项目全面展开,需求显著上升),本文采用线积分对沿线连续分布的线状物资需求进行刻画,构建两阶段设施选址模型,第一阶段进行部分设施的选址,第二阶段对剩余设施选址,以实现系统总成本最低的目标。针对模型特点,设计基于Voronoi图的两阶段交替定位-分配(alternative location-allocation,ALA)算法进行求解,并进行实例分析,比较两阶段模型与传统模型的异同。研究表明,将两阶段设施选址模型应用于交通线路建设项目多物流节点选址,能节约运营成本,并缓解初期资金压力。  相似文献   

我国体育彩票销售额的空间异质性及影响因素分析          下载免费PDF全文

王成龙 《河北体育学院学报》2019,33(6):41-48

收集2013—2017年我国31个省份体育彩票销售额的面板数据和2017年各省份社会发展指标的相关数据，运用探索性空间数据分析方法（ESDA），对31个省份体育彩票销售额的空间异质性特征及影响因素进行研究。全局空间自相关分析结果表明，体育彩票销售额具有明显的空间聚集性，且呈显著的空间正相关性。经分析，发现31个省份体育彩票销售额不均衡，东部沿海省份的销售额高于西部省份，提出应重点关注西部地区的体育彩票事业发展。31个省份体育彩票销售呈现出以次冷区域为主，个别热点区域和次热区域相结合的特点，说明我国体育彩票销售的区域性寡占市场特征较明显，提出应加强省份间的沟通交流，促进各区域联动均衡发展。在影响因素方面，人均GDP、体育场地面积对体育彩票销售额具有显著性影响，其中人均GDP对南方的影响大于北方，体育场地面积对东北地区的影响大于西南地区，提出应进一步加大对体育场地建设的投入力度，以促进我国体育彩票事业的发展。  相似文献   

Effect of different loading conditions on running mechanics at different velocities     

Germán Carretero-Navarro  Doménico Cherubini  Wolfgang Taube 《European Journal of Sport Science》2019,19(5):595-602

Weighted vests are widely used to improve running economy and performance. However, it is not well-studied how running mechanics are adapted to counteract the higher peak vertical ground reaction forces (Fpeak) while running with such a device. Therefore, the present study aimed to investigate the effects of different loading conditions on running mechanics at different velocities. Thirteen subjects participated in two separate sessions one week apart. In the first session, maximal aerobic speed (MAS) was determined through a maximal incremental running test while in the second session, they were instructed to run during one minute under different loading (0%, +10% and +20% of body mass [BM]) and velocity (60%, 80% and 100% of MAS) conditions in a random order. Spatiotemporal data were recorded and then running mechanics modelled using the spring-mass model. The main results indicated that vertical and leg stiffness (Kvert and Kleg, respectively) were increased (P?P?>?.05) when load was changed. At the same time, alterations of the running kinematics were observed such as longer contact times, reduced flight times, stride frequencies and step lengths, as well as an increase of the centre of mass dynamics. Based on these results it is assumed that runners maintain a certain stiffness level for each velocity despite different loading conditions. As a consequence, Fpeak increases and this probably causes spatiotemporal adjustments in the movement kinematics.  相似文献   

基于新鲜度和配送成本的易腐食品配送路径   总被引：1，自引：0，他引：1  

李畅  陈淮莉 《上海海事大学学报》2019,40(1):14-20

为解决企业在制定易腐食品配送计划时难以权衡配送总成本与交付产品的新鲜度的问题,建立以配送总成本最低和交付产品平均新鲜度最大为目标,带时间窗的易腐食品配送路径规划模型。采用自适应差分进化(differential evolution,DE)算法求解模型,通过数值算例验证模型和算法的有效性。与基本DE算法和基本蚁群算法的求解结果进行对比,自适应DE算法的求解结果更优,收敛速度更快。求得的帕累托解集表明配送总成本与交付产品平均新鲜度相悖,增加少量的配送成本可以使交付产品的平均新鲜度得到大幅提升。对易腐食品保质期、时间窗宽度和车辆装载量进行灵敏度分析,为企业在不同配送情景下在配送总成本与交付产品的平均新鲜度之间的权衡提供参考。  相似文献   

基于DE-SVM的船舶航迹预测模型     

刘娇  史国友  杨学钱  朱凯歌 《上海海事大学学报》2020,41(1):34-39

为提高船舶航迹预测精度,解决准确建模难度大和神经网络易陷入局部最优的问题,考虑实时获取目标船AIS数据较少的特点,提出一种基于支持向量机(support vector machine,SVM)的航迹预测模型。选择AIS数据中的航速、航向和船舶经纬度作为样本特征变量;采用小波阈值去噪的方法处理训练数据;采用差分进化(differential evolution,DE)算法对模型内部参数寻优以提高模型收敛速度和预测精度。选取天津港实船某段航迹的AIS数据,比较基于DE-SVM与基于BP神经网络的航迹预测模型的仿真结果。结果表明,基于DE-SVM的航迹预测模型具有更高的预测精度,简单、可行、高效,且耗时少。  相似文献   

基于NSGAⅡ的带时间窗生鲜品配送路径优化     

李善俊  陈淮莉 《上海海事大学学报》2020,41(2):58-64

考虑到生鲜品的易腐性、高时效性和对运输环境要求高的特点,为减少生鲜品配送企业的配送成本,提高客户满意度,在车辆载重、时间窗、生鲜品保质期等约束条件下,提出将配送总成本最小化和生鲜品新鲜度最大化的多目标车辆路径优化模型。利用改进的非支配排序遗传算法(non-dominated sorting genetic algorithmⅡ,NSGAⅡ)对模型进行求解。求出满足配送总成本最小和生鲜品新鲜度最大的相对较优解,证明模型的有效性。该模型在大规模配送和客户地理位置分布较为分散的情况下,对生鲜品新鲜度的优化较为明显。  相似文献   

动态接力点模式下双自动化轨道吊协调调度     

邱亚  梁承姬  张悦 《上海海事大学学报》2020,41(3):53-59

为提高自动化集装箱码头堆场中自动化轨道吊(automated stacking crane,ASC)的作业效率,研究箱区动态接力点对双ASC作业效率的影响。考虑双ASC作业过程中的安全距离、相互冲突等因素,建立以最小化双ASC最长完工时间为目标的混合整数规划模型,利用遗传算法对该模型进行求解。与固定贝位接力模式进行对比,结果表明,相较于固定贝位接力模式,动态接力点模式下双ASC的作业效率更高。在不同规模算例背景下,将遗传算法与CPLEX的计算结果进行对比,验证了遗传算法的有效性。  相似文献   

基于短时交通流量预测对路网微观交通仿真的研究     

张虎 《苏州市职业大学学报》2021,(1)

针对早高峰短时交通流量预测数据少、波动大的特点,提出用灰色模型进行预测。将灰色GM(2,1)改进为灰色GM(2,1,λ,ρ)预测模型,以提高预测精度。针对粒子群算法(PSO)的早熟现象,将Logistic混沌搜索嵌入到PSO算法,应用混沌粒子群算法(CPSO)寻找灰色GM(2,1,λ,ρ)预测模型最优的参数λ和ρ。结合两者提出了基于CPSO–GM(2,1,λ,ρ)的早高峰短时交通流预测模型。利用VISSIM对研究路网进行微观交通仿真,通过VISSIM–Excel、VBA–Matlab平台实现了短时交通流量预测和路网微观交通仿真数据的交互,对集成交通控制系统的架构进行了方案设计。仿真结果表明,结合流量预测的路网优于固定信号配时下的路网仿真。  相似文献