Multiple Empirical Kernel Learning with Majority Projection for imbalanced problems

Traditional Multiple Empirical Kernel Learning (MEKL) expands the expressions of the sample and brings better classification ability by using different empirical kernels to map the original data space into multiple kernel spaces. To make MEKL suit for the imbalanced problems, this paper introduces a weight matrix and a regularization term into MEKL. The weight matrix assigns high misclassification cost to the minority samples to balanced misclassification cost between minority and majority class. The regularization term named Majority Projection (MP) is used to make the classification hyperplane fit the distribution shape of majority samples and enlarge the between-class distance of minority and majority class. The contributions of this work are: (i) assigning high cost to minority samples to deal with imbalanced problems, (ii) introducing a new regularization term to concern the property of data distribution, (iii) and modifying the original PAC-Bayes bound to test the error upper bound of MEKL-MP. Through analyzing the experimental results, the proposed MEKL-MP is well suited to the imbalanced problems and has lower generalization risk in accordance with the value of PAC-Bayes bound.     

Common due date assignment scheduling for a no-wait flowshop with convex resource allocation and learning effect

This article addresses the no-wait flowshop scheduling problem with simultaneous consideration of common due date assignment, convex resource allocation and learning effect in a two machine setting. The processing time of each job can be controlled by its position in a sequence and also by allocating extra resource, which is a convex function of the amount of a common continuously divisible resource allocated to the job. The objective is to determine the optimal common due date, the resource allocation and the schedule of jobs such that the total earliness, tardiness and common due date cost (the total resource consumption cost) are minimized under the constraint condition that the total resource consumption cost (the total earliness, tardiness and common due date cost) is limited. Polynomial time algorithms are developed for two versions of the problem.     

面向不平衡数据流的自适应加权在线超限学习机算法

一般的在线学习算法对不平衡数据流的分类识别会遇到较大困难,特别是当数据流发生概念漂移时,对其进行分类会变得更困难.文中提出面向不平衡数据流的自适应加权在线超限学习机算法,自动调整实时到达的训练样本的惩罚参数,达到在线学习不平衡数据流的目的.文中算法可以适用于不同偏斜程度的静态数据流的在线学习和发生概念漂移时数据流的在线学习.理论分析和在多个真实数据流上的实验表明文中算法的正确性和有效性.     

基于PU学习的链接预测方法

基于分类的链接预测方法中,由于链接未知节点对的大规模性与不确定性,选择可靠负例成为构造链接预测分类器的难点问题.为此,文中提出基于正例和无标识样本(PU)学习的链接预测方法.首先,提取节点对的拓扑信息以构造样本集.再利用社区结构确定候选负例的分布,基于分布进行多次欠采样,获得多个候选负例子集,集成多个负例集与正例集中构建的分类器选择可靠负例.最后基于正例与可靠负例构造链接预测分类器.在4个网络数据集上的实验表明文中方法预测结果较优.     

基于多尺度高斯核的分布式正则化回归学习算法

针对工业、信息等领域出现的基于较大规模、非平稳变化复杂数据的回归问题,已有算法在计算成本及拟合效果方面无法同时满足要求.因此,文中提出基于多尺度高斯核的分布式正则化回归学习算法.算法中的假设空间为多个具有不同尺度的高斯核生成的再生核Hilbert空间的和空间.考虑到整个数据集划分的不同互斥子集波动程度不同,建立不同组合系数核函数逼近模型.利用最小二乘正则化方法同时独立求解各逼近模型.最后,通过对所得的各个局部估计子加权合成得到整体逼近模型.在2个模拟数据集和4个真实数据集上的实验表明,文中算法既能保证较优的拟合性能,又能降低运行时间.     

基于结构化遮挡编码和极限学习机的局部遮挡人脸识别

提出使用结构化遮挡编码（SOC）结合极限学习机（ELM）的算法来处理人脸识别中的遮挡问题。首先，使用SOC去除图像上的遮挡物，将遮挡物体与人脸分离开；同时，通过局部性约束字典（LCD）来估计遮挡物的位置，建立遮挡字典和人脸字典。然后，将建立好的人脸字典矩阵进行归一化处理，并利用ELM对归一化的数据进行分类识别。最后，在AR人脸库上进行的仿真实验结果表明，所提方法对不同遮挡物和不同区域遮挡的图像具有较好的识别率和鲁棒性。     

基于ELM的英语教学质量评价研究

对英语教学质量实施科学化的评价关系到英语教学质量的提升,为减少主观因素的影响,本文提出一种ELM的教学质量评价方法,该种方法将可以在很大程度上提高教学质量评价的精度。对影响英语教学质量的12个指标进行打分,以此作为ELM的输入,同时将英语教学质量的综合得分作为ELM的输出。评价结果表明,基于EIM的英语教学质量评价结果比SVM、BP具有了很大程度的提升,为实施英语教学质量的评价提供了参考。     

MOOC study group: Facilitation strategies,influential factors,and student perceived gains

“Join a Meetup Group” (face-to-face study group) has been propagated by Coursera to build rapport and provide mutual support among MOOC learners; however, studies remain scant regarding its effectiveness and sustainability. This interpretive case study documents our facilitation process, key influential factors, as well as student perceived gains in a six-week MOOC study group. Data sources include discussion recordings, end-of-course interviews, goal setting sheets, weekly reflection journals, and researchers' observation notes. Results showed that, cognitively, participants broadened their perspective of thinking, raised cultural awareness, and shared many learning strategies. Affectively, they established a strong sense of community and gained motivation for learning. Participants also increased action tendencies toward trying out Coursera functions, new courses, and learning strategies, and they became more cognizant of the benefits and procedures of the MOOC study group. Our findings suggest that, with proper design and facilitation, face-to-face study group would be a practicable and effective approach to leverage MOOC students' motivation, engagement, and deeper learning. Implications are discussed in terms of potential gains, challenges, key influential factors, as well as future design and implementation of MOOC study groups.     

Learning fuzzy controllers in mobile robotics with embedded preprocessing

The automatic design of controllers for mobile robots usually requires two stages. In the first stage, sensorial data are preprocessed or transformed into high level and meaningful values of variables which are usually defined from expert knowledge. In the second stage, a machine learning technique is applied to obtain a controller that maps these high level variables to the control commands that are actually sent to the robot. This paper describes an algorithm that is able to embed the preprocessing stage into the learning stage in order to get controllers directly starting from sensorial raw data with no expert knowledge involved. Due to the high dimensionality of the sensorial data, this approach uses Quantified Fuzzy Rules (QFRs), that are able to transform low-level input variables into high-level input variables, reducing the dimensionality through summarization. The proposed learning algorithm, called Iterative Quantified Fuzzy Rule Learning (IQFRL), is based on genetic programming. IQFRL is able to learn rules with different structures, and can manage linguistic variables with multiple granularities. The algorithm has been tested with the implementation of the wall-following behavior both in several realistic simulated environments with different complexity and on a Pioneer 3-AT robot in two real environments. Results have been compared with several well-known learning algorithms combined with different data preprocessing techniques, showing that IQFRL exhibits a better and statistically significant performance. Moreover, three real world applications for which IQFRL plays a central role are also presented: path and object tracking with static and moving obstacles avoidance.     

基于特征选择和深度信念网络的文本情感分类算法

由于人类语言的复杂性，文本情感分类算法大多都存在因为冗余而造成的词汇量过大的问题。深度信念网络（DBN）通过学习输入语料中的有用信息以及它的几个隐藏层来解决这个问题。然而对于大型应用程序来说，DBN是一个耗时且计算代价昂贵的算法。针对这个问题，提出了一种半监督的情感分类算法，即基于特征选择和深度信念网络的文本情感分类算法（FSDBN）。首先使用特征选择方法（文档频率（DF）、信息增益（IG）、卡方统计（CHI）、互信息（MI））过滤掉一些不相关的特征从而使词汇表的复杂性降低；然后将特征选择的结果输入到DBN中，使得DBN的学习阶段更加高效。将所提算法应用到中文以及维吾尔语中，实验结果表明在酒店评论数据集上，FSDBN在准确率方面比DBN提高了1.6%，在训练时间上比DBN缩短一半。