A new e-learning achievement evaluation model based on RBF-NN and similarity filter

As Internet rises fast in recent decades, teaching and learning tools based on Internet technology are rapidly applied in education. Learning through Internet can make learners absorb knowledge without the limitations on learning time and distance. Therefore, in academy, e-learning is one of the popular learning assistant instruments. Recently, “student-centered” instruction has become one of the primary approaches in education, and the e-learning system, which can provide the learning environment of personalization and adaptability, is more and more popular. By using e-learning system, teachers can adjust the learning schedule instantly for learners according to their learning achievements, and build more adaptive learning environments. However, in some cases, bias assessments are given for student achievements under specific uncontrollable conditions (i.e. tiredness, preference). In dire need of overcoming this predicament, a new model based on radial basis function neural networks (RBF-NN) and similarity filter to evaluate learning achievements is proposed. The proposed model includes three phases to reduce bias assessments: (1) preprocess: select important features (attributes) to enhance classification performance by feature selection methods and utilize minimal entropy principle approach (MEPA) to fuzzify the quantitative data, (2) similarity filter: select linguistic values for each feature and delete inconsistent data by the similarity threshold (similarity filter) and (3) construct classification model and accuracy evaluation: build the proposed model based on RBF-NN and evaluate model performance. To verify the proposed model, a practical achievement dataset, collected from e-learning online examination system in a university of Taiwan, is used as experiment dataset, and the performance of the proposed model is compared with the listing models in this paper. From the empirical study, it is shown that the proposed model provided more proper achievement evaluations than the listing models.     

A hybrid trust model using reinforcement learning and fuzzy logic

Multiagent systems (MASs) are increasingly popular for modeling distributed environments that are highly complex and dynamic, such as e‐commerce, smart buildings, and smart grids. Typically, agents assumed to be goal driven with limited abilities, which restrains them to working with other agents for accomplishing complex tasks. Trust is considered significant in MASs to make interactions effectively, especially when agents cannot assure that potential partners share the same core beliefs about the system or make accurate statements regarding their competencies and abilities. Due to the imprecise and dynamic nature of trust in MASs, we propose a hybrid trust model that uses fuzzy logic and Q‐learning for trust modeling. as an improvement over Q‐learning‐based trust evaluation. Q‐learning is used to estimate trust on the long term, fuzzy inferences are used to aggregate different trust factors, and suspension is used as a short‐term response to dynamic changes. The performance of the proposed model is evaluated using simulation. Simulation results indicate that the proposed model can help agents select trustworthy partners to interact with. It has a better performance compared to some of the popular trust models in the presence of misbehaving interaction partners.     

FEATURE SELECTION BASED ON COMPACTNESS AND SEPARABILITY: COMPARISON WITH FILTER‐BASED METHODS

Selecting a subset of salient features for performing clustering using a clustering learning algorithm has been explored extensively in many real‐world applications. To select salient features during training, the filter model evaluates the intrinsic characteristics of each individual feature but is not permitted to use a clustering learning algorithm that provides clustered information to train the features. In particular, the filter model aims to predict unobservable clusters and measure how the features help provide satisfactory within‐cluster and between‐cluster scatters to achieve a good clustering quality. However, it is generally difficult to achieve both scatters in the filter model. For example, a random variable with a large variance may raise only the between‐cluster scatter, whereas another variable following a uniform distribution may raise only the within‐cluster scatter. In this paper, we present a new filter‐based method to quantify features that consider feature compactness and separability to ensure that both scatters are raised. Moreover, our method adopts a new search strategy to locate the best feature salience vector instead of visiting the space of all the possible feature subsets. After the benchmark data sets are tested, the experimental results indicate that our method performs better than many benchmark filter‐based methods at selecting a feature subset to perform clustering.     

Neighborhood based sample and feature selection for SVM classification learning

Support vector machines (SVMs) are a class of popular classification algorithms for their high generalization ability. However, it is time-consuming to train SVMs with a large set of learning samples. Improving learning efficiency is one of most important research tasks on SVMs. It is known that although there are many candidate training samples in some learning tasks, only the samples near decision boundary which are called support vectors have impact on the optimal classification hyper-planes. Finding these samples and training SVMs with them will greatly decrease training time and space complexity. Based on the observation, we introduce neighborhood based rough set model to search boundary samples. Using the model, we firstly divide sample spaces into three subsets: positive region, boundary and noise. Furthermore, we partition the input features into four subsets: strongly relevant features, weakly relevant and indispensable features, weakly relevant and superfluous features, and irrelevant features. Then we train SVMs only with the boundary samples in the relevant and indispensable feature subspaces, thus feature and sample selection is simultaneously conducted with the proposed model. A set of experimental results show the model can select very few features and samples for training; in the mean time the classification performances are preserved or even improved.     

Improving the effectiveness of English vocabulary review by integrating ARCS with mobile game‐based learning

Memorizing English vocabulary is often considered uninteresting, and a lack of motivation exists during learning activities. Moreover, most vocabulary practice systems automatically select words from articles and do not provide integrated model methods for students. Therefore, this study constructed a mobile game‐based English vocabulary practice system, which imitates the popular block elimination game and combines test items of article, difficulty, and teacher model in accordance with curriculum objectives and demands. The participants were first‐year students and classified into traditional group (Group T) and game‐based group (Group G), and learning effectiveness and motivation were analysed in this study. According to statistical analysis results, students who used the game‐based vocabulary practice system, which could captivate their attention and interest, had higher learning effectiveness and provided positive feedback of learning motivation. Furthermore, English teachers could select teaching materials that are consistent with the features, knowledge, and cultural background of learners to improve the relevance dimension of learning motivation.     

基于粗糙集和改进鲸鱼优化算法的特征选择方法

随着互联网和物联网技术的发展,数据的收集变得越发容易。但是,高维数据中包含了很多冗余和不相关的特征,直接使用会徒增模型的计算量,甚至会降低模型的表现性能,故很有必要对高维数据进行降维处理。特征选择可以通过减少特征维度来降低计算开销和去除冗余特征,以提高机器学习模型的性能,并保留了数据的原始特征,具有良好的可解释性。特征选择已经成为机器学习领域中重要的数据预处理步骤之一。粗糙集理论是一种可用于特征选择的有效方法,它可以通过去除冗余信息来保留原始特征的特性。然而,由于计算所有的特征子集组合的开销较大,传统的基于粗糙集的特征选择方法很难找到全局最优的特征子集。针对上述问题,文中提出了一种基于粗糙集和改进鲸鱼优化算法的特征选择方法。为避免鲸鱼算法陷入局部优化,文中提出了种群优化和扰动策略的改进鲸鱼算法。该算法首先随机初始化一系列特征子集,然后用基于粗糙集属性依赖度的目标函数来评价各子集的优劣,最后使用改进鲸鱼优化算法,通过不断迭代找到可接受的近似最优特征子集。在UCI数据集上的实验结果表明,当以支持向量机为评价所用的分类器时,文中提出的算法能找到具有较少信息损失的特征子集,且具有较高的分类精度。因此,所提算法在特征选择方面具有一定的优势。     

Fault diagnosis of rolling bearing based on optimized soft competitive learning Fuzzy ART and similarity evaluation technique

In this work, a new classification method called Soft Competitive Learning Fuzzy Adaptive Resonance Theory (SFART) is proposed to diagnose bearing faults. In order to solve the misclassification caused by the traditional Fuzzy ART based on hard competitive learning, a soft competitive learning ART model is established using Yu’s norm similarity criterion and lateral inhibition theory. The proposed SFART is based on Yu’s norm similarity criterion and soft competitive learning mechanism. In SFART, Yu’s similarity criterion and the lateral inhibition theory were employed to measure the proximity and select winning neurons, respectively. To further improve the classification accuracy, a feature selection technique based on Yu’s norms is also proposed. In addition, Particle Swarm Optimization (PSO) is introduced to optimize the model parameters of SFART. Meanwhile, the validity of the feature selection technique and parameter optimization method is demonstrated. Finally, fuzzy ART/ ARTMAP (FAM) as well as the feasibility of the proposed SFART algorithm are validated by comparing the diagnosis effectiveness of the proposed algorithm with the classic Fuzzy c-means (FCM), Fuzzy ART and fuzzy ARTMAP (FAM).     

Using Rough Sets with Heuristics for Feature Selection

Practical machine learning algorithms are known to degrade in performance (prediction accuracy) when faced with many features (sometimes attribute is used instead of feature) that are not necessary for rule discovery. To cope with this problem, many methods for selecting a subset of features have been proposed. Among such methods, the filter approach that selects a feature subset using a preprocessing step, and the wrapper approach that selects an optimal feature subset from the space of possible subsets of features using the induction algorithm itself as a part of the evaluation function, are two typical ones. Although the filter approach is a faster one, it has some blindness and the performance of induction is not considered. On the other hand, the optimal feature subsets can be obtained by using the wrapper approach, but it is not easy to use because of the complexity of time and space. In this paper, we propose an algorithm which is using rough set theory with greedy heuristics for feature selection. Selecting features is similar to the filter approach, but the evaluation criterion is related to the performance of induction. That is, we select the features that do not damage the performance of induction.     

基于特征选择与集成学习的钓鱼网站检测方法

针对目前大部分钓鱼网站检测方法存在检测准确率低、误判率高等问题,提出了一种基于特征选择与集成学习的钓鱼网站检测方法。该检测方法首先使用FSIGR算法进行特征选择,FSIGR算法结合过滤和封装模式的优点,从信息相关性和分类能力两个方面对特征进行综合度量,并采用前向递增后向递归剔除策略对特征进行选择,以分类精度作为评价指标对特征子集进行评价,从而获取最优特征子集;然后使用最优特征子集数据对随机森林分类算法模型进行训练。在UCI数据集上的实验表明,所提方法能够有效提高钓鱼网站检测的正确率,降低误判率,具有实际应用意义。     

基于粗概念格模型的电子商务领域本体的构建研究

以构建电子商务系统中的本体为出发点,分析现有的本体构建技术中存在的缺陷。针对这些不足,综合考虑变精度粗糙集模型和形式概念分析的相关理论,提出基于粗概念格模型来构建本体。将变精度粗糙集的β选取算法和可辨识矩阵属性约简算法进行了改进,使β 上、下分布的约简方法适用于形式背景的约简,从而提出基于变精度粗糙集的概念格约减算法;然后计算语义概念相似度,并以联合国标准产品与服务分类代码的本体元模型为核心本体,结合领域专家知识,建立电子商务领域本体模型。实验表明了粗概念格构建本体的高效性。     

基于BP神经网络在物联网多敏感信息同源检索

针对物联网中多用户相似源敏感信息检索方法运算速度慢、结果质量较差问题,提出一种物联网多敏感信息同源检索算法.准确分析物联网各应用层,构建隐层的3层BP神经网络模型,引入动态因子的学习算法,结合多维仿生信息理论的点同源连续性规则,推导出多敏感信息同源相似度特征,排序信息同源相似度特征的相关性级别并输出,最终完成物联网多敏...     

Assessing mathematics learning achievement using hybrid rough set classifiers and multiple regression analysis

Education is recognized as the key to individual success. Particularly, elementary education is vital for providing students with basic literacy and numeracy, as well as establishing foundations in mathematics, language, science, geography, history, and other social sciences. Mathematics is fundamental to numerous fields with real life applications, including natural science, engineering, medicine, and social sciences; therefore, student mathematics-learning achievement (MLA) in elementary school is valuable. This study aims to eliminate wastage of educational resources and seek suitable hybrid models for application to education. This study proposes an integrated hybrid model based on rough set classifiers and multiple regression analysis and provides a new trial of such a hybrid model to process MLA problems for elementary schools and their teachers. The proposed model is illustrated by examining a dataset from an elementary school in Taiwan. The experimental results reveal that the proposed model outperforms the listing methods in both classification accuracy and standard deviation. The rough set LEM2 (Learning from Examples Module, version 2) algorithm generates a set of comprehensible decision rules that can be applied in a knowledge-based education system designed for interested parties. Consequently, the analytical results have important implications for strategies related to mathematics teaching/learning and support to achieve teaching goals related to mathematics education.     

基于卷积特征深度融合的海上目标跟踪算法

针对海上复杂环境下深度学习方法跟踪速度慢和尺度变化问题,以及现有跟踪算法仅使用单层深度特征或手动融合多层特征的问题,提出一种基于卷积神经网络特征深度融合的多尺度相关滤波海上目标跟踪算法。以VGG-NET-16深度模型为基础,加入多层特征融合结构,实现深度卷积融合网络,用于特征提取,通过相关滤波算法构建定位滤波器,确定目标的中心位置,通过多尺度采样构建尺度滤波器,实现对目标的判断。实验结果表明,该算法可对海上移动目标实现多尺度的有效跟踪。     

Mining students activities from a computer supported collaborative learning system based on peer to peer network

As Information & Communication Technology (ICT) is rapidly evolved, educational paradigms have been changing. The ultimate goal of education with the aid of ICT is to provide customized training for learners to improve the effectiveness of their learning at anytime and anywhere. In the online learning environment where the Internet, mobile devices, peer-to-peer (P2P) and the cloud technology are leveraged, all the information in learning activities is converted into digital data and stored in the Computer Supported Collaborative Learning (CSCL) system. The data in the CSCL system contains various learners’ information including the learning objectives, learning preferences, competences and achievements. Thus, by analyzing the activity information of learners in an online CSCL system, meaningful and useful information can be extracted and provided for learners, teachers and administrators as feedback. In this paper, we propose a learner activity model that represents the learner’s activity information stored in a CSCL system. As for the proposed learner activity model, we classified the learning activities in a CSCL system into three categories: vivacity, learning and relationship; then we created quotients to represent them accordingly. In addition, we developed a CSCL System, which we termed as COLLA, applied the proposed learner activity model and analyzed the results.     

基于深度学习与特征融合的恶意网页识别方法研究

互联网环境的高度开放性和无序性导致了网络安全问题的普遍性和不可预知性, 网络安全问题已成为当前国际社会关注的热点问题。基于机器学习的恶意网页识别方法虽然卓有成就, 但随着对恶意网页识别需求的不断提高, 在识别效率上仍然表现出较大的局限性。本文提出一种基于深度学习与特征融合的识别方法, 将图卷积神经网络(Generalized connection network,GCN)与一维卷积神经网络(Convolution neural network, CNN)、支持向量机(Support vector machine, SVM)相结合。首先, 考虑到传统神经网络只适用于处理结构化数据以及无法很好的捕获单词间非连续和长距离依赖关系, 从而影响网页识别准确率的缺点,通过 GCN 丰富的关系结构有效捕获并保持网页文本的全局信息; 其次, CNN 可以弥补 GCN 在局部特征信息提取方面的不足,通过一维 CNN 对网页 URL(Uniform resource locator, URL)进行局部信息提取, 并进一步将捕获到的 URL 局部特征与网页文本全局特征进行融合, 从而选择出兼顾 CNN 模型和 GCN 模型特点的更具代表性的网页特征; 最终, 将融合后的特征输入到 SVM分类器中进行网页判别。本文首次将 GCN 应用于恶意网页识别领域, 通过组合模型有效兼顾了深度学习与机器学习的优点, 将深度学习网络模型作为特征提取器, 而将机器学习分类算法作为分类器, 通过实验证明, 测试准确率达到 92.5%, 高于已有的浅层的机器学习检测方法以及单一的神经网络模型。本文提出的方法具有更高的稳定性, 以及在精确率、召回率、 F1 值等多项检测指标上展现出更加优越的性能。     

基于改进模糊决策粗糙集的最小化决策代价特征选择算法

模糊决策粗糙集是决策粗糙集理论在模糊集环境下的重要延伸,然而该模型对含噪声的数据不具有很好的容忍性。为此在传统的模糊相似关系中引入一个限定阈值,提出一种改进的模糊相似关系。在其基础上对原始的模糊决策粗糙集进行重构,提出一种改进的模糊决策粗糙集模型。根据不同的特征选择方式,利用所提出的改进模型设计出两种搜索策略的最小化决策代价特征选择算法。实验分析表明,该算法比传统算法具有更高的优越性。     

Evolving Least Squares Support Vector Machines for Stock Market Trend Mining

In this paper, an evolving least squares support vector machine (LSSVM) learning paradigm with a mixed kernel is proposed to explore stock market trends. In the proposed learning paradigm, a genetic algorithm (GA), one of the most popular evolutionary algorithms (EAs), is first used to select input features for LSSVM learning, i.e., evolution of input features. Then, another GA is used for parameters optimization of LSSVM, i.e., evolution of algorithmic parameters. Finally, the evolving LSSVM learning paradigm with best feature subset, optimal parameters, and a mixed kernel is used to predict stock market movement direction in terms of historical data series. For illustration and evaluation purposes, three important stock indices, S&P 500 Index, Dow Jones Industrial Average (DJIA) Index, and New York Stock Exchange (NYSE) Index, are used as testing targets. Experimental results obtained reveal that the proposed evolving LSSVM can produce some forecasting models that are easier to be interpreted by using a small number of predictive features and are more efficient than other parameter optimization methods. Furthermore, the produced forecasting model can significantly outperform other forecasting models listed in this paper in terms of the hit ratio. These findings imply that the proposed evolving LSSVM learning paradigm can be used as a promising approach to stock market tendency exploration.     

CAT-RFE:点击欺诈的集成检测框架

点击欺诈是近年来最常见的网络犯罪手段之一,互联网广告行业每年都会因点击欺诈而遭受巨大损失。为了能够在海量点击中有效地检测欺诈点击,构建了多种充分结合广告点击与时间属性关系的特征,并提出了一种点击欺诈检测的集成学习框架——CAT-RFE集成学习框架。CAT-RFE集成学习框架包含3个部分:基分类器、递归特征消除(RFE,recursive feature elimination)和voting集成学习。其中,将适用于类别特征的梯度提升模型——CatBoost(categorical boosting)作为基分类器;RFE是基于贪心策略的特征选择方法,可在多组特征中选出较好的特征组合;Voting集成学习是采用投票的方式将多个基分类器的结果进行组合的学习方法。该框架通过CatBoost和RFE在特征空间中获取多组较优的特征组合,再在这些特征组合下的训练结果通过voting进行集成,获得集成的点击欺诈检测结果。该框架采用了相同的基分类器和集成学习方法,不仅克服了差异较大的分类器相互制约而导致集成结果不理想的问题,也克服了RFE在选择特征时容易陷入局部最优解的问题,具备更好的检测能力。在实际互联网点击欺诈数据集上的性能评估和对比实验结果显示,CAT-RFE集成学习框架的点击欺诈检测能力超过了CatBoost模型、CatBoost和RFE组合的模型以及其他机器学习模型,证明该框架具备良好的竞争力。该框架为互联网广告点击欺诈检测提供一种可行的解决方案。     

Road Safety Performance Function Analysis With Visual Feature Importance of Deep Neural Nets

Road safety performance function (SPF) analysis using data-driven and nonparametric methods, especially recent developed deep learning approaches, has gained increasing achievements. However, due to the learning mechanisms are hidden in a “black box” in deep learning, traffic features extraction and intelligent importance analysis are still unsolved and hard to generate. This paper focuses on this problem using a deciphered version of deep neural networks (DNN), one of the most popular deep learning models. This approach builds on visualization, feature importance and sensitivity analysis, can evaluate the contributions of input variables on model’s “black box” feature learning process and output decision. Firstly, a visual feature importance (ViFI) method that describes the importance of input features is proposed by adopting diagram and numerical-analysis. Secondly, by observing the change of weights using ViFI on unsupervised training and fine-tuning of DNN, the final contributions of input features are calculated according to importance equations for both steps that we proposed. Sequentially, a case study based on a road SPF analysis is demonstrated, using data collected from a major Canadian highway, Highway 401. The proposed method allows effective deciphering of the model’s inner workings and allows the significant features to be identified and the bad features to be eliminated. Finally, the revised dataset is used in crash modeling and vehicle collision prediction, and the testing result verifies that the deciphered and revised model achieves state-of-the-art performance.      

Online identification of nonlinear multivariable processes using self‐generating RBF neural networks

This paper addresses the problem of online model identification for multivariable processes with nonlinear and time‐varying dynamic characteristics. For this purpose, two online multivariable identification approaches with self‐organizing neural network model structures will be presented. The two adaptive radial basis function (RBF) neural networks are called as the growing and pruning radial basis function (GAP‐RBF) and minimal resource allocation network (MRAN). The resulting identification algorithms start without a predefined model structure and the dynamic model is generated autonomously using the sequential input‐output data pairs in real‐time applications. The extended Kalman filter (EKF) learning algorithm has been extended for both of the adaptive RBF‐based neural network approaches to estimate the free parameters of the identified multivariable model. The unscented Kalman filter (UKF) has been proposed as an alternative learning algorithm to enhance the accuracy and robustness of nonlinear multivariable processes in both the GAP‐RBF and MRAN based approaches. In addition, this paper intends to study comparatively the general applicability of the particle filter (PF)‐based approaches for the case of non‐Gaussian noisy environments. For this purpose, the Unscented Particle Filter (UPF) is employed to be used as alternative to the EKF and UKF for online parameter estimation of self‐generating RBF neural networks. The performance of the proposed online identification approaches is evaluated on a highly nonlinear time‐varying multivariable non‐isothermal continuous stirred tank reactor (CSTR) benchmark problem. Simulation results demonstrate the good performances of all identification approaches, especially the GAP‐RBF approach incorporated with the UKF and UPF learning algorithms. Copyright © 2010 John Wiley and Sons Asia Pte Ltd and Chinese Automatic Control Society