海面红外目标智能跟踪算法研究

根据海面运动目标及成像传感器的特点,本文提出了一种红外目标图像序列跟踪点自动识别的智能跟踪算法.以平均梯度为特征进行目标图像分割;并根据目标最亮区与吃水线的空间关系选择目标最亮区中心点的垂线与吃水线的交点作为跟踪点;利用D-S证据理论融合帧间关联特征的跟踪点的自动识别以及可信度确认.实验证明本方法合理而且十分有效.     

基于改进DS证据融合与ELM的入侵检测算法

为了提高检测率,采用D-S证据融合技术融合多个ELM,能够提高整个检测系统的精确性。但是传统的D-S技术处理冲突信息源时并不理想。因此,本文采过引入证据之间的冲突强度将信息源划分成可接受冲突和不可接受冲突,给出了新的证据理论(Improved Dempster-Shafer,I-DS),同时针对ELM随机产生隐层神经元对算法性能造成影响的缺陷做出了改进。通过实验表明,结合I-DS和改进的ELM能够更高速,更有效的判别入侵行为。     

新的分布交互式多模型广义概率数据关联算法

为有效解决密集杂波环境下分布式多传感器多机动目标跟踪问题,提出了一种基于改进D-S证据组合规则的分布交互式多模型多传感器广义概率数据关联（DIMM-MSGPDA-IDS）算法。该算法首先对各局部节点均应用单传感器的IMM-GPDA算法跟踪多机动目标,并将其各模型的状态估计、协方差估计、模型概率、组合新息及其协方差矩阵等滤波结果送至融合中心;在航迹关联判决结束后,融合中心根据各模型对应似然函数的大小融合不同传感器关于同一目标的模型状态估计及其协方差矩阵,并提出利用三维（3-D）证据进行直接融合的改进D-S算法对来源于同一目标的不同传感器的各模型概率进行有效融合,然后依此概率来更新各目标的状态估计并反馈至各局部节点,使之获得更为精确的状态预测;最后,将该算法与基于D-S证据组合规则的分布交互式多模型多传感器联合概率数据关联（DIMM-MSJPDA-DS）算法进行仿真对比分析。理论分析和仿真结果表明,该算法能够很好地对强机动目标进行跟踪,且其计算量相对较小,是一种有效的分布交互式多模型多传感器多机动目标跟踪算法。     

扩展D-S证据理论在入侵检测中的应用

网络异常行为检测是入侵检测中不可或缺的部分,单一的检测方法很难获得较好的检测结果。针对经典D-S证据理论不能有效合成高度冲突证据的不足,提出将基于改进的加权D-S证据组合方法应用到网络异常行为检测中,并融合多个SVM,建立新的入侵检测模型。该方法通过引入平均证据得到权重系数,以此区分各证据在D-S融合中的影响程度,因此能有效解决证据的高度冲突。仿真结果表明,与传统的基于D-S证据理论的异常检测相比,本模型能够有效提高融合效率,进而提高检测性能。     

扩展D-S证据理论在入侵检测中的应用

网络异常行为检测是入侵检测中不可或缺的部分,单一的检测方法很难获得较好的检测结果。针对经典D-S证据理论不能有效合成高度冲突证据的不足,提出将基于改进的加权D-S证据组合方法应用到网络异常行为检测中,并融合多个SVM,建立新的入侵检测模型。该方法通过引入平均证据得到权重系数,以此区分各证据在D-S融合中的影响程度,因此能有效解决证据的高度冲突。仿真结果表明,与传统的基于D-S证据理论的异常检测相比,本模型能够有效提高融合效率,进而提高检测性能。     

基于改进D-S证据理论的室内环境控制决策系统

针对室内环境因子多且相互作用关系复杂,影响室内环境舒适度的控制精准决策,设计了一种基于改进D-S证据理论的室内环境控制决策系统。首先采用箱线图法和均值替代法检测修复异常采集数据,然后利用距离自适应加权融合算法实现同类传感器数据一级融合,最后利用改进D-S证据理论算法,实现全局融合决策。实验结果表明,改进D-S证据理论算法能够有效避免冲突证据带来的融合决策误差,系统可以实现室内环境控制的精准决策,融合决策精度高,具有一定的推广应用价值。     

一种新型非线性滤波的多特征融合跟踪算法

为有效解决非线性系统的状态估计问题,提出一种新型非线性滤波算法。该算法通过在积分卡尔曼滤波中引入修正因子,对积分点进行优化重组,并采用修正后的积分卡尔曼滤波产生优选建议分布函数,较好地克服了粒子退化现象。在新算法的框架内,利用颜色和运动边缘特征作为观测模型进行视频目标跟踪,并通过D-S证据理论的方法进行权值融合,较好地克服了单一颜色特征在姿态改变、相似背景遮挡等情况下跟踪稳定性较差的问题。实验表明本方法对复杂条件下的目标跟踪问题在保持较强鲁棒性的同时,跟踪精度提升了近32%。     

基于最大互信息区域跟踪的人体行为检测算法

人体行为检测问题不仅需要判断行为的类别,而且需要估计行为发生的时间和位置,有重要的现实应用意义. 人体行为检测的主要难点在于参数空间维度高以及背景运动干扰. 针对上述难点,本文提出了一种基于最大互信息区域跟踪的人体行为检测算法. 该算法将行为区域定义为最大互信息矩形区域,采用稠密轨迹作为底层特征,利用随机森林学习轨迹特征与行为类别的互信息函数,利用轨迹的时间连续性对行为区域进行大时间跨度的预测和跟踪. 实验结果表明,该算法不仅能够有效地识别不同类别的行为,而且能够适应现实场景中背景运动的干扰,从而准确地检测和跟踪行为区域.     

基于堆栈式消噪自编码机的分块目标跟踪

在视觉目标跟踪系统中, 特征的表达和提取是重要的组成部分. 本文提出基于多个自编码机网络相联合的特征提取机, 通过对输入数据进行一定程度的重组, 采用深度学习的理论对其局部特征进行描述并对结果进行联合决策. 结合该网络结构, 本文提出一种融合局部特征的深度信息进行目标跟踪的算法. 将输入图像分块使得大量的乘法运算转化为加法和乘法的混合运算, 相对于全局的特征表达, 大幅降低了运算复杂度. 在跟踪过程中, 目标候选区的各分块权重能够根据相应网络的置信度进行自适应的调整, 提升了跟踪器对光照变化、目标姿态和遮挡的适应. 实验表明, 该跟踪算法在鲁棒性和跟踪速度上表现优秀.     

用D-S证据理论实现多传感器信息融合的一种改进算法

阐述了基于D-S证据理论的多传感器信息融合算法,提供一种基于D-S理论的改进方法以解决融合信息的相关性问题.用滑觉和热觉传感器作实验,对该方法的有效性进行了验证.     

基于云模型的智能教学系统中学习质量的评价

提出将云模型理论应用于智能教学系统的学生模型构建中,建立一种基于云模型的学生学习质量评价方法,利用云对概念的贡献程度进行数据离散化,并引入云变换计算隶属云,最后结合极大判定算法求出更加符合实际的学习质量等级划分。实验结果表明,这种新的评价方法得出的隶属概念不仅能够反映出学生对知识点的掌握程度,还能够反映出学生在学习中的发挥稳定性、心理素质等情况,有利于提高智能教学系统的应用效率。     

PeRSIVA: An empirical evaluation method of a student model of an intelligent e-learning environment for computer programming

In this paper an evaluation method for assessing the effectiveness, accuracy and validity of a student model was presented. Our method is called PeRSIVA and is a combination of the well-known evaluation method of Kirkpatrick and the layered evaluation framework. These well-known and commonly used evaluation techniques have been selected in order to design an accurate and correct evaluation methodology, since there are no clear guidelines in the literature for the evaluation of the student model of an adaptive tutoring system. Furthermore, PeRSIVA method was used to evaluate the hybrid student model, which combines an overlay model with stereotypes and fuzzy logic techniques, of an e-learning system. Particularly, PeRSIVA assesses the results of student modeling in students' satisfaction, performance, progress, behavior and state, as well as the validity of the conclusions drawn by the student model and the validity of the adaptation decision making. The e-learning system was used by the students of a postgraduate program in the field of informatics in the University of Piraeus and the evaluation results demonstrated learning improvements in students and adaptation success to students' needs.     

线上体育课堂在线人数的智能评估方法

由于线上教学时,学生与教师无法面对面交流,给教师监察学生行为与课堂表现造成极大不便,为此研究线上体育课堂在线人数智能评估方法.使用量化分析法采集在线人数学习行为相关信息数据,获得学习行为信息数据集;构建线上课堂在线人数信息处理系统,并利用可视化与平行坐标方法将学习行为数据进行分段评估处理,实现各段学习行为的准确评估.实...     

Intelligent preference selection model based on NRE for evaluating student learning achievement

In order to evaluate student learning achievement, several aspects should be considered, such as exercises, examinations, and observations. Traditionally, such an evaluation calculates a final score using a weighted average method after awarding numerical scores, and then determines a grade according to a set of established crisp criteria. However, this approach lacks the potential to reflect the individual characteristics of a class compared to others. Several researches have used fuzzy techniques to devise practical methods for evaluating student learning achievement to ascertain linguistic terms that are usually used by teachers to assess student learning achievement. However, these approaches are largely based on expert opinions and require complicated computational processes. In this paper, we present a new method for evaluating student learning achievement using an adaptive ordered weighted averaging operator and K-nearest-neighbor classification method. The proposed method simulates the evaluation behavior of teachers when performing a student achievement evaluation based on a norm-referenced evaluation by identifying situations involving the application of intelligence and provides a useful means to award a reasonable grade to students. Furthermore, the proposed method provides a feedback mechanism to update the norm dataset. Therefore, the repetitious use of the feedback mechanism will gradually strengthen the representativeness of the norm dataset.     

Comparing strategies for modeling students learning styles through reinforcement learning in adaptive and intelligent educational systems: An experimental analysis

A huge number of studies attest that learning is facilitated if teaching strategies are in accordance with students learning styles, making the learning process more effective and improving students performances. In this context, this paper presents an automatic, dynamic and probabilistic approach for modeling students learning styles based on reinforcement learning. Three different strategies for updating the student model are proposed and tested through experiments. The results obtained are analyzed, indicating the most effective strategy. Experiments have shown that our approach is able to automatically detect and precisely adjust students’ learning styles, based on the non-deterministic and non-stationary aspects of learning styles. Because of the probabilistic and dynamic aspects enclosed in automatic detection of learning styles, our approach gradually and constantly adjusts the student model, taking into account students’ performances, obtaining a fine-tuned student model.     

Development of an adaptive learning system with two sources of personalization information

Previous research of adaptive learning mainly focused on improving student learning achievements based only on single-source of personalization information, such as learning style, cognitive style or learning achievement. In this paper, an innovative adaptive learning approach is proposed by basing upon two main sources of personalization information, that is, learning behavior and personal learning style. To determine the initial learning styles of the students, the [Keefe, J. W. (1987). Learning Styles: Theory and Practice. Reston, VA: National Association of Secondary School Principals.] questionnaire is employed in our approach. To more precisely reflect the learning behaviors of each student, the interactions and learning results of each student are analyzed when adjusting the subject materials. Based on the innovative approach, an adaptive learning system has been developed; moreover, an experiment was conducted to evaluate the performance of our approach. By analyzing the results from three groups of students using different adaptive learning approaches, it can be found that the innovative approach is helpful in improving both the learning achievement and learning efficiency of individual students.     

基于深度时空残差卷积神经网络的课堂教学视频中多人课堂行为识别

针对课堂教学场景遮挡严重、学生众多,以及目前的视频行为识别算法并不适用于课堂教学场景,且尚无学生课堂行为的公开数据集的问题,构建了课堂教学视频库以及学生课堂行为库,提出了基于深度时空残差卷积神经网络的课堂教学视频中实时多人学生课堂行为识别算法.首先,结合实时目标检测和跟踪,得到每个学生的实时图片流;接着,利用深度时空残...     

Recommender system in collaborative learning environment using an influence diagram

Giving useful recommendations to students to improve collaboration in a learning experience requires tracking and analyzing student team interactions, identifying the problems and the target student. Previously, we proposed an approach to track students and assess their collaboration, but it did not perform any decision analysis to choose a recommendation for the student. In this paper, we propose an influence diagram, which includes the observable variables relevant for assessing collaboration, and the variable representing whether the student collaborates or not. We have analyzed the influence diagram with two machine learning techniques: an attribute selector, indicating the most important attributes that the model uses to recommend, and a decision tree algorithm revealing four different scenarios of recommendation. These analyses provide two useful outputs: (a) an automatic recommender, which can warn of problematic circumstances, and (b) a pedagogical support system (decision tree) that provides a visual explanation of the recommendation suggested.     

A contextual game-based learning approach to improving students' inquiry-based learning performance in social studies courses

Inquiry-based learning, an effective instructional strategy, can be in the form of a problem or task for triggering student engagement. However, how to situate students in meaningful inquiry activities remains to be settled, especially for social studies courses. In this study, a contextual educational computer game is developed to improve students' learning performance based on an inquiry-based learning strategy. An experiment has been conducted on an elementary school social studies course to evaluate the effects of the proposed approach on the inquiry-based learning performances of students with different learning styles. The experimental results indicate that the proposed approach effectively enhanced the students' learning effects in terms of their learning achievement, learning motivation, satisfaction degree and flow state. Furthermore, it is also found that the proposed approach benefited the “active” learning style students more than the “reflective” style students in terms of learning achievement. This suggests the need to provide additional supports to students with particular learning styles in the future.     

Evaluation and management of new service concepts: An ANP-based portfolio approach

One of the most crucial decisions in service development is concept selection. Nevertheless, little attention has been paid to evaluation of new service concepts (NSCs). This study proposes an analytic network process (ANP) approach to evaluation of NSCs. ANP is a multiple criteria decision making (MCDM) method that can accommodate interdependency among decision attributes. The proposed approach measures feasibility of NSCs in terms of strategy, technology, market, implementation, and operation. The derived feasibility values of NSC alternatives are then employed to construct the NSC portfolio matrix, together with customers’ preference. The NSC portfolio matrix is expected to aid decision making on concept selection and provide managerial implications for service development. A case of the mobile information and entertainment service is presented to illustrate the proposed approach.