Data spread-based entropy clustering method using adaptive learning

Clustering analysis is to identify inherent structures and discover useful information from large amount of data. However, the decision makers may suffer insufficient understanding the nature of the data and do not know how to set the optimal parameters for the clustering method. To overcome the drawback above, this paper proposes a new entropy clustering method using adaptive learning. The proposed method considers the data spreading to determine the adaptive threshold within parameters optimized by adaptive learning. Four datasets in UCI database are used as the experimental data to compare the accuracy of the proposed method with the listing clustering methods. The experimental results indicate that the proposed method is superior to the listing methods.     

Incentive compatible regression learning

We initiate the study of incentives in a general machine learning framework. We focus on a game-theoretic regression learning setting where private information is elicited from multiple agents with different, possibly conflicting, views on how to label the points of an input space. This conflict potentially gives rise to untruthfulness on the part of the agents. In the restricted but important case when every agent cares about a single point, and under mild assumptions, we show that agents are motivated to tell the truth. In a more general setting, we study the power and limitations of mechanisms without payments. We finally establish that, in the general setting, the VCG mechanism goes a long way in guaranteeing truthfulness and economic efficiency.     

对步态空时数据的连续特征子空间分析

提出一种基于空时特征提取的人体步态识别算法。连续的特征子空间学习依次提取出步态的时间与空间特征：第一次特征子空间学习对步态的频域数据进行主成分分析,步态数据被转化为周期特征矢量;第二次特征子空间学习对步态数据的周期特征矢量形式进行主成分分析加线性判别分析的联合分析,步态数据被进一步转化为步态特征矢量。步态特征矢量同时包含运动的周期特征以及人体的形态特征,具有很强的识别能力。在USF步态数据库上的实验结果显示,该算法识别率较其他同类算法有明显提升。     

Chinese undergraduate students' perceptions of mobile learning: Conceptions,learning profiles,and approaches

Close links between students' conceptions of and approaches to learning were established in the past research. However, only a few quantitative studies investigated this relationship particularly with regard to mobile learning (m‐learning). The correlation between learners' conceptions and approaches to m‐learning was analysed using a partial least squares analysis applied to data obtained from a sample of 971 undergraduate students in China. The results indicated that students' conceptions of m‐learning could be classified into reproductive, transitional, and constructive levels. Students may hold multiple m‐learning applications than a predominant one; hence, examining m‐learning as one monolithic entity may provide limited information. Latent profile analysis identified four learning profiles based on students' preferred m‐learning applications: passive, mixed, surface‐supportive, and high‐engagement.. Moreover, a general trend was observed, whereby students with reproductive and surface‐supportive learning profiles showed a tendency to adopt surface approaches, whereas those expressing constructive and mixed learning profiles were more inclined to adopt deep approaches. Interestingly, students with transitional conceptions and high‐engagement learning profiles tended to take both surface and deep approaches.     

Analyzing collaborative learning processes automatically: Exploiting the advances of computational linguistics in computer-supported collaborative learning

In this article we describe the emerging area of text classification research focused on the problem of collaborative learning process analysis both from a broad perspective and more specifically in terms of a publicly available tool set called TagHelper tools. Analyzing the variety of pedagogically valuable facets of learners’ interactions is a time consuming and effortful process. Improving automated analyses of such highly valued processes of collaborative learning by adapting and applying recent text classification technologies would make it a less arduous task to obtain insights from corpus data. This endeavor also holds the potential for enabling substantially improved on-line instruction both by providing teachers and facilitators with reports about the groups they are moderating and by triggering context sensitive collaborative learning support on an as-needed basis. In this article, we report on an interdisciplinary research project, which has been investigating the effectiveness of applying text classification technology to a large CSCL corpus that has been analyzed by human coders using a theory-based multi-dimensional coding scheme. We report promising results and include an in-depth discussion of important issues such as reliability, validity, and efficiency that should be considered when deciding on the appropriateness of adopting a new technology such as TagHelper tools. One major technical contribution of this work is a demonstration that an important piece of the work towards making text classification technology effective for this purpose is designing and building linguistic pattern detectors, otherwise known as features, that can be extracted reliably from texts and that have high predictive power for the categories of discourse actions that the CSCL community is interested in.

From learning in the limit to stochastic finite learning

Inductive inference can be considered as one of the fundamental paradigms of algorithmic learning theory. We survey results recently obtained and show their impact to potential applications.     

基于机器学习算法的主机恶意代码检测技术研究

对机器学习算法下主机恶意代码检测的主流技术途径进行了研究,分别针对静态、动态这2种分析模式下的检测方案进行了讨论,涵盖了恶意代码样本采集、特征提取与选择、机器学习算法分类模型的建立等要点。对机器学习算法下恶意代码检测的未来工作与挑战进行了梳理。为下一代恶意代码检测技术的设计和优化提供了重要的参考。     

半监督学习在链接预测问题中的应用

链接预测是社会网络分析领域的关键问题,研究如何从已知网络中预测可能存在的新链接。现实网络中存在了大量未连接的节点对,从中挖掘潜在信息可以帮助实现链接预测任务。将链接预测视为二类分类问题,使用半监督学习技术,利用网络中的未标记数据帮助学习。使用了两种半监督范式:自我训练和协同训练。在现实数据集Enron和DBLP中的实验结果表明,链接预测任务中采用未标记数据能够有效提高预测的准确率。     

Image retrieval based on incremental subspace learning

Many problems in information processing involve some form of dimensionality reduction, such as face recognition, image/text retrieval, data visualization, etc. The typical linear dimensionality reduction algorithms include principal component analysis (PCA), random projection, locality-preserving projection (LPP), etc. These techniques are generally unsupervised which allows them to model data in the absence of labels or categories. In this paper, we propose a semi-supervised subspace learning algorithm for image retrieval. In relevance feedback-driven image retrieval system, the user-provided information can be used to better describe the intrinsic semantic relationships between images. Our algorithm is fundamentally based on LPP which can incorporate user's relevance feedbacks. As the user's feedbacks are accumulated, we can ultimately obtain a semantic subspace in which different semantic classes can be best separated and the retrieval performance can be enhanced. We compared our proposed algorithm to PCA and the standard LPP. Experimental results on a large collection of images have shown the effectiveness and efficiency of our proposed algorithm.     

Systems for strategic learning

An important decision support system component is machine learning/data mining. Classical machine learning methods implicitly assume that attributes of instances under classification do not change to acquire a positive classification. However, in many situations these instances represent people or organizations that can proactively seek to alter their characteristics to gain a positive classification. We argue that the learning mechanism should take this possible strategic learning into consideration during the induction process. We call this strategic learning. In this paper we define this concept, summarize related research, and present a number of future research areas. This article is part of the “Handbook on Decision Support Systems” edited by Frada Burstein and Clyde W. Holsapple (2008) Springer.     

Productive use of learning resources in an online problem-based learning environment

This study examined students’ use of learning resources in a technologically-mediated online learning environment. Undergraduate student groups were engaged in an online problem-based learning (PBL) environment, rich with pre-selected video and knowledge resources. Quantitative and qualitative analyses showed that students accessed resources fairly frequently and benefited from them. Resources helped students construct a rich understanding of the problem and provided ideas for problem solutions. Detailed analyses of resource exploration along with contrasting case analyses between high-achieving and low-achieving student groups suggested that for learning to be effective in resource-rich environments, students first need to develop an understanding of the resources and learn how to access them efficiently. Second, students need to learn to process the contents of resources in meaningful ways so that they can integrate diverse resources to form a coherent understanding and apply them to solve problems. Finally, students need to develop knowledge and skills to use resources collaboratively, such as sharing and relating to each other’s resources. The results indicated that students, especially low-achieving students, need guidance to use resources effectively in resource-rich learning environments.     

Machine learning techniques for business blog search and mining

Weblogs, or blogs, have rapidly gained in popularity over the past few years. In particular, the growth of business blogs that are written by or provide commentary on businesses and companies opens up new opportunities for developing blog-specific search and mining techniques. In this paper, we propose probabilistic models for blog search and mining using two machine learning techniques, latent semantic analysis (LSA) and probabilistic latent semantic analysis (PLSA). We implement the models in our database of business blogs, BizBlogs07, with the aim of achieving higher precision and recall. The probabilistic model is able to segment the business blogs into separate topic areas, which is useful for keywords detection on the blogosphere. Various term-weighting schemes and factor values were also studied in detail, which reveal interesting patterns in our database of business blogs. Our multi-functional business blog system is indeed found to be very different from existing blog search engines, as it aims to provide better relevance and precision of the search.     

Stability-based PAC-Bayes analysis for multi-view learning algorithms

Multi-view learning exploits structural constraints among multiple views to effectively learn from data. Although it has made great methodological achievements in recent years, the current generalization theory is still insufficient to prove the merit of multi-view learning. This paper blends stability into multi-view PAC-Bayes analysis to explore the generalization performance and effectiveness of multi-view learning algorithms. We propose a novel view-consistency regularization to produce an informative prior that helps to obtain a stability-based multi-view bound. Furthermore, we derive an upper bound on the stability coefficient that is involved in the PAC-Bayes bound of multi-view regularization algorithms for the purpose of computation, taking the multi-view support vector machine as an example. Experiments provide strong evidence on the advantageous generalization bounds of multi-view learning over single-view learning. We also explore strengths and weaknesses of the proposed stability-based bound compared with previous non-stability multi-view bounds experimentally.     

Phonetic-enriched text representation for Chinese sentiment analysis with reinforcement learning

The Chinese pronunciation system offers two characteristics that distinguish it from other languages: deep phonemic orthography and intonation variations. In this paper, we hypothesize that these two important properties can play a major role in Chinese sentiment analysis. In particular, we propose two effective features to encode phonetic information and, hence, fuse it with textual information. With this hypothesis, we propose Disambiguate Intonation for Sentiment Analysis (DISA), a network that we develop based on the principles of reinforcement learning. DISA disambiguates intonations for each Chinese character (pinyin) and, hence, learns precise phonetic representations. We also fuse phonetic features with textual and visual features to further improve performance. Experimental results on five different Chinese sentiment analysis datasets show that the inclusion of phonetic features significantly and consistently improves the performance of textual and visual representations and surpasses the state-of-the-art Chinese character-level representations.     

ROC分析技术在机器学习中的应用

ROC(受试者工作特征)分析技术是一种用来衡量分类算法和图示它们性能的技术。与传统的正确率相比,ROC分析更能够全面地描述分类算法的分类性能。该方法具有可信度高,描述客观精确,特别是不受数据环境影响等优势。对国内外这一方法的研究成果进行了较为系统地介绍,详细分析了它的优缺点,最后对这一技术的发展进行了展望。     

主题关注网络的表示学习

针对异质网络表示学习仅从结构方面考虑社交关系而忽略语义这一问题,结合用户间的社交关系和用户对主题的偏好两个方面,提出基于主题关注网络的表示学习算法。首先,针对主题关注网络的特点,结合集对分析理论的同异反（确定与不确定）思想,给出转移概率模型;然后,在转移概率模型的基础上提出了一种基于两类节点的随机游走算法,以得到相对高质量的随机游走序列;最后,基于序列中两类节点建模得到主题关注网络的嵌入向量空间表示。理论分析和在豆瓣数据集上的实验结果表明,结合转移概率模型的随机游走算法能更全面地分析网络中节点的连接关系,当划分社区的个数为13时,所提算法的模块度为0.699 8,相比metapath2vec算法提高了近5%,可以更详细地捕获网络中的信息。     

Approximate adjoint-based iterative learning control

This paper characterises stochastic convergence properties of adjoint-based (gradient-based) iterative learning control (ILC) applied to systems with load disturbances, when provided only with approximate gradient information and noisy measurements. Specifically, conditions are discussed under which the approximations will result in a scheme which converges to an optimal control input. Both the cases of time-invariant step sizes and cases of decreasing step sizes (as in stochastic approximation) are discussed. These theoretical results are supplemented with an application on a sequencing batch reactor for wastewater treatment plants, where approximate gradient information is available. It is found that for such case adjoint-based ILC outperforms inverse-based ILC and model-free P-type ILC, both in terms of convergence rate and measurement noise tolerance.     

Label distribution for multimodal machine learning

Multimodal machine learning(MML)aims to understand the world from multiple related modalities.It has attracted much attention as multimodal data has become increasingly available in real-world application.It is shown that MML can perform better than single-modal machine learning,since multi-modalities containing more information which could complement each other.However,it is a key challenge to fuse the multi-modalities in MML.Different from previous work,we further consider the side-information,which reflects the situation and influences the fusion of multi-modalities.We recover multimodal label distribution(MLD)by leveraging the side-information,representing the degree to which each modality contributes to describing the instance.Accordingly,a novel framework named multimodal label distribution learning(MLDL)is proposed to recover the MLD,and fuse the multimodalities with its guidance to learn an in-depth understanding of the jointly feature representation.Moreover,two versions of MLDL are proposed to deal with the sequential data.Experiments on multimodal sentiment analysis and disease prediction show that the proposed approaches perform favorably against state-of-the-art methods.     

基于深度学习的安卓恶意应用检测

针对传统安卓恶意程序检测技术检测准确率低,对采用了重打包和代码混淆等技术的安卓恶意程序无法成功识别等问题,设计并实现了DeepDroid算法。首先,提取安卓应用程序的静态特征和动态特征,结合静态特征和动态特征生成应用程序的特征向量;然后,使用深度学习算法中的深度置信网络（DBN）对收集到的训练集进行训练,生成深度学习网络;最后,利用生成的深度学习网络对待测安卓应用程序进行检测。实验结果表明,在使用相同测试集的情况下,DeepDroid算法的正确率比支持向量机（SVM）算法高出3.96个百分点,比朴素贝叶斯（Naive Bayes）算法高出12.16个百分点,比K最邻近（KNN）算法高出13.62个百分点。DeepDroid算法结合了安卓应用程序的静态特征和动态特征,采用了动态检测和静态检测相结合的检测方法,弥补了静态检测代码覆盖率不足和动态检测误报率高的缺点,在特征识别的部分采用DBN算法使得网络训练速度得到保证的同时还有很高的检测正确率。