基于隐半马尔可夫模型的用户兴趣特征提取

针对网络用户兴趣行为特征的抽取,提出了一种基于隐半马尔可夫模型的用户兴趣特征提取模型,通过用状态驻留时间的概率来控制用户浏览行为,使描述兴趣特征的隐状态和时间的相关性更紧密地结合起来,并且根据隐半马尔可夫模型可以产生多观察值序列的特性,把文本信息划分成多个文本块子区域,使每个子区域的特征和其中一个观察值序列对应起来。实验结果表明,利用隐半马尔可夫模型进行特征提取比HMM方法有更高的准确率和召回率。     

A novel approach to equipment health management based on auto-regressive hidden semi-Markov model （AR-HSMM）

As a new maintenance method, CBM （condition based maintenance） is becoming more and more important for the health management of complicated and costly equipment. A prerequisite to widespread deployment of CBM technology and prac- tice in industry is effective diagnostics and prognostics. Recently, a pattern recog- nition technique called HMM （hidden Markov model） was widely used in many fields. However, due to some unrealistic assumptions, diagnositic results from HMM were not so good, and it was difficult to use HMM directly for prognosis. By relaxing the unrealistic assumptions in HMM, this paper presents a novel approach to equip- ment health management based on auto-regressive hidden semi-Markov model （AR-HSMM）. Compared with HMM, AR-HSMM has three advantages： 1） It allows explicitly modeling the time duration of the hidden states and therefore is capable of prognosis. 2） It can relax observations＇ independence assumption by accom- modating a link between consecutive observations. 3） It does not follow the unre- alistic Markov chain＇s memoryless assumption and therefore provides more pow- erful modeling and analysis capability for real problems. To facilitate the computa- tion in the proposed AR-HSMM-based diagnostics and prognostics, new forward- backward variables are defined and a modified forward-backward algorithm is de- veloped. The evaluation of the proposed methodology was carried out through a real world application case study： health diagnosis and prognosis of hydraulic pumps in Caterpillar Inc. The testing results show that the proposed new approach based on AR-HSMM is effective and can provide useful support for the decision- making in equipment health management.     

基于HSMM的人脸表情识别

针对表情识别中存在人脸semi-Markov models,HSMM)的人脸表情识别模型.该模型具有每个状态产生多个观察值、允许观察值缺省等特性,据此识别那些由于局部被遮挡或其它原因引起的丢失特征的人脸表情.实验结果表明,该模型提高了部分遮挡人脸的表情识别效果,同时对无遮挡人脸的表情识别也有所改善.     

An improved noise-robust voice activity detector based on hidden semi-Markov models

To improve the performance of voice activity detector (VAD) in noisy environments, this paper concentrates on three critical aspects related to noise robustness including speech features, feature distributions and temporal dependence. Based on the statistic on TIMIT and NOIZEUS, Mel-frequency cepstrum coefficients (MFCCs) are selected as speech features, Gaussian Mixture distributions (GMD) are applied to associate the observations in MFCC domain with both speech and non-speech states, and Weibull and Gamma distributions are used to explicitly model noise and speech durations, respectively. To integrate these aspects into VAD, the hidden semi-Markov model (HSMM) as a generalized hidden Markov model (HMM) is introduced first. Then the VAD decision is made according to the likelihood ratio test (LRT) incorporating state prior knowledge and modified forward variables of HSMM. We design a recursive way to efficiently calculate modified forward variables. Finally a series of experiments demonstrate: (1) the positive effect of different robustness-related schemes adopted in the proposed VAD; (2) better performance against the standard ITU-T G.729B, Adaptive MultiRate VAD phase 2 (AMR2), Advanced Front-end (AFE), HMM-based VAD and VAD using Laplacian-Gaussian model (LD-GD based VAD).     

Toward finding hidden communities based on user profile

We consider the community detection problem from a partially observable network structure where some edges are not observable. Previous community detection methods are often based solely on the observed connectivity relation and the above situation is not explicitly considered. Even when the connectivity relation is partially observable, if some profile data about the vertices in the network is available, it can be exploited as auxiliary or additional information. We propose to utilize a graph structure (called a profile graph) which is constructed via the profile data, and propose a simple model to utilize both the observed connectivity relation and the profile graph. Furthermore, instead of a hierarchical approach, based on the modularity matrix of the network structure, we propose an embedding approach which utilizes the regularization via the profile graph. Various experiments are conducted over two social network datasets and comparison with several state-of-the-art methods is reported. The results are encouraging and indicate that it is promising to pursue this line of research.     

A novel user identification scheme with key distribution preserving user anonymity for distributed computer networks

Recently, Yang et al. proposed an efficient user identification scheme with key distribution, in which it is possible for the user to anonymously log into a system and establish a secret key shared with the system. Mangipudi and Katti later demonstrated a Deniable-of-Service (DoS) attack on the Yang et al. scheme and then proposed an improvement to withstand such an attack. However, this paper demonstrates an identity disclosure attack to show that neither schemes’ claimed user anonymity requirement can be achieved. We further propose a novel user identification scheme with key distribution preserving user anonymity for distributed computer networks. The proposed scheme not only withstands the attacks mentioned above, but also achieves the following: (i) user anonymity, (ii) key distribution, (iii) mutual authentication, and (iv) key confirmation. The performance of our scheme is of greater efficiency than that of previously proposed schemes in terms of communication costs and computational complexities.     

基于隐马尔科夫模型的用户行为异常检测方法

提出了一种基于HMM的用户行为异常检测的新方法,用shell命令序列作为审计数据,但在数据预处理、用户行为轮廓的表示方面与现有方法不同。仿真实验结果表明,本方法的检测效率和实时性相对较高,在检测准确率方面也有较大优势。     

Fuzzy-genetic approach to recommender systems based on a novel hybrid user model

The main strengths of collaborative filtering (CF), the most successful and widely used filtering technique for recommender systems, are its cross-genre or ‘outside the box’ recommendation ability and that it is completely independent of any machine-readable representation of the items being recommended. However, CF suffers from sparsity, scalability, and loss of neighbor transitivity. CF techniques are either memory-based or model-based. While the former is more accurate, its scalability compared to model-based is poor. An important contribution of this paper is a hybrid fuzzy-genetic approach to recommender systems that retains the accuracy of memory-based CF and the scalability of model-based CF. Using hybrid features, a novel user model is built that helped in achieving significant reduction in system complexity, sparsity, and made the neighbor transitivity relationship hold. The user model is employed to find a set of like-minded users within which a memory-based search is carried out. This set is much smaller than the entire set, thus improving system’s scalability. Besides our proposed approaches are scalable and compact in size, computational results reveal that they outperform the classical approach.     

基于用户的QoS管理模型

随着IP电话、视频点播和电视电话这些网络应用的出现,传统的网络“尽力转发”服务已经不能满足需要。区分服务模型的出现在一定程度上解决了这个问题,但是由于缺乏有效的管理方法,使得网络资源很快耗尽。提出了一种基于用户的服务质量(QoS)管理模型,通过它能够对一个Intranet网内通过认证的用户提供有保证的网络服务。在该模型中,有一个QoS管理服务器,它负责对整个的QoS服务进行管理,网内的网络设备都按照它发送的命令和规则来转发数据分组。     

基于隐马尔科夫模型的J波自动识别检测

J波检测在临床上可以作为判定某些心脏病的一种非创性的标记手段。主要定义了5个精确反映J波特性的特征向量,包括3个时域特征向量和两个基于小波的特征向量,并使用主成分分析减少特征向量的维数,作为分类器的输入。利用这些特征向量训练隐马尔可夫模型作为分类器,输出最终的判定结果。结果表明,提出的方法提供了93.8%的平均准确度、94.2%的平均敏感性、93.3%的平均特异性和93.4%的平均阳性预测值,揭示了很高的评价标准,表明该方法有能力准确地检测识别J波,并且可以利用该方法检测心电图中的其他病变波形。     

Exploring the state sequence space for hidden Markov and semi-Markov chains

The knowledge of the state sequences that explain a given observed sequence for a known hidden Markovian model is the basis of various methods that may be divided into three categories: (i) enumeration of state sequences; (ii) summary of the possible state sequences in state profiles; (iii) computation of a global measure of the state sequence uncertainty. Concerning the first category, the generalized Viterbi algorithm for computing the top L most probable state sequences and the forward-backward algorithm for sampling state sequences are derived for hidden semi-Markov chains and hidden hybrid models combining Markovian and semi-Markovian states. Concerning the second category, a new type of state (and state change) profiles is proposed. The Viterbi forward-backward algorithm for computing these state profiles is derived for hidden semi-Markov chains and hidden hybrid models combining Markovian and semi-Markovian states. Concerning the third category, an algorithm for computing the entropy of the state sequence that explains an observed sequence is proposed. The complementarity and properties of these methods for exploring the state sequence space (including the classical state profiles computed by the forward-backward algorithm) are investigated and illustrated with examples.     

An efficient user identification scheme based on ID-based cryptosystem

Tseng-Jan modified a non-interactive public key distribution system and also proposed several applications based on the Maurer–Yacobi scheme. In their scheme, a user can prove his identity to another user without revealing his secret key. They use a challenge-response-type interactive protocol to achieve their objective. However, in wireless environment, waiting for a corresponding response from the other is time-wasting and consumes the battery of the mobile device. The ability of computing and the capacity of the battery of a mobile device are limited. Therefore, we propose an efficient scheme based on ID-based cryptosystem that is more suitable to be applied in the mobile environment.     

基于隐马尔科夫模型的时空序列预测方法

提出了一种基于时空密度聚类的隐马尔科夫模型对时空序列进行预测的方法。时空序列与一般的时间序列相比,最主要的特征是其时空依赖性以及时空非平稳性。针对如何有效地预测不同尺度分布的时空序列的问题,本文采用基于时空密度聚类的隐马尔科夫模型,该模型不仅能分析时空序列在时间和空间上的相关性,而且可以通过时空序列的分段有效地去除噪声,提高模型预测的精度。本文采用该模型对药品冷藏库中的时空序列温度数据进行分析预测,并与其他预测模型比较,结果显示本文提出的方法更准确有效。     

A novel rating prediction method based on user relationship and natural noise

Rating prediction is a hot spot in the research of recommender systems. There are lots of methods in this field such as collaborative filtering. However, few of these approaches take users’ friendship relationships into consideration, which actually contain significant information for rating prediction. Besides, there exists natural noise in users’ ratings. In this paper, we propose a rating prediction algorithm named NF-SVM based on the analysis of users’ natural noise and relationships. We cluster users to sharpen the similarity attribute among users, and use an iterative algorithm to obtain the rank of users’ rating quality. Then, we analyze users’ rating history to obtain the attributes of users’ natural noise. All these attributes are used to build a training set for SVM to get a prediction model. We also tested our algorithm in a data set which is crawled down from Douban, one of the largest movie rating web sites in China. Then we compared our algorithm with other state-of-the-art rating prediction methods. Extensive experiments show that our algorithm outperforms the other algorithms.     

基于社交媒体用户评论和关注度的电影票房预测模型

通过研究电影票房与社交媒体用户行为的关系,揭示在线口碑(word-of-mouth)对业绩表现的作用。与之前的研究不同,将社交媒体用户评论、用户关注等用户行为数据作为内生变量进行研究,认为用户行为既影响业绩,又被业绩影响。首先,以电影产业为研究对象,分析了每周票房与用户评论、用户评分、用户关注度等之间的关系,通过样板(Panel)数据分析,构建了电影票房预测模型。接着,将票房作为自变量,分析了作为在线口碑表现形式的用户评论、用户关注度与票房的关系。最后,分析了在线口碑自身的特点,得出了多个有意义的结论,如用户评分仅仅是票房收入的反映,其本身并不显著影响票房。本研究具有良好的理论价值和实践意义。     

A statistical model for user preference

Modeling user preference is one of the challenging issues in intelligent information systems. Extensive research has been performed to automatically analyze user preference and to utilize it. One problem still remains: The representation of preference, usually given by measure of vector similarity or probability, does not always correspond to common sense of preference. This problem gets worse in the case of negative preference. To overcome this problem, this paper presents a preference model using mutual information in a statistical framework. This paper also presents a method that combines information of joint features and alleviates problems arising from sparse data. Experimental results, compared with the previous recommendation models, show that the proposed model has the highest accuracy in recommendation tests.     

基于BA模型的微博谣言传播机理研究

基于BA无标度网络模型,融合微博用户网络的特征,构造一种微博谣言传播网络模型。提出一种新的SIR(susceptible-infected-removed)模型,通过数值仿真探讨微博谣言的传播动力学行为。研究表明,当微博用户规模以及节点间新建连接数量不断增大时,微博谣言的最终感染程度显著增大,网络的传播临界值却大幅降低。研究还发现,拥有较大吸引度的节点比例以及微博用户之间添加反向关注的概率基本不影响谣言的传播特性。     

一种基于隐Markov模型的异常检测技术

给出了一种建立隐Markov异常检测模型的算法,并从序列支持度分析、序列预测两个方面研究了该模型在异常检测中的应用,通过实验,分析了影响这一检测方法效果和效率的因素。实验表明,该方法能在不需要任何安全方面背景知识的情况下,有效地检测出入侵行为。     

基于用户界面模型的界面模板

界面模板是一种崭新的界面设计模式，提出了基于界面模型的界面模板概念，在支持界面自动生成的界面开发方法中实现从抽象界面到具体界面的转化。讨论了界面模板的构成与表达、界面模板的分类以及界面模板库体系结构，说明了界面模板的用法。