Mining sequential patterns from probabilistic databases

Knowledge and Information Systems - This paper considers the problem of sequential pattern mining (SPM) in probabilistic databases. Specifically, we consider SPM in situations where there is...     

基于混合概率背景模型的视频分割方法

提出一种新的基于混合概率模型的背景建模方法,用于视频中前景物体的检测与分割。主要利用两个概率模型：隐马尔可夫模型和概率图模型建立一个混合的贝叶斯网概率模型,对视频输入中背景变化的时间和空间局部相关性(同现性)进行学习。在建立正确模型参数的基础上,贝叶斯信念传播算法根据图像输入预测当前背景状态的后验分布,并根据预测得到的背景状态对输入图像进行分割。实验结果验证了该方法的有效性和在复杂背景变化下的鲁棒性。     

Mining atomic Chinese abbreviations with a probabilistic single character recovery model

An HMM-based single character recovery (SCR) model is proposed in this paper to extract a large set of atomic abbreviations and their full forms from a text corpus. By an “atomic abbreviation,” it refers to an abbreviated word consisting of a single Chinese character. This task is important since Chinese abbreviations cannot be enumerated exhaustively but the abbreviation process for compound words seems to be compositional. One can often decode an abbreviated word character by character to its full form. With a large atomic abbreviation dictionary, one may be able to handle multiple character abbreviation problems more easily based on the compositional property of abbreviations.     

数据流中基于计数的频繁模式挖掘

频繁项集是挖掘流数据挖掘的基本任务。许多近似算法能够有效进行频繁项挖掘,但不能有效控制内存资源消耗。文章提出并实现了0—δ算法,能够有效控制内存消耗问题。在充分的理论分析基础上,还用翔实的实验证明了新方法的有效性。     

A flexible approach to model coupling through probabilistic pooling

Model coupling is an important approach to studying the dynamics of complex systems, but by introducing new feedback loops, the dynamics of coupled models can be artificially distorted. This paper describes a new method of model coupling which addresses this problem through a dynamic form of regularization. The method allows the time series evolution of model variables to be mutually informed by multiple models, and models to influence each other in proportion to their degree of certainty. Uncoupled forms of the coupled models can act as dynamic priors on the trajectory of coupled variables, strengthening model stability and offering additional calibration of the coupling process. Finally, models that describe different spatial scales can be coupled into multi-scale models, so that, for example, spatially-distributed models can be coupled with aggregate models, and influence one another. We apply this technique to a coupled socio-ecological system of population growth and ecosystem harvesting.     

Improving urban classification through fuzzy supervised classification and spectral mixture analysis

In this study, a fuzzy‐spectral mixture analysis (fuzzy‐SMA) model was developed to achieve land use/land cover fractions in urban areas with a moderate resolution remote sensing image. Differed from traditional fuzzy classification methods, in our fuzzy‐SMA model, two compulsory statistical measurements (i.e. fuzzy mean and fuzzy covariance) were derived from training samples through spectral mixture analysis (SMA), and then subsequently applied in the fuzzy supervised classification. Classification performances were evaluated between the ‘estimated’ landscape class fractions from our method and the ‘actual’ fractions generated from IKONOS data through manual interpretation with heads‐up digitizing option. Among all the sub‐pixel classification methods, fuzzy‐SMA performed the best with the smallest total_MAE (MAE, mean absolute error) (0.18) and the largest Kappa (77.33%). The classification results indicate that a combination of SMA and fuzzy logic theory is capable of identifying urban landscapes at sub‐pixel level.     

Information filtering using a probabilistic model

We propose an information filtering system based on a probabilistic model. We make an assumption that a document consists of words which occur according to a probability distribution, and regard a document as a sample drawn according to that distribution. In this article, we adopt a multinomial distribution and represent a document as probability which has random values as the words in the document. When an information filtering system selects information, it uses the similarity between the user's interests (a user profile) and a document. Since our proposed system is constructed under the probabilistic model, the similarity is defined using the Kullback Leibler divergence. To create the user profile, we must optimize the Kullback Leibler divergence. Since the Kullback Leibler divergence is a nonlinear function, we use a genetic algorithm to optimize it. We carry out experiments and confirm effectiveness of the proposed method. This work was presented in part at the 10th International Symposium on Artificial Life and Robotics, Oita, Japan, February 4–6, 2005     

Robust semi-supervised mixture probabilistic principal component regression model development and application to soft sensors

Traditional data-based soft sensors are constructed with equal numbers of input and output data samples, meanwhile, these collected process data are assumed to be clean enough and no outliers are mixed. However, such assumptions are too strict in practice. On one hand, those easily collected input variables are sometimes corrupted with outliers. On the other hand, output variables, which also called quality variables, are usually difficult to obtain. These two problems make traditional soft sensors cumbersome. To deal with both issues, in this paper, the Student's t distributions are used during mixture probabilistic principal component regression modeling to tolerate outliers with regulated heavy tails. Furthermore, a semi-supervised mechanism is incorporated into traditional probabilistic regression so as to deal with the unbalanced modeling issue. For simulation, two case studies are provided to demonstrate robustness and reliability of the new method.     

Constructing probabilistic graphical model from predicate formulas for fusing logical and probabilistic knowledge

Expressing knowledge as expert experience and discovering knowledge implied in data are two important ways for knowledge acquisition. Consistent combination of these two kinds of knowledge has attracted much attention due to the potential applications to knowledge fusion and wide requirements of decision support. In this paper, we focus on the probabilistic modeling of expert experience represented as logical predicate formulas, aiming at the effective fusion of logical and probabilistic knowledge. Taking qualitative probabilistic network (QPN) as the underlying framework of probabilistic knowledge implied in data as well as the abstraction of general Bayesian networks (BNs), we are to construct the probabilistic graphical model for both the given predicate formulas and the ultimate result of knowledge fusion. We first propose the concept and the construction algorithm of predicate graph (PG) to describe the dependence relations among predicate formulas, and discuss PG’s probabilistic semantics correspondingly. We then prove that PG is a probability dependency model and has the same semantics with a general probabilistic graphical model. Consequently, we give the method for fusing PG and QPN. Experimental results show the effectiveness of our methods.     

基于混合高斯模型的运动阴影抑制算法

复杂场景的背景建模、运动目标检测、运动目标所投射阴影的检测与抑制在智能监控、机器人视觉、视频会议等领域有着广泛的应用。在运动前景检测阶段,给出了一种改进的混合高斯算法进行场景的背景建模,根据各点像素值出现的混乱程度采取不同的高斯函数参数更新机制,缓解了混合高斯算法计算量大的问题。在运动目标的阴影检测与抑制中,提出了一种基于混合高斯的阴影抑制算法,该算法先利用阴影在HSV颜色空间的特点,判断被检测为运动前景的像素是否为疑似阴影,然后用混合高斯阴影模型对所有疑似阴影值进行聚类,进一步完成阴影抑制。仿真结果表明：该算法可更有效地抑制阴影对运动目标检测的影响,并具有较强的实时性。     

Instrument identification and pitch estimation in multi-timbre polyphonic musical signals based on probabilistic mixture model decomposition

In this paper, we propose a method based on probabilistic mixture model decomposition that can simultaneously identify musical instrument types, estimate pitches and assign each pitch to its source instrument in monaural polyphonic audio containing multiple sources. In the proposed system, the probability density function (PDF) of the observed mixture note is treated as a weighted sum approximation of all possible note models. These note models, covering 14 instruments and all their possible pitches, describe their dynamic frequency envelopes in terms of probability. The weight coefficients, indicating the probabilities of the existence of pitches of a certain type of instrument, are estimated using the Expectation-Maximization (EM) algorithm. The weight coefficients are used to detect the types of source instruments and the pitches. The results of experiments involving 14 instruments within a designated pitch range F3–F6 (37 pitches) demonstrate a good discrimination capability, especially in instrument identification and instrument-pitch identification. For the entire system including the note onset detection tool, using quartet polyphonic recordings, the average F-measure values of instrument-pitch identification, instrument identification and pitch estimation were 55.4, 62.5 and 86 % respectively.     

Continuous-valued probabilistic behavior in a VLSI generative model

This paper presents the VLSI implementation of the continuous restricted Boltzmann machine (CRBM), a probabilistic generative model that is able to model continuous-valued data with a simple and hardware-amenable training algorithm. The full CRBM system consists of stochastic neurons whose continuous-valued probabilistic behavior is mediated by injected noise. Integrating on-chip training circuits, the full CRBM system provides a platform for exploring computation with continuous-valued probabilistic behavior in VLSI. The VLSI CRBM's ability both to model and to regenerate continuous-valued data distributions is examined and limitations on its performance are highlighted and discussed.     

License plate localization based on a probabilistic model

Extraction of the license plate region is the challenging first step in the license plate recognition system. We propose a novel feature fusion concept for plate extraction. The image-feature extraction process is modeled as a feature-detection problem in noise. The geometric features are probabilistically modeled and detected under various detection thresholds. These detection results are then fused within the Bayesian framework to obtain the features for further processing. Along with a probabilistic model, a pixels voting algorithm is also tested through threshold variation.     

A study of a Gaussian mixture model for urban land-cover mapping based on VHR remote sensing imagery

This article proposes a Gaussian-mixture-model (GMM)-based method with optimal Gaussian components to address the high intra-class spectral variability in urban land-cover mapping using remote sensing images with very high resolution (VHR). GMMs can simulate and approximate any data distribution provided the optimal Gaussian components can be found. Through improving the model parameters in view of the characteristic of VHR remote sensing images, the parameter space of GMM is optimized significantly, and the model can find the optimal Gaussian components that are suitable for remote sensing images with different resolutions. Experimental results of Wuhan urban area using two images with different resolutions have demonstrated the efficiency and effectiveness of the model. The optimized GMM-based method performs at least comparably or superior to the state-of-the-art classifiers such as support vector machines (SVMs), characterizes man-made land-cover types better than conventional methods, fuses spectral and textural features of VHR image properly, and meanwhile has lower computational complexity.     

Knowledge discovery through directed probabilistic topic models: a survey

Graphical models have become the basic framework for topic based probabilistic modeling. Especially models with latent variables have proved to be effective in capturing hidden structures in the data. In this paper, we survey an important subclass Directed Probabilistic Topic Models (DPTMs) with soft clustering abilities and their applications for knowledge discovery in text corpora. From an unsupervised learning perspective, “topics are semantically related probabilistic clusters of words in text corpora; and the process for finding these topics is called topic modeling”. In topic modeling, a document consists of different hidden topics and the topic probabilities provide an explicit representation of a document to smooth data from the semantic level. It has been an active area of research during the last decade. Many models have been proposed for handling the problems of modeling text corpora with different characteristics, for applications such as document classification, hidden association finding, expert finding, community discovery and temporal trend analysis. We give basic concepts, advantages and disadvantages in a chronological order, existing models classification into different categories, their parameter estimation and inference making algorithms with models performance evaluation measures. We also discuss their applications, open challenges and future directions in this dynamic area of research.     

Topic evolution based on the probabilistic topic model: a review

Accurately representing the quantity and characteristics of users’ interest in certain topics is an important problem facing topic evolution researchers, particularly as it applies to modern online environments. Search engines can provide information retrieval for a specified topic from archived data, but fail to reflect changes in interest toward the topic over time in a structured way. This paper reviews notable research on topic evolution based on the probabilistic topic model from multiple aspects over the past decade. First, we introduce notations, terminology, and the basic topic model explored in the survey, then we summarize three categories of topic evolution based on the probabilistic topic model: the discrete time topic evolution model, the continuous time topic evolutionmodel, and the online topic evolution model. Next, we describe applications of the topic evolution model and attempt to summarize model generalization performance evaluation and topic evolution evaluation methods, as well as providing comparative experimental results for different models. To conclude the review, we pose some open questions and discuss possible future research directions.     

Estimating the mixing proportion in a semiparametric mixture model

In this paper, we investigate methods of estimating the mixing proportion in the case when one of the probability densities is not specified analytically in a mixture model. The methodology we propose is motivated by a sequential clustering algorithm. After a sequential clustering algorithm finds the center of a cluster, the next step is to identify observations belonging to that cluster. If we assume that the center of the cluster is known and that the distribution of observations not belonging to the cluster is unknown, the problem of identifying observations in the cluster is similar to the problem of estimating the mixing proportion in a special two-component mixture model. The mixing proportion can be considered as the proportion of observations belonging to the cluster. We propose two estimators for parameters in the model and compare the performance of these two estimators in several different cases.     

Real-time traffic incident detection using a probabilistic topic model

Traffic congestion occurs frequently in urban settings, and is not always caused by traffic incidents. In this paper, we propose a simple method for detecting traffic incidents from probe-car data by identifying unusual events that distinguish incidents from spontaneous congestion. First, we introduce a traffic state model based on a probabilistic topic model to describe the traffic states for a variety of roads. Formulas for estimating the model parameters are derived, so that the model of usual traffic can be learned using an expectation–maximization algorithm. Next, we propose several divergence functions to evaluate differences between the current and usual traffic states and streaming algorithms that detect high-divergence segments in real time. We conducted an experiment with data collected for the entire Shuto Expressway system in Tokyo during 2010 and 2011. The results showed that our method discriminates successfully between anomalous car trajectories and the more usual, slowly moving traffic patterns.