Universum, a collection of nonexamples that do not belong to any class of interest, has become a new research topic in machine learning. This paper devises a semi-supervised learning with Universum algorithm based on boosting technique, and focuses on situations where only a few labeled examples are available. We also show that the training error of AdaBoost with Universum is bounded by the product of normalization factor, and the training error drops exponentially fast when each weak classifier is slightly better than random guessing. Finally, the experiments use four data sets with several combinations. Experimental results indicate that the proposed algorithm can benefit from Universum examples and outperform several alternative methods, particularly when insufficient labeled examples are available. When the number of labeled examples is insufficient to estimate the parameters of classification functions, the Universum can be used to approximate the prior distribution of the classification functions. The experimental results can be explained using the concept of Universum introduced by Vapnik, that is, Universum examples implicitly specify a prior distribution on the set of classification functions.

Processing large volumes of streaming data in near-real-time is becoming increasingly important as the Internet, sensor networks and network traffic grow. Online machine learning is a typical means of dealing with streaming data, since it allows the classification model to learn one instance of data at a time. Although many online learning methods have been developed since the development of the Perceptron algorithm, existing online methods assume that the number of classes is available in advance of classification process. However, this assumption is unrealistic for large scale or streaming data sets. This work proposes an online Chinese restaurant process (CRP) algorithm, which is an online and nonparametric algorithm, to tackle this problem. This work proposes a relaxing function as part of the prior and updates the parameters with the likelihood function in terms of the consistency between the true label information and predicted result. This work presents two Gibbs sampling algorithms to perform posterior inference. In the experiments, the online CRP is applied to three massive data sets, and compared with several online learning and batch learning algorithms. One of the data sets is obtained from Wikipedia, which comprises approximately two million documents. The experimental results reveal that the proposed online CRP performs well and efficiently on massive data sets. Finally, this work proposes two methods to update the hyperparameter $\alpha$ of the online CRP. The first method is based on the posterior distribution of $\alpha$, and the second exploits the property of online learning, namely adapting to change, to adjust $\alpha$ dynamically.

This paper devises a semi-supervised learning method called semi-supervised linear discriminant clustering (Semi-LDC). The proposed algorithm considers clustering and dimensionality reduction simultaneously by connecting K -means and linear discriminant analysis (LDA). The goal is to find a feature space where the K -means can perform well in the new space. To exploit the information brought by unlabeled examples, this paper proposes to use soft labels to denote the labels of unlabeled examples. The Semi-LDC uses the proposed algorithm, called constrained-PLSA, to estimate the soft labels of unlabeled examples. We use soft LDA with hard labels of labeled examples and soft labels of unlabeled examples to find a projection matrix. The clustering is then performed in the new feature space. We conduct experiments on three data sets. The experimental results indicate that the proposed method can generally outperform other semi-supervised methods. We further discuss and analyze the influence of soft labels on classification performance by conducting experiments with different percentages of labeled examples. The finding shows that using soft labels can improve performance particularly when the number of available labeled examples is insufficient to train a robust and accurate model. Additionally, the proposed method can be viewed as a framework, since different soft label estimation methods can be used in the proposed method according to application requirements.

In this paper, we propose an algorithm called coherence hidden Markov model (HMM) to extract coherence features and rank content. Coherence HMM is a variant of HMM and is used to model the stochastic process of essay writing and identify topics as hidden states, given sequenced clauses as observations. This study uses probabilistic latent semantic analysis for parameter estimation of coherence HMM. In coherence-feature extraction, support vector regression (SVR) with surface features and coherence features is used for essay grading. The experimental results indicate that SVR can benefit from coherence features. The adjacent agreement rate and the exact agreement rate are 95.24% and 59.80%, respectively. Moreover, this study submits high-scoring essays to the same experiment and finds that the adjacent agreement rate and exact agreement rate are 98.33% and 64.50%, respectively. In content ranking, we design and implement an intelligent assisted blog writing system based on the coherence-HMM ranking model. Several corpora are employed to help users efficiently compose blog articles. When users finish composing a clause or sentence, the system provides candidate texts for their reference based on current clause or sentence content. The experimental results demonstrate that all participants can benefit from the system and save considerable time on writing articles.

In this paper, we design and develop a movie-rating and review-summarization system in a mobile environment. The movie-rating information is based on the sentiment-classification result. The condensed descriptions of movie reviews are generated from the feature-based summarization. We propose a novel approach based on latent semantic analysis (LSA) to identify product features. Furthermore, we find a way to reduce the size of summary based on the product features obtained from LSA. We consider both sentiment-classification accuracy and system response time to design the system. The rating and review-summarization system can be extended to other product-review domains easily.

Subtopic segmentation is a critical task in numerous applications, including information retrieval, automatic summarization, essay scoring, and others. Although several approaches have been developed, many are ineffective for specific domains with a small corpus because of the fuzziness of the semantics of words and sentences in the corpus. This paper explores the problem of subtopic segmentation by proposing a fuzzy model for the semantics of both words and sentences. The model has three characteristics. First, it can deal with the uncertainty in the semantics of words and sentences. Secondly, it can measure the fuzzy similarity between the fuzzy semantics of sentences. Thirdly, it can develop a fuzzy algorithm for segmenting a text into several subtopic segments. The experiments, especially for a short text with a small corpus in a specific domain, indicate that the method can efficiently increase the accuracy of subtopic segmentation over previous methods.

In this paper, we extend the work of Kraft et al. to present a new method for fuzzy information retrieval based on fuzzy hierarchical clustering and fuzzy inference techniques. First, we present a fuzzy agglomerative hierarchical clustering algorithm for clustering documents and to get the document cluster centers of document clusters. Then, we present a method to construct fuzzy logic rules based on the document clusters and their document cluster centers. Finally, we apply the constructed fuzzy logic rules to modify the user's query for query expansion and to guide the information retrieval system to retrieve documents relevant to the user's request. The fuzzy logic rules can represent three kinds of fuzzy relationships (i.e., fuzzy positive association relationship, fuzzy specialization relationship and fuzzy generalization relationship) between index terms. The proposed fuzzy information retrieval method is more flexible and more intelligent than the existing methods due to the fact that it can expand users' queries for fuzzy information retrieval in a more effective manner.

In this paper, an intelligent fuzzy information retrieval system with an automatically constructed knowledge base is presented. The knowledge base is represented by a multi-relationship fuzzy concept network that can depict the relationships and their associating relevance degrees between concepts clearly. Based on the multi-relationship fuzzy concept network architecture, the users of the fuzzy information retrieval system can submit a fuzzy contextual query that specifies the search context in the query formula. The fuzzy information retrieval system retrieves documents whose contents are relevant to the user's query by some required relationships beneath a specified search context. The proposed fuzzy information retrieval method is more intelligent and more flexible than the existing methods since it can automatically construct knowledge bases (i.e, multi-relationship fuzzy concept networks) and it provides contextual search capability that allows users to specify fuzzy contextual queries in a more flexible manner.

This paper presents a new method for document retrieval using fuzzy-valued concept networks, where the relevant degrees between the concepts in a fuzzy-valued concept network are represented by arbitrary shapes of fuzzy numbers. There are two kinds of relevant relationships between any two concepts in a fuzzy-valued concept network, i.e., fuzzy positive association and fuzzy negative association. The relevant matrices and the relationship matrices are used to model the fuzzy-valued concept network. The elements in a relevant matrix represent the relevant degrees between concepts. The elements in a relationship matrix represent the relevant relationships between concepts. Furthermore, ne also allow users' queries to be represented by arbitrary shapes of fuzzy numbers and to use fuzzy positive association relationship and fuzzy negative association relationship for formulating their queries for increasing the flexibility of fuzzy information retrieval systems. We also present an information retrieval method in the Internet environment based on the network-type fuzzy-valued concept network architecture.

We present a fuzzy information retrieval method based on fuzzy-valued concept networks, where the relevant degree between any two concepts in a fuzzy-valued concept network is represented by arbitrary shapes of fuzzy numbers. There are two kinds of relevant relationships between any two concepts in a fuzzy-valued concept network (i.e., fuzzy positive association and fuzzy negative association). In order to reduce the time of fuzzy inference, the relevant matrices and the relationship matrices are used to model the fuzzy-valued concept networks. The elements of a relevant matrix represent the relevant degrees between concepts. The elements of a relationship matrix represent the relevant relationships between concepts. Furthermore, we also allow users' queries to be represented by arbitrary shapes of fuzzy numbers and to use the fuzzy positive association relationship and the fuzzy negative association relationship for formulating their queries in order to increase the flexibility of fuzzy information retrieval systems. We also present a fuzzy information retrieval method based on the network-type fuzzy-valued concept network architecture in the Internet environment.

In this work, the authors propose a novel method to obtain correspondence between range data across image frames using neural like mechanisms. The method is computationally efficient and tolerant of noise and missing points. Elastic nets, which evolved out of research into mechanisms to establish ordered neural projections between structures of similar geometry, are used to cast correspondence as an optimization problem. This formulation is then used to obtain approximations to the motion parameters under the assumption of rigidity (inelasticity). These parameter scan be used to recover correspondence. Experimental results are presented to establish the veracity of the scheme and the method is compared to earlier attempts in this direction.

S. Ullman's classical (1979) results on motion/structure from orthographic views are revisited. Specifically, two results are presented: (1) two orthographic views allow an uncountably infinite number of solutions to motion/structure of a rigid body, and (2) a linear algorithm for solving motion/structure from four point correspondences over three views.>

The authors present a method which reduces an n-point problem to a set of four-point problems. The effort of reduction takes O(n) steps and it also takes O(n) steps to construct all possible mappings of an n-point set from the solution to a four-problem. Other results include (1) coplanarity condition of four points in two views, (2) recovering the tilt direction of the rotational axis using four points in two views, and (3) recovering the scaling factor.>

We present a method which uses both a planar patch and points for an analysis of time-varying imagery.From our theory.two pieces of information are required for the unique recovery of Structure from motion.One is the image motion of a plane and the other ODe is the image motion of points not on the plane.For the former.one needs four points while for the latter one need two points.Our technique has the following characteristics: (i) it is efficient, (ii) it appears to be very robust over a wide range of simulations, (iii) it sheds some new light on some previous work, and (iv) it has geometric meaning for each step of the derivation of the solution.

This report shows how highlights on three types of objects-cylinders, spheres, and generalised cylinders-can be used to provide constraints on their size and location. This information can then be used in conjunction with other constraints to solve the problem of object acquisition.