A Fuzzy Neural Network Model of Linguistic Dynamic Systems Based on Computing with Words

Linguistic dynamic systems(LDS)are dynamic processes involving computing with words(CW)for modeling and analysis of complex systems.In this paper,a fuzzy neural network(FNN)structure of LDS was proposed.In addition,an improved nonlinear particle swarm optimization was employed for training FNN.The experiment results on logistics formulation demonstrates the feasibility and the efficiency of this FNN model.

Auto Composition System for Two Voice Part Inventions Based on Soft Computational Methods

Algorithmic composition is a very popular research field today. Bach's "two voice part invention" is the research object in this paper. The grammar and compositional rules of "invention" are introduced first. Then two soft computational methods,genetic algorithms and back propagation (BP) neural network technology,are combined to the experiment on assisting in composing "two voice part inventions". The system presented in this paper is quite effective and satisfactory.

Automatic Partition of Chinese Sentence Group

Automatic partition of Chinese sentence group is very important to the statistical machine translation system based on discourse. This paper presents an approach to this issue： first, each sentence in a discourse is expressed as a feature vector; second, a special hierarchical clustering algorithm is applied to present a discourse as a sentence group tree. In this paper, local reoccurrence measure is proposed to the selection of key phras and the evaluation of the weight of key phrases. Experimental results show our approach promising.

Vitality Condition Judgment through Eye Quantification in Traditional Chinese Medicine

There are four main vitality states of human body in Traditional Chinese Medicine (TCM). Eye is the most important part on face for TCM doctors to diagnose patient. In this paper,we present several methods of eye quantification. We quantify eye movement,venation in white part of eye,tears in eye and the shape of upper eyelid,and then use cloud model to deduce volunteers' vitality condition by these measurements. We collect more than one thousand face images and dozens of face videos of patients',and some rules we mined with the highest support and interesting scores are right correspond to TCM diagnosis.

An Automatic Segmentation of Kidney in Serial Abdominal CT Scans Using Region Growing Approach

Automatic kidney segmentation from abdominal CT images is a key step in computer-aided diagnosis for kidney CT as well as computeraided surgery. However, kidney segmentation from CT images is generally performed manually or semi-autornatically because of gray levels similarities of adjacent organs/tissues in abdominal CT images. This paper presents an efficient algorithm for segmenting kidney from serials of abdominal CT images. First, we extracted estimated kidney position （EKP） according to the statistical geometric location of kidney within the abdomen. Second, we analyzed the intensity distribution of EKP for several abdominal CT images and exploit an adaptive threshold searching algorithm to eliminate many other organs/tissues in the EKP. Finally, a novel region growing approach based on labeling is used to obtain the fine kidney regions. Experimental results are comparable to those of manual tracing radiologist and shown to be efficient.

Gender Recognition with Face Images Based on Partially Connected Neural Evolutionary

In this paper,a new type of neural network model - Partially Connected Neural Evolutionary (PARCONE) was introduced to recognize a face gender. The neural network has a mesh structure in which each neuron didn't connect to all other neurons but maintain a fixed number of connections with other neurons. In training,the evolutionary computation method was used to improve the neural network performance by change the connection neurons and its connection weights. With this new model,no feature extraction is needed and all of the pixels of a sample image can be used as the inputs of the neural network. The gender recognition experiment was made on 490 face images (245 females and 245 males from Color FERET database),which include not only frontal faces but also the faces rotated from-40°-40° in the direction of horizontal. After 300-600 generations' evolution,the gender recognition rate,rejection rate and error rate of the positive examples respectively are 96.2%,1.1%,and 2.7%. Furthermore,a large-scale GPU parallel computing method was used to accelerate neural network training. The experimental results show that the new neural model has a better pattern recognition ability and may be applied to many other pattern recognitions which need a large amount of input information.

A New Word Detection Method for Chinese Based on Local Context Information

Finding out out-of-vocabulary words is an urgent and difficult task in Chinese words segmentation. To avoid the defect causing by offline training in the traditional method, the paper proposes an improved prediction by partical match （PPM） segmenting algorithm for Chinese words based on extracting local context information, which adds the context information of the testing text into the local PPM statistical model so as to guide the detection of new words. The algorithm focuses on the process of online segmentatien and new word detection which achieves a good effect in the close or opening test, and outperforms some well-known Chinese segmentation system to a certain extent.

Song Ci Style Automatic Identification

To identify Song Ci style automatically,we put forward a novel stylistic text categorization approach based on words and their semantic in this paper. And a modified special word segmentation method,a new semantic relativity computing method based on HowNet along with the corresponding word sense disambiguation method are proposed to extract words and semantic features from Song Ci. Experiments are carried out and the results show that these methods are effective.

LGAA: a Lattice-Gas Automata with Aggregation for Composing Guqin Music

Guqin music has been viewed as the symbol of Chinese music. Using artificial intelligence approaches to study Guqin music's composition will have an important theoretical and practical value. For the characteristics of Guqin tablature, a new model of lattice-gas automata with aggregation (LGAA) was constructed to generate melody, based on the theory of lattice-gas cellular automata (LGCA) and diffusion limited aggregation (DLA). Firstly, music segments were composed by the model of LGAA based on an emotional database. Then, based on the same pitch database, they were made smoother by a balance principle, which was followed by almost all Chinese traditional music. After that, composition music could be regarded as a knapsack problem, and the smooth music segments were seemed as the items. Therefore, the generation of the Guqin piece equaled the optimal solution to the knapsack problem. In the end, five musicians were invited to judge the results by two criteria and they all agreed that the automatic generated pieces of Guqin were success ful .

Improving Phrase-Based Statistical Machine Translation Models by Incorporating Syntax-Based Language Models

This paper proposed a method to incorporate syntax-based language models in phrase-based statistical machine translation (SMT) systems. The syntax-based language model used in this paper is based on link grammar,which is a high lexical formalism. In order to apply language models based on link grammar in phrase-based models,the concept of linked phrases,an extension of the concept of traditional phrases in phrase-based models was brought out. Experiments were conducted and the results showed that the use of syntax-based language models could improve the performance of the phrase-based models greatly.

Recognition of Chinese Organization Name Using Co-training

Chinese organization name recognition is hard and important in natural language processing. To reduce tagged corpus and use untagged corpus,we presented combing Co-training with support vector machines (SVM) and conditional random fields (CRF) to improve recognition results. Based on principles of uncorrelated and compatible,we constructed different classifiers from different views within SVM or CRF alone and combination of these two models. And we modified a heuristic untagged samples selection algorithm to reduce time complexity. Experimental results show that under the same tagged data,Co-training has 10% F-measure higher than using SVM or CRF alone; under the same F-measure,Co-training saves at most 70% of tagged data to achieve the same performance.

A Compression Algorithm of Hyperspectral Remote Sensing Image Based on 3-D Wavelet-Fractal Coder

In this paper, the 3-D Wavelet-Fractal coder was used to compress the hyperspectral remote sensing image, which is a combination of 3-D improved set partitioning in hierarchical trees （SPIHT） coding and 3-D fractal coding. Hyperspectral image date cube was first translated by 3-D wavelet and the 3-D fractal compression ceding was applied to lowest frequency subband. The remaining coefficients of higher frequency sub-bands were encoding by 3-D improved SPIHT. We used the block set instead of the hierarchical trees to enhance SPIHT＇s flexibility. The classical eight kinds of affme transformations in 2-D fractal image compression were generalized to nineteen for the 3-D fractal image compression. The new compression method had been tested on MATLAB. The experiment results indicate that we can gain high compression ratios and the information loss is acceptable.

Action Recognition from Videos with Complex Background via Transfer Learning

Classifier learning methods commonly assume that the training data and the testing data are drawn from the same underlying distribution. However, in many practical situations, this assumption is violated. One example is the practical action videos with complex background and the universal human action databases of Kangliga Tekniska Hogskolan （KTH）. When training data are very scarce, supervised learning is difficult. However, it will cost lots of human and material resources to establish a labeled video set which includes a large amount of videos with complex backgrounds. In this paper, we propose an action recognition framework which uses transfer boosting learning algorithm. By using this algorithm, we can train an action recognition model fitting for most practical situations just relaying on the universal action video dataset and a tiny set of action videos with complex background. And the experiment results show that the performance is improved.

Mining User Role in Social Community Application of Web 2.0

With the development of web 2.0, more and more social community applications appeared. The classical type of this kind of application is blog and facebook. The most important feature of these applications is that it is a self-media and users can post their own ideas in Internet. By using these social community applications, a big social network is formed. To study the feature of social network, it is important to mine the individual information at the beginning. In this paper, we propose a User Role based method to mine the relation between the user and object thing. First, we extract the User Role from the semantic dictionary Wordnet. Then, the feature of User Role is also mined by considering the hypemymy and hyponymy relation. Finally, we can use these features to deduce the User Role. In our experiments, we use a big corpus from TREC 2006 to test the mining performance. The experiment results show that the User Role effectively explores the feature of user.

Research on the Cognitive Comprehension Logic and Its Application in Understanding of Metaphor

In this article,a novel logic which concerns with the natural property of comprehension is presented,and the cognitive state of the agent is also considered. The cognitive comprehension operator is put forward,the logical system is established,and then the axioms and the properties of the system are discussed. Finally,the application of the cognitive comprehension logic in the understanding of the metaphor is showed.

Interframe Variation Vector:A Novel Feature for Gait Recognition

Gait representation is an important issue in gait recognition. A simple yet efficient approach, called Interframe Variation Vector （IW）, is proposed. IW considers the spatiotemporal motion characteristic of gait, and uses the shape variation information between successive frames to represent gait signature. Different from other features, IVV rather than condenses a gait sequence into single image resulting in spatial sequence lost; it records the whole moving process in an IVV sequence. IVV can encode whole essential features of gait and preserve all the movements of limbs. Experimental results show that the proposed gait representation has a promising recognition performance.

A classification-based method to estimate event-related potentials from single trial EEG

A novel method based on machine learning is developed to estimate event-related potentials from single trial electroencephalography. This paper builds a basic framework using classification and an optimization model based on this framework for estimating event-related potentials. Then the SingleTrialEM algorithm is derived by introducing a logistic regression model, which could be obtained by training before SingleTrialEM is used, to instantiate the optimization model. The simulation tests demonstrate that the proposed method is correct and solid. The advantage of this method is verified by the comparison between this method and the Woody filter in simulation tests. Also, the cognitive test results are consistent with the conclusions of cognitive science.

Harmonizing Melody with Meta-Structure of Piano Accompaniment Figure

In this paper, a meta-structure of piano accompaniment figure （meta-structure for short） is proposed to harmonize a melodic piece of music so as to construct a multi-voice music. Here we approach melody harmonization with piano accompaniment as a machine learning task in a probabilistic framework. A series of piano accompaniment figures are collected from the massive existing sample scores and converted into a set of meta-structure. After the procedure of samples training, a model is formulated to generate a proper piano accompaniment figure for a harmonizing unit in the context. This model is flexible in harmonizing a melody with piano accompaniment. The experimental results are evaluated and discussed.

Survey of visual sentiment prediction for social media analysis

Recent years have witnessed a rapid spread of multi-modality microblogs like Twitter and Sina Weibo composed of image, text and emoticon. Visual sentiment prediction of such microblog based social media has recently attracted ever-increasing research focus with broad application prospect. In this paper, we give a systematic review of the recent advances and cutting-edge techniques for visual senti- ment analysis. To this end, in this paper we review the most recent works in this topic, in which detailed comparison as well as experimental evaluation are given over the cuttingedge methods. We further reveal and discuss the future trends and potential directions for visual sentiment prediction.