http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
The analysis of Japanese vowel by using Real-Time Spectral Analysis
Hideto Nakatsuji,Sigeru Omatu 한국멀티미디어학회 2009 한국멀티미디어학회 국제학술대회 Vol.2009 No.-
In recent year, wavelet transform has been paid attention as an analysis in time-frequency domain. Furthermore, a new analytical method with the same function of wavelet transform was presented by authors. In this method, the spectrum calculation is performed by only one input data and high-speed processing performed. In this paper, by using this analytical method for the analysis of the Japanese vowel sound, and we show how a Japanese vowel sound consists of it. There are five vowel sounds in Japanese, and these vowels consist of fundamental, harmonics and high frequency waves, and so we show how these waves affect the decision of the Japanese vowel sound.
Particle Swarm Optimization for Gene Selection Using Microarray Data
Mohd Saberi Mohamad,Sigeru Omatu,Safaai Deris,Michifumi Yoshioka 한국멀티미디어학회 2009 한국멀티미디어학회 국제학술대회 Vol.2009 No.-
In order to find and select possible informative genes for cancer classification, recently, many researchers are analyzing micro array data using various computational intelligence methods. However, due to a small number of samples compared to the huge number of genes, irrelevant genes, and noisy genes, most of these methods face difficulties to select the informative genes. In this paper, we propose an improved binary particle swarm optimization to select a small subset of informative genes that is relevant for the cancer classification. Instead of the existing rule of position update in binary particle swarm optimization (BPSO), we modify the rule so that it selects efficiently the small subset from the microarray data. By performing experiments on two different public cancer data sets, we have found that the performance of the proposed method is superior to other related previous works, including BPSO in terms of classification accuracy and the number of selected genes.
Tag Clustering with Social Bookmarking Data
Hidekazu Yanagimoto,Michifumi Yoshioka,Sigeru Omatu 한국멀티미디어학회 2009 한국멀티미디어학회 국제학술대회 Vol.2009 No.-
We propose a tag clustering with social bookmarking data in this paper. In social bookmarking services, various kinds of data, for example, text data, image data, and movie data, and users use tags, which are keywords added to registered web pages, to administer the registered web pages. However, the tags are not limited to a particular vocabulary and are added to the web pages freely. Hence, though the same tag is added to web pages, it might have a difference of meaning for each web page. Thesauruses are not useful to solve it because of including neologisms and symbols. Hence, our goal is to solve the ambiguity of tags and to classify them according to their meanings. To achieve the goal we regard adding a tag to a web page as a link between the tag and the web page and construct a weighted bipartite graph between tags and web pages without their contents. To classify the tags according to their meanings, Probabilistic Latent Semantic Indexing is used to analyze the weighted bipartite graph. We carried out evaluation experiments using real social bookmarking data, Buzzurl and confirmed the proposed method classifies the tags precisely regardless of the ambiguity of description and meaning.
Improvement of Face Detection Method using Haar-Like Features
Ryuji Tanaka,Michifumi Yoshioka,Sigeru Omatu 한국멀티미디어학회 2009 한국멀티미디어학회 국제학술대회 Vol.2009 No.-
Recently, utilization of the face information for human detection becomes the focus of attention in many situations. In any case, object detection system needs a fast and high-accuracy ability. Viola et al. introduced a rapid object detection system using a Haar-Like features based on rectangle features [1]. These features are so simple that they can apply everything to detection. However, these features take a large computational cost to detect complex objects. This paper has proposed adaptive Haar-Like features which are novel sets of features using target's texture. Experimental result showed that our method had less computational cost and higher performance than these of the conventional method.
Pooi San Chua,Abdul Hakim Mohamed Salleh,Mohd Saberi Mohamad,Safaai Deris,Sigeru Omatu,Michifumi Yoshioka 한국생물공학회 2015 Biotechnology and Bioprocess Engineering Vol.20 No.2
The current problem for metabolic engineering is how to identify a suitable set of genes for knockout that can improve the production of certain metabolites and sustain the growth rate from the thousands of metabolic networks which are complex and combinatorial. Some approaches, such as OptKnock and OptGene, are developed to enhance the production of desired metabolites. However, the performances of these approaches are suboptimal and the obtained results are unsatisfactory because of computational limitations such as local minima. In this paper, we propose a hybrid of Bat Algorithm and Flux Balance Analysis (BATFBA) to enhance succinate and lactate production by identifying a set of genes for knock out. The Bat Algorithm is an optimisation algorithm, whereas Flux Balance Analysis (FBA) is a mathematical approach to analyse the flow of metabolites through a metabolic network. The Escherichia coli iJR904 dataset was used to determine optimal knockout genes, production rate, and growth rate. By applying this hybrid method to the iJR904 dataset, we found that BATFBA yielded better results than existing methods, such as OptKnock and a hybrid of Artificial Bee Colony algorithms and Flux Balance Analysis (ABCFBA), at predicting succinate and lactate production.
Abdul Hakim Mohamed Salleh,Mohd Saberi Mohamad,Safaai Deris,Sigeru Omatu,Florentino Fdez-Riverola,Juan Manuel Corchado 한국생물공학회 2015 Biotechnology and Bioprocess Engineering Vol.20 No.4
The increasing demand of biochemical supply for various industries has spurred the development of metabolic engineering to find the optimal design of the microbial cell factories. Traditional method of chemical synthesis using the natural producer leads to the production far below their theoretical maximums. Gene knockout strategy is then introduced to improve the metabolite production. To aid the process, many computational algorithms have been developed to design the optimal microbial strain as cell factories to increase the production of the desired metabolite. However, due to the size of the genome scale model of the microbial strain, finding the optimal combination of genes to be knocked out is not an easy task. In this paper, we propose a hybrid of Genetic Ant Colony Optimization (GACO) and Flux Balance Analysis (FBA) namely GACOFBA to find the optimal gene knockout that increase the production of the target metabolite. Using E. coli and S. cerevisiae genome scale model, we test our proposed hybrid algorithm to increase the production of four different metabolites. By comparing with the results from existing method OptKnock as well as the conventional Ant Colony Optimization (ACO), the results show that our proposed hybrid algorithm able to identify the best set of genes and increase the production while maintaining the optimal growth rate.