• KIISE Transactions on Computing Practices
• /
• v.23 no.4
• /
• pp.226-231
• /
• 2017

Word representation, an important area in natural language processing(NLP) used machine learning, is a method that represents a word not by text but by distinguishable symbol. Existing word embedding employed a large number of corpora to ensure that words are positioned nearby within text. However corpus-based word embedding needs several corpora because of the frequency of word occurrence and increased number of words. In this paper word embedding is done using dictionary definitions and semantic relationship information(hypernyms and antonyms). Words are trained using the feature mirror model(FMM), a modified Skip-Gram(Word2Vec). Sense similar words have similar vector. Furthermore, it was possible to distinguish vectors of antonym words.

• Journal of KIISE:Software and Applications
• /
• v.28 no.9
• /
• pp.681-687
• /
• 2001

In order to raise a class discrimination power by combining multiple classifiers under the Bayesian decision theory, the upper bound of a Bayes error rate bounded by the conditional entropy of a class variable and decision variables obtained from training data samples should be minimized. Wang and Wong proposed a tree dependence first-order approximation scheme of a high order probability distribution composed of the class and multiple feature pattern variables for minimizing the upper bound of the Bayes error rate. This paper presents an extended high order product approximation scheme dealing with higher order dependency more than the first-order tree dependence, based on the minimization of the upper bound of the Bayes error rate. Multiple recognizers for unconstrained handwritten numerals from CENPARMI were combined by the proposed approximation scheme using the Bayesian formalism, and the high recognition rates were obtained by them.

• The Journal of the Korea Contents Association
• /
• v.13 no.10
• /
• pp.584-593
• /
• 2013

The purpose of this research is to analyze students' perception of smart learning focusing on its definitions, roles and values in distance higher education. In the online survey, 1,950 students of 'A' open university were participated. The results show that the students viewed the smart learning to be more 'absorbing', 'interactive' and 'collaborative' than the existing e-learning, as it compiles their experiences into learning. However, the respondents' perceptions of smart learning varied among different age groups: more students in their 40s and 50s responded that smart learning was 'customized', 'humanlike', 'interactive', 'comfortable', 'stable', 'familiar', 'unstressful', and 'practical' than students in their 20s and 30s, and they tend to view the main feature of smart learning to be the compilation of learner experiences.

• Tunnel and Underground Space
• /
• v.13 no.4
• /
• pp.294-303
• /
• 2003

The structural behaviour of rock mass structure, such as tunnel or slope is critically dependent on the various characteristics of discontinuities. Therefore, it is important to survey and analyze discontinuities correctly for the design and construction of rock mass structure. One inevitable Procedure of discontinuity survey and analysis is joint set identification from a lot of raw directional joint data. The identification procedure is generally done by a graphical method. This type of analysis has some shortcomings such as subjective identification results, inability to use extra information on discontinuity, and so on. In this study, a computer program for joint set identification based on the fuzzy clustering algorithm was implemented and tested using two kinds of joint data. It was confirmed that fuzzy clustering method is effective and valid for joint set identification and estimation of mean direction and degree of clustering of huge joint data through the applications.

• The Korean Journal of Food And Nutrition
• /
• v.9 no.4
• /
• pp.404-408
• /
• 1996

One of the major objectives of the food industry is the enrichment of the functional properties and nutritional value of soybean protein. To attain this goal, an expression system of cDNA encoding native and protein-engineered soybean proteins in a microorganism must be developed and the function then ability of self-assembly and the functionalities of the expressed proteins should be evaluated before the modified genes are transfered to soybean plants. The pro-$\beta$-conglycinin synthesized in E. coli BL21(DE3) comprised approximately 20% of the total bacterial proteins and the expressed protein are formed soluble and trimer such as native protein in E. coli cells. The highly expressed protein was purified to homogeneity by salt precipitation with 20~40$ Ammonium sulfate ion-exchange chromatography with Q-Sepharose and hydrophobic column chromatography with Butyltoyopearl. Therefore, we concluded that the high-level expression system of $\beta$-conglycinin cDNA was established and a relatively simple and rapid method for purifying pro-$\beta$-conglicinin was also developed.

• Microbiology and Biotechnology Letters
• /
• v.13 no.1
• /
• pp.1-5
• /
• 1985

File fish fillets collected from markets were analyzed for their microbial load and microflora, resulting in the level of aerobic mesophiles ranged from $3.5{\times}10^3$ to $1.1{\times}10^8CFU/g$, total coliforms from less than 2.3 to $4.6{\times}10^5MPN/g$, fecal coliforms from 1ess than 2.3 to $1.1{\times}10^5MPN/g$, Enterobacteriaceae from $3{\times}10$ to $4.4{\times}10^5CFU/g$ and fecal Streptococci from $1.6{\times}10^2$ to $7.6{\times}10^6CFU/g$ having Staphylococcus hominis, Staphylococcus hemolyticus, Micrococcus varians, Micrococcus luteus and Streptococcus cremoris as constituent microorganisms of aerobic mesophiles, Escherichia coli and Proteus rettgeri of Family Enterobacteriaceae, and Streptococcus faecium and Streptococcus avium of fecal Streptococci.

• Applied Biological Chemistry
• /
• v.45 no.4
• /
• pp.190-194
• /
• 2002

Expression of Bacillus subtilis glutamyl-tRNA synthetase (GluRS) in Escherichia coli is lethal for the host, probably because this enzyme misaminoacylates ${tRNA_l}^{Gln}$ with glutamate in vivo. In order to overexpress B. subtilis GluRS, encoded by the gltX gene, in E. coli, this gene was amplified from B. subtilis 168 chromosomal DNA using PCR method and the entire coding region was cloned into a pET11a expression vector so that it was expressed under the control or the T7 Promoter. The resulting recombinant pEBER plasmid was transformed into E. coli Novablue (DE3) bearing the T7 RNA polymerase gene for expression. After IPTG treatment, the overproduced enzyme was purified using ammonium sulfate fractionation, Source Q anion exchange chromatography, Superdex-200 gel filtration, and Mono Q anion exchange chromatography. The purified enzyme yielded 18-fold increase in specific activity over the crude cell extract and its molecular weight was approximately 55 kDa on SDS-PAGE.

• The Korean Journal of Pesticide Science
• /
• v.5 no.4
• /
• pp.62-67
• /
• 2001

In order to investigate the harmful effects of Fusarium, producing mycotoxins, contaminated cereals, two months feeding study was carried out in SPF-ICR mice. Mice diets were incorporated with 30% Fusarium infected wheat, nepal barley or barley. The wheat was processed to flour. The nepal barley and barley were polished by 68% and 58%, respectively. The cereal incorporating amount in mice diet were 0, 10, 30 and 50% for each processed cereal. Five week-old mice were fed with the prepared diet for 2 months. The effects of Fusarium contaminated cereals on the mice were observed after the feeding. The rates of body weight gain, diet and water consumption were not changed. There were no significant changes on hematology, blood biochemistry, gross and histopathological evaluation, organ weights in all treatment groups. These results suggest that the diets have no deleterious effects to ICR mice.

• Applied Chemistry for Engineering
• /
• v.9 no.2
• /
• pp.159-164
• /
• 1998

Polycarbosilane(PCS) as a SiC precursor was synthesized from the rearrangement reaction of polydimethysilane(PDMS) in an autoclave, which prepared by dehalocoupling reaction of dichlorodimethylsilane. After fractional precipitation into three fractions in n-hexane-methanol mixture, they were characterized by FT-IR, NMR, GPC, TGA/DSC and XRD, and compared with the commercial product. We found that the molecular weight distributions of the PCS depended on the reaction pressures, temperatures and the reaction times, and affected thermal property and ceramic yield of the polymer. The monodispersed PCS containing less amount of oligomers and nonsoluble products was prepared by reaction of PDMS at $420^{\circ}C$ for 10 hrs, and it also gave the greatest amount of medium molecular weight($M_n=4,000$) fraction.

• Journal of KIISE:Computer Systems and Theory
• /
• v.35 no.8
• /
• pp.372-377
• /
• 2008

Multiclass cancer classification has been actively investigated based on gene expression profiles, where it determines the type of cancer by analyzing the large amount of gene expression data collected by the DNA microarray technology. Since gene expression data include many genes not related to a target cancer, it is required to select informative genes in order to obtain highly accurate classification. Conventional rank-based gene selection methods often use ideal marker genes basically devised for binary classification, so it is difficult to directly apply them to multiclass classification. In this paper, we propose a novel method for multiclass gene selection, which does not use ideal marker genes but directly analyzes the distribution of gene expression. It measures the class-discriminability by discretizing gene expression levels into several regions and analyzing the frequency of training samples for each region, and then classifies samples by using the naive Bayes classifier. We have demonstrated the usefulness of the proposed method for various representative benchmark datasets of multiclass cancer classification.