• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• v.9 no.1
• /
• pp.453-456
• /
• 2005

In this paper, we propose method to improve recognition performance using the most effective algorithm selectively after clustering various face data, because recognition performance of each algorithm according to facial attribute is change. The proposed face recognition is divided into two steps. First step is the clustering integrated various data to be optimized in algorithm. Second is that classify input image by a similar cluster, select suitable algorithm and recognize the target. This thesis takes the first step towards the creation of a synthetic classifier fusiontesting environment. The effects of data correlation on three classifier fusion techniques were examined. We proposed fusion method for each recognition algorithm's result. This research explores how the degree of correlation in classification data affects the degree of accuracy in a fusion context.

• Journal of KIISE:Databases
• /
• v.32 no.6
• /
• pp.650-661
• /
• 2005

As the Internet is growing, the use of XML which is a standard of semi-structured document is increasing. Therefore, there are on going works about integration and retrieval of XML documents. However, the basis of efficient integration and retrieval of documents is to cluster XML documents with similar structure. The conventional XML clustering approaches use the hierarchical clustering algorithm that produces the demanded number of clusters through repeated merge, but it have some problems that it is difficult to compute the similarity between XML documents and it costs much time to compare similarity repeatedly. In order to address this problem, we use clustering algorithm for transactional data that is scale for large size of data. In this paper we use common structures from XML documents that don't have DTD or schema. In order to use common structures of XML document, we extract representative structures by decomposing the structure from a tree model expressing the XML document, and we perform clustering with the extracted structure. Besides, we show efficiency of proposed method by comparing and analyzing with the previous method.

• The KIPS Transactions:PartC
• /
• v.10C no.6
• /
• pp.665-674
• /
• 2003

In this paper, we propose an approach to correlate alerts using a clustering analysis of data mining techniques in order to support intrusion detection system. Intrusion detection techniques are still far from perfect. Current intrusion detection systems cannot fully detect novel attacks. However, intrucsion detection techniques are still far from perfect. Current intrusion detection systems cannot fully detect novel attacks or variations of known attacks without generating a large amount of false alerts. In addition, all the current intrusion detection systems focus on low-level attacks or anomalies. Consequently, the intrusion detection systems to underatand the intrusion behind the alerts and take appropriate actions. The clustering analysis groups data objects into clusters such that objects belonging to the same cluster are similar, while those belonging to different ones are dissimilar. As using clustering technique, we can analyze alert data efficiently and extract high-level knowledgy about attacks. Namely, it is possible to classify new type of alert as well as existed. And it helps to understand logical steps and strategies behind series of attacks using sequences of clusters, and can potentially be applied to predict attacks in progress.

• Korean Journal of Soil Science and Fertilizer
• /
• v.47 no.4
• /
• pp.275-283
• /
• 2014

The ${\gamma}$-oryzanol, ferulic acid esters, which are well-known for their function on cholesterol lowering and prevention of inflammation, diabetes and cancer, are found in the seeds of cereal crops such as rice, corn, wheat and rye. Among them, rice has been reported to contain the highest ferulic acid esters. Since rice cultivation with green manure as a N source is an environmental friendly agricultural practice, it is necessary to identify and quantify as well as evaluate the variations in these compounds in rice samples as affected by different green manure conditions. A total of ten components of ${\gamma}$-oryzanol were isolated and cycloartenyl ferulate, 24-methylenecycloartanyl ferulate, campesteryl ferulate and sitosteryl ferulate were identified as the major components in Korean rice cultivars, 'Unkwang' and 'Hopum'. Comparing the ${\gamma}$-oryzanol contents of these varieties, 'Unkwang' showed clearly similar pattern with conventional type. With the PLS-DA (partial least squares of discriminant analysis) using SIMCA 11.0 ver., the specific pattern and cluster of ${\gamma}$-oryzanol scores with green manure conditions were confirmed, and thus distinguishing green manure effects were possible.

• Journal of Korean Society of Transportation
• /
• v.27 no.5
• /
• pp.135-144
• /
• 2009

This study estimates the expected number of accidents in Kyungbuk Province to capitalize on experience gained from four years of accident history using the Empirical Bayes (EB) Method. The number of accidents of each site in Kyungbuk Province is recalculated using the Equivalent Property Damage Only (EPDO) method to reflect the severities of the accidents. A cluster analysis is performed to determine similar sites and a unique Safety Performance Function (SPF) is established for each site. The overdispersion parameter is built to correct the difference between the actual number of accidents and the underlying probability distribution. To adjust for varying traffic characteristics of each site, a relative weight is applied and eventually estimates the expected number of accidents. The results show that the highest accident sites are Kimcheon, Youngcheon, and Chilgok, but on the other hand the lowest is Gunwi.

• Journal of KIISE:Software and Applications
• /
• v.30 no.11
• /
• pp.1054-1061
• /
• 2003

Q-Learning, a representative algorithm of reinforcement learning, experiences repeatedly until estimation values about all state-action pairs of state space converge and achieve optimal policies. When the state space is high dimensional or continuous, complex reinforcement learning tasks involve very large state space and suffer from storing all individual state values in a single table. We introduce Q-Map that is new function approximation method to get classified policies. As an agent learns on-line, Q-Map groups states of similar situations and adapts to new experiences repeatedly. State-action pairs necessary for fine control are treated in the form of rule. As a result of experiment in maze environment and mountain car problem, we can achieve classified knowledge and extract easily rules from Q-Map

• Journal of The Korean Astronomical Society
• /
• v.35 no.3
• /
• pp.111-121
• /
• 2002

We have measured the correlation functions of the optically selected clusters of galaxies in the Abell and the APM catalogs, and of the X-ray clusters in the X-ray-Brightest Abell-type Clusters of galaxies (XBACs) catalog and the Brightest Clusters Sample (BCS). The same analysis method and the same method of characterizing the resulting correlation functions are applied to all observational samples. We have found that the amplitude of the correlation function of the APM clusters is much higher than what has been previously claimed, in particular for richer subsamples. The correlation length of the APM clusters with the richness R $\ge$ 70 (as defined by the APM team) is found to be $r_0 = 25.4_{-3.0}^{+3.1}\;h^{-1}$ Mpc. The amplitude of correlation function is about 2.4 times higher than that of Croft et al. (1997). The correlation lengths of the Abell clusters with the richness class RC $\ge$ 0 and 1 are measured to be $r_0 = 17.4_{-1.1}^{+1.2}$ and $21.0_{-2.8}^{+2.8}\;h^{-1}$ Mpc, respectively, which is consistent with our results for the APM sample at the similar level of richness. The richness dependence of cluster correlations is found to be $r_0= 0.40d_c + 3.2$ where $d_c$ is the mean intercluster separation. This is identical in slope with the Bahcall & West (1992)'s estimate, but is inconsistent with the weak dependence of Croft et al. (1997). The X-ray bright Abell clusters in the XBACs catalog and the X-ray selected clusters in the BCS catalog show strong clustering. The correlation length of the XBACs clusters with $L_x {\ge}0.65{\times} 10^{44}\;h^{-2}erg\;s^{-1}$ is $30.3_{-6.5}^{+8.2}\;h^{-1}$ Mpc, and that of the BCS clusters with $L_x {\ge}0.70{\times} 10^{44}\;h^{-2}erg\;s^{-1}$ is $30.2_{-8.9}^{+9.8}\;h^{-1}$ Mpc. The clustering strength of the X-ray clusters is much weaker than what is expected from the optical clusters.

• Bulletin of the Korean Chemical Society
• /
• v.14 no.1
• /
• pp.113-117
• /
• 1993

Good yield synthetic routes for the production of new B-alkyl-dithiaborane clusters are reported. The syntheses of the B-alkyl-dithiaboranes are based on the use of Fischer-type carbene reagents to activate the B-H bonds of dithiaborane for alkyl-addition reactions and are the first examples of transition-mediated reactions of dithiaborane to be reported. Thus, reactions employing arachno-$S_2B_7H_8$- and $(CO)_5M{C(R_1)R_2}$ (M = Cr, W; $R_1 = CH_3,\;C_6H_5;\; R_2 = OCH_3,\;SC_6H_5)$ were found to yield the intermidiate anions I, $[(CO)_5M{C(R_1)R_2S_2B_7H_8}]^-$, which upon protonation gave the corresponding neutral, air-sensitive cluster arachno-4-$RCH_2-6,8-S_2B_7H_8(R=CH_3,\;IIa;\;C_6H_5,\;IIb)$ range from 30 to 35% yield. Complexes IIa and IIb are isoelectronic with arachno-6,8-$S_2B_7H_9$ and, on the basis of the spectroscopic data, are proposed to adopt a similar arachno cage geometry in which an $RCH_2$ units are substituted to 4 position boron atom of the arachno-6,8-$S_2B_7H_9$.

• Journal of Radiation Industry
• /
• v.2 no.3
• /
• pp.121-128
• /
• 2008

Flavanone-3-hydroxylase (F3H) is one of the key enzymes for the biosynthesis of flavonals, anthocyanins, catechins and proanthocyanins. F3H catalyzes the $3{\beta}$-hydroxylation of (2S)-flavonones to form (2R, 3R)-dihydroflavonols. In this report, we isolated a full-length cDNA of RocF3H from black raspberry (Rubus occidentalis L.) using a reverse transcriptase-PCR and rapid amplification of the cDNA ends (RACE)-PCR. The full-length cDNA of RocF3H contains a 1,098 bp open reading frame (ORF) encoding a 365 amino acid protein with a calculated molecular weight of about 41.1 kDa and isoelectric point (pI) of 5.45. The genomic DNA analysis revealed that the RocF3H gene had three exons and two introns. Comparison of the deduced amino acid sequence of the RocF3H with other F3Hs revealed that the protein is highly homologous with various plant species. The conserved amino acids ligating the ferrous iron and the residues participating in the 2-oxoglutarate binding (R-X-S) were found in RocF3H at the similar positions to other F3Hs. Southern blot analysis indicated that RocF3H exist a multi-gene family. The isolation of RocF3H gene will be helpful to further study the role of F3H gene in the biosynthesis of flavonoids in R. occidnetalis.

• Korean Journal of Breeding Science
• /
• v.40 no.3
• /
• pp.243-249
• /
• 2008

In the previous study, 141 $BC_3F_2$ lines from a cross between the Oryza sativa cv. Milyang 23 and O. glaberrima were used to identify favorable wild QTL alleles for yield component traits. In this study, we carried out QTL analysis of four grain morphology as well as four yield component traits using 141 $BC_3F_5$ lines from the same cross and compared QTLs detected in two different generations. The mean number of O. glaberrima segments in the 141 $BC_3F_5$ lines ranged from 1 to 13 with 2.69 and 5.71 of the average means of homozygous and heterozygous segments, respectively. There was a three-fold difference in the number of QTLs detected for four traits commonly evaluated in two generations (seven QTLs in the $BC_3F_5$ vs 21 in the $BC_3F_2$ population). The percentages of the phenotypic variance explained by QTLs in the BC3F5 population were similar to or less than those in the $BC_3F_2$ population. This is probably due to the difference in the genetic composition of two populations and the environmental effects. The locations of the QTLs commonly detected in both generations were in good agreement except for one QTL for spikelets per panicle. The yield QTL, yd3 was colocalized with the spikelets per panicle, spp3. Yield increase at this locus is due to the increase in spikelets per panicle, because both traits were associated with increase in spikelets per panicle and yield due to the presence of an O. glaberrima allele. Clusters of QTLs for grain morphology traits were observed in two chromosome regions. One cluster harboring five QTLs near SSR markers RM106 and RM263 was detected on chromosome 2. This population would serve as a foundation for development of the introgression line population from a cross between Milyang 23 and O. glaberrima.