• 한국융합학회논문지
• /
• 제10권11호
• /
• pp.159-164
• /
• 2019

ICT 기술 발전으로 강아지와 고양이등 반려동물 돌보기와 건강에 대한 관심도가 높아지고 있다. 본 논문에서는 반려동물 산업의 다양한 분야에 활용될 수 있도록 반려동물 사료의 성분 데이터를 기반으로 군집분석을 수행하고 적합한 서비스에 대해 고찰한다. 군집분석을 위해 시중에서 유통되고 있는 300여 개의 강아지 및 고양이 펫푸드를 대상으로 성분별 상관관계를 분석하여 유사성을 측정하며, Hierarchical, K-Means, Partitioning around medoids(PAM), Density-based, Mean-Shift 등의 다양한 클러스터링 기법을 활용하여 군집화 하여 분석한다. 또한 반려동물의 개인화 추천시스템도 제안한다. 본 논문의 연구 결과는 반려동물을 대상으로 한 사료 추천시스템 등의 맞춤형 개인화 서비스에 활용할 수 있다.

• 한국도로학회논문집
• /
• 제7권4호
• /
• pp.211-219
• /
• 2005

교통사고 분석구간 선정은 유사한 사고유형을 하나의 동일집단으로 처리함으로써 분석의 효율성을 높이고, 효과적인 개선 사업을 위해 그 순위를 결정하는데 필요한 작업이다. 기존에는 도로이정에 기초하여 균등하게 분할하는 방법을 주로 사용하여 왔는데 사고간의 유사성을 전혀 고려하지 못하는 단점이 있다. 따라서 최근 도로이정보다는 사고간의 유사성을 고려하여 구간을 선정하는 방식인 Slider length 적용기법이 사용되고 있다. 본 연구에서는 Slider length 적용기법의 한 방법론으로써, 군집분석에 사용되는 비계층적 분류기법인 K-평균법을 사용하여 가장 인근거리에 발생된 사고들이 최대한 하나의 집단으로 분류될 수 있는 기법을 제시하고자 한다. 또한 이의 검증을 위해 경부고속도로 부산방향으로 총연장 25.6km구간에 대하여 균일간격에 의한 분할방식과 K-평균법을 이용한 분할방식에 대해 상호 비교를 통해 그 효율성을 검토하였으며, K-평균법이 사고의 유사성이나 인접성을 감안하여 효율적으로 분석구간 선정에 적용될 수 있음을 검증하였다.

• 천문학회보
• /
• 제35권1호
• /
• pp.53.2-53.2
• /
• 2010

Using the JHK 2MASS near-infrared data, we present photometry results for the old open cluster (OC) Trumpler 5 (Tr 5). From the near-infrared color-magnitude diagrams of Tr 5, we have located the position of the red giant clump (RGC) stars, and used the mean magnitude of the RGC stars in K-band to estimate the distance to Tr 5. From the Padova isochrone fittings, we have estimated the reddening, metallicity, and age of Tr 5. These parameters generally agree well with those obtained from the previous studies on Tr 5 and confirms that this cluster is an old OC with metallicity being metal-poorer than solar abundance, located in the anti-Galactic center region.

• 한국전기전자재료학회논문지
• /
• 제20권11호
• /
• pp.959-964
• /
• 2007

To investigate the influence of partial discharge(PD) distribution characteristics due to various defects on the power cable joints interface, we used the K-means clustering method. As the result of PD number(n) distribution analyzing on $\Phi-n$ graph, the phase angle($\Phi$) of cluster centroid shifted to $0^{\circ}\;and\;180^{\circ}$ increasing with applying voltage. It was confirmed that the PD quantify(q) and euclidean distance of centroid were increased with applying voltage from the centroid distribution analyzing of $\Phi-q$ plane. The dispersion degree was increased with calculated standard deviation of the $\Phi-q$ cluster centroid. The PD number and mean value on $\Phi-q$ graph were some different by electric field concentration with defect types.

• Asian-Australasian Journal of Animal Sciences
• /
• 제17권8호
• /
• pp.1047-1052
• /
• 2004

The genomes of Chinese native chicken populations were screened using microsatellites as molecular markers. A total of, 528 individuals comprisede12 Chinese native chicken populations were typed for 7 microsatellite markers covering 5 linkage groups and genetic variations and genetic distances were also determined. In the 7 microsatellite loci, the number of alleles ranged from 2 to 7 per locus and the mean number of alleles was 4.6 per locus. By using fuzzy cluster, 12 Chinese native chicken populations were divided into three clusters. The first cluster comprised Taihe Silkies, Henan Game Chicken, Langshan Chicken, Dagu Chicken, Xiaoshan Chicken, Beijing Fatty Chicken and Luyuan Chicken. The second cluster included Chahua Chicken, Tibetan Chicken, Xianju Chicken and Baier Chicken. Gushi Chicken formed a separate cluster and demonstrated a long distance when comparing with other chicken populations.

• 한국환경복원기술학회지
• /
• 제13권3호
• /
• pp.36-45
• /
• 2010

To analyze the relationship between climatic factors (monthly mean temperature and total precipitation) and tree-ring growths of Quercus mongolica Fischer (Mongolian oak) with different topographic sites in Mt. Worak, more than 10 trees were selected from each of seven stands. Two cores from each tree were measured for ring width. After crossdating, each ring-width series was double standardized by fitting first a negative exponential or straight regression line and secondly a 60-year cubic spline. Seven stands were categorized in two groups using cluster analysis for tree-ring index patterns. Cluster I (four stands) was located in higher elevation (550-812 m) with aspects of east, west and northwest, and cluster II (three stands) was located in rather lower election (330-628 m) with aspects of north and northwest. The aspects of two clusters were not significantly different. Response-function analysis showed a significant positive response to March precipitation for both clusters. It indicates that moisture supply during early spring season is important to radial growth because the cambial growths of ring-porous species, such as Mongolian oak, start before leaf growth. Cluster II showed a positive response to the precipitation of middle and late growing season, too.

• Asia pacific journal of information systems
• /
• 제25권4호
• /
• pp.686-711
• /
• 2015

The performance of a novel medical question-answering engine called CliniCluster and existing search engines, such as CQA-1.0, Google, and Google Scholar, was evaluated using known-item searching. Known-item searching is a document that has been critically appraised to be highly relevant to a therapy question. Results show that, using CliniCluster, known-items were retrieved on average at rank 2 ($MRR@10{\approx}0.50$), and most of the known-items could be identified from the top-10 document lists. In response to ill-defined questions, the known-items were ranked lower by CliniCluster and CQA-1.0, whereas for Google and Google Scholar, significant difference in ranking was not found between well- and ill-defined questions. Less than 40% of the known-items could be identified from the top-10 documents retrieved by CQA-1.0, Google, and Google Scholar. An analysis of the top-ranked documents by strength of evidence revealed that CliniCluster outperformed other search engines by providing a higher number of recent publications with the highest study design. In conclusion, the overall results support the use of CliniCluster in answering therapy questions by ranking highly relevant documents in the top positions of the search results.

• International journal of advanced smart convergence
• /
• 제4권1호
• /
• pp.93-98
• /
• 2015

Recently, Radon has been gotten attention for problems of Nuclear Generating Station and a variety of nuclear. It is naturally arises that is accumulated in the interior through the soil with radioactive materials. People exposed to indoor a Radon increase the high risks of lung cancer. The data are consisted of regional Country, The Location, Average Radon pCi/L, Geo Mean and Geo S.D etc. The research is experimented using E-M algorithm. The research result appears to make a division of soil distance, regional and cluster. It requires in effort to minimize exposure to people who live in areas with high radon levels. A country must apprise to people about Radon risk and needs to work out measures plan.

• 한국전기전자재료학회:학술대회논문집
• /
• 한국전기전자재료학회 2002년도 하계학술대회 논문집
• /
• pp.232-235
• /
• 2002

An 150 kV gas cluster ion accelerator was fabricated and assessed. The change of surface morphology and surface roughness were examined by an atom force microscope (AFM) after irradiation of $CO_2$ gas clusters on Si (100) surfaces at the acceleration voltages of 50 kV. The density of hillocks induced by cluster ion impact was gradually increased with the dosage up to 5$\times$10$^{11}$ ions/$\textrm{cm}^2$. At the boundary of the ion dosage of 10$^{12}$ ions/$\textrm{cm}^2$, the density of the induced hillocks was decreased and RMS (root mean square) surface roughness was not deteriorated further. At the dosage of 5x10$^{13}$ ions/$\textrm{cm}^2$, the induced hillocks completely disappeared and the surface became very flat. In addition, the irradiated region was sputtered. $CO_2$ cluster ions are irradiated at the acceleration voltage of 25 kV to remove hillocks on indium tin oxide (ITO) surface and thus to attain highly smooth surfaces. $CO_2$ monomer ions are also bombarded on the ITO surface at the same acceleration voltage to compare sputtering phenomena. From the AFM results, the irradiation of monomer ions make the hillocks sharper and the surfaces rougher On the other hand, the irradiation of $CO_2$ cluster ions reduces the hight of hillocks and planarize the ITO surfaces. From the experiment of isolated cluster ion impact on the Si surfaces, the induced hillocks m high had the surfaces embossed at the lower ion dosages. The surface roughness was slightly increased with the hillock density and the ion dosage. At higher than a critical ion dosage, the induced hillocks were sputtered and the sputtered particles migrated in order to fill valleys among the hillocks. After prolonged irradiation of cluster ions, the irradiated region was very flat and etched.

• 한국멀티미디어학회논문지
• /
• 제21권7호
• /
• pp.779-786
• /
• 2018

In this paper, we are applied a nerve network to allow for the reflection of data learning methods in their overall forms by using cluster data rather than data learning by the stages and then selected a nerve network model and analyzed its variables through learning by the cluster. The CkLR algorithm was proposed to analyze the reaction variables of clustering outcomes through an approach to the initialization of K-means clustering and build a model to assess the prediction rate of clustering and the accuracy rate of prediction in case of new data inputs. The performance evaluation results show that the accuracy rate of test data by the class was over 92%, which was the mean accuracy rate of the entire test data, thus confirming the advantages of a specialized structure found in the proposed learning nerve network by the class.