http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Combined Artificial Bee Colony for Data Clustering
Bum-Su Kang(강범수),Sung-Soo Kim(김성수) 한국산업경영시스템학회 2017 한국산업경영시스템학회지 Vol.40 No.4
Data clustering is one of the most difficult and challenging problems and can be formally considered as a particular kind of NP-hard grouping problems. The K-means algorithm is one of the most popular and widely used clustering method because it is easy to implement and very efficient. However, it has high possibility to trap in local optimum and high variation of solutions with different initials for the large data set. Therefore, we need study efficient computational intelligence method to find the global optimal solution in data clustering problem within limited computational time. The objective of this paper is to propose a combined artificial bee colony (CABC) with K-means for initialization and finalization to find optimal solution that is effective on data clustering optimization problem. The artificial bee colony (ABC) is an algorithm motivated by the intelligent behavior exhibited by honeybees when searching for food. The performance of ABC is better than or similar to other population-based algorithms with the added advantage of employing fewer control parameters. Our proposed CABC method is able to provide near optimal solution within reasonable time to balance the converged and diversified searches. In this paper, the experiment and analysis of clustering problems demonstrate that CABC is a competitive approach comparing to previous partitioning approaches in satisfactory results with respect to solution quality. We validate the performance of CABC using Iris, Wine, Glass, Vowel, and Cloud UCI machine learning repository datasets comparing to previous studies by experiment and analysis. Our proposed KABCK (K-means+ABC+K-means) is better than ABCK (ABC+K-means), KABC (K-means+ABC), ABC, and K-means in our simulations.
Efficient Data Clustering using Fast Choice for Number of Clusters
Sung-Soo Kim(김성수),Bum-Su Kang(강범수) 한국산업경영시스템학회 2018 한국산업경영시스템학회지 Vol.41 No.2
K-means algorithm is one of the most popular and widely used clustering method because it is easy to implement and very efficient. However, this method has the limitation to be used with fixed number of clusters because of only considering the intra-cluster distance to evaluate the data clustering solutions. Silhouette is useful and stable valid index to decide the data clustering solution with number of clusters to consider the intra and inter cluster distance for unsupervised data. However, this valid index has high computational burden because of considering quality measure for each data object. The objective of this paper is to propose the fast and simple speed-up method to overcome this limitation to use silhouette for the effective large-scale data clustering. In the first step, the proposed method calculates and saves the distance for each data once. In the second step, this distance matrix is used to calculate the relative distance rate (Vj) of each data j and this rate is used to choose the suitable number of clusters without much computation time. In the third step, the proposed efficient heuristic algorithm (Group search optimization, GSO, in this paper) can search the global optimum with saving computational capacity with good initial solutions using probabilistically for the data clustering. The performance of our proposed method is validated to save significantly computation time against the original silhouette only using Ruspini, Iris, Wine and Breast cancer in UCI machine learning repository datasets by experiment and analysis. Especially, the performance of our proposed method is much better than previous method for the larger size of data.
빠른 클러스터 개수 선정을 통한 효율적인 데이터 클러스터링 방법
Sung-Soo Kim,Bum-Su Kang 한국산업경영시스템학회 2018 한국산업경영시스템학회지 Vol.41 No.2
K-means algorithm is one of the most popular and widely used clustering method because it is easy to implement and very efficient. However, this method has the limitation to be used with fixed number of clusters because of only considering the intra-cluster distance to evaluate the data clustering solutions. Silhouette is useful and stable valid index to decide the data clustering solution with number of clusters to consider the intra and inter cluster distance for unsupervised data. However, this valid index has high computational burden because of considering quality measure for each data object. The objective of this paper is to propose the fast and simple speed-up method to overcome this limitation to use silhouette for the effective large-scale data clustering. In the first step, the proposed method calculates and saves the distance for each data once. In the second step, this distance matrix is used to calculate the relative distance rate (Vj) of each data j and this rate is used to choose the suitable number of clusters without much computation time. In the third step, the proposed efficient heuristic algorithm (Group search optimization, GSO, in this paper) can search the global optimum with saving computational capacity with good initial solutions using Vj probabilistically for the data clustering. The performance of our proposed method is validated to save significantly computation time against the original silhouette only using Ruspini, Iris, Wine and Breast cancer in UCI machine learning repository datasets by experiment and analysis. Especially, the performance of our proposed method is much better than previous method for the larger size of data.
A Phase 1 Study Using Autologous Natural Killer Cells in Patients with HAIC-Hepatocellular Carcinoma
( Sung Bum Cho ),( Chung Hwan Jun ),( Sung Kyu Choi ),( Woo Kyun Bae ),( Je Jung Lee ),( Yang Jun Kang ),( Cheol Kyun Cho ),( Yang Seok Ko ) 대한간학회 2018 춘·추계 학술대회 (KASL) Vol.2018 No.1
Aims: Natural killer (NK) cell-based immunotherapy has recently been tried with advances of understanding the role of immune defense against hepatocellular carcinoma (HCC). To improve NK cells therapy, we focused to increasing delivery of NK cells and synergic effect combined with hepatic arterial infusion chemotherapy (HAIC). Methods: We did a prospective, open label, phase 1 trial of the safety and efficacy of autologous NK cells through hepatic arterial infusion as sequential therapy after HAIC in advanced HCC patients. Between March 2016 and July 2017, 11 patients were included who showed favorable response more than stable disease (SD) after 2 sessions of HAIC in advanced HCC patients with child A. A total 4 sessions of HAIC were performed the protocols of infusion of cisplatin (25/m2) and 5-fluorouracil (750/m<sup>2</sup>) for 4 days every 3-4 weeks interval. The peripheral blood mononuclear cells of patients by leukapheresis were ob-tained after 3<sup>rd</sup> HAIC and NK cells were expanded for 2 weeks under Current Good Manufacturing Practices (cGMP). Patients received planned dosage of NK cells through chemoport into hepatic artery for 5 days after 4<sup>th</sup> HAIC (3 patients; 2.5x108, 3 patients; 5x10<sup>8</sup>, 5 patients; injection of 10x10<sup>8</sup> NK cells). The primary end point was safety of NK cell injection; secondary endpoint included objective response rate (modified Response Evaluation Criteria In Solid Tumors), time to progression, dura-tion of response and immunologic efficacy. Results: Any adverse events of NK cells injection were none according to dosage. An objective response was observed in 7 patients (63.6%) included three complete responses and four partial responses. Stable disease was observed in 2 patients and progressive disease was in 2 patients and thus disease control rate was 81.8%. The mean duration of time to progression was 9.7±5.3 month and duration of response without chemotherapy was 6.1±5.2 month. The newly metastatic lesion was occurred in 3 patents (27.2%; lymph node 1 patients, Lung 2 patients). Two patients were died by tumor progression and others were still alive. The increasing immunologic response was observed in 5 patients (55 %) to evaluate cytotoxicity and NK cell proportion of peripheral mononuclear cells after NK cell injection. Conclusions: The HAIC and NK cells immunotherapy is safe and effective treatment in the advance HCC patient with favorable liver function. The additional studies are urgently required to establish the new novel treatment.
Prevalence of Reflux Esophagitis During 12 Years in Daegu and Gyeongbuk Provinces
( Sung Jae Kim ),( Seok Keun Lee ),( Kang Wook Chung ),( Eun Soo Kim ),( Byoung Kuk Jang ),( Woo Jin Chung ),( Kwang Bum Cho ),( Jae Seok Hwang ),( Kyung Sik Park ) 대한소화기기능성질환·운동학회 2009 Journal of Neurogastroenterology and Motility (JNM Vol.15 No.2
Background/Aims: The prevalence of reflux esophagitis (RE) has been reported up to 12% in Korea, and seems being increased. However, epidemiological data for long-term trend are scarce. Moreover, there is no published report about the prevalence of RE in Daegu and Gyeongbuk provinces. Therefore, we analyzed the prevalence of RE during the recent 12 years in Daegu and Gyeongbuk provinces and also analyzed associated risk factors. Methods: We retrospectively reviewed 8,446 age and gender adjusted subjects who had visited a health promotion center for health check-up including esophagogastroduodenoscopy. Results: The mean age was 43.3±13.4 years. The overall prevalence of RE was 2.8%. The prevalences of RE in health check up subject from 1997 to 1999, from 2000 to 2002, from 2003 to 2005, and from 2006 to 2008 were 0.3%, 1.2%, 2.8%, and 6.8%, respectively. On the univariate analysis, male gender, high body mass index (BMI), obesity (BMI≥25 kg/m2), and hypertriglyceridemia were significantly related with presence of RE. On the multivariate analysis, male gender and obesity were independent risk factors of RE. Conclusion: The prevalence of RE is increasing recently in Daegu and Gyeongbuk provinces and male gender and obesity are independent risk factors of this disease.(Korean J Neurogastroenterol Motil 2009;15:124-129)
( Sung Hyub Han ),( Ji Won Byun ),( Won Soo Lee ),( Hoon Kang ),( Yong Chul Kye ),( Ki Ho Kim ),( Do Won Kim ),( Moon Bum Kim ),( Seong Jin Kim ),( Hyung Ok Kim ),( Woo Young Sim ),( Tae Young Yoon ) 대한피부과학회 2012 Annals of Dermatology Vol.24 No.3
Background: Androgenetic alopecia (AGA) is a common hair loss disease with genetic predisposition among men and women, and it may commence at any age after puberty. It may significantly affect a variety of psychological and social aspects of one`s life and the individual`s overall quality of life (QoL). Objective: This study aimed to investigate the QoL of AGA patients and discover the factors that can influence the QoL of AGA patients, including previous experience in non-medical hair care, reasons for hospital visits, age, duration, and the severity of AGA. Methods: A total of 998 male patients with AGA were interviewed, using the Hair Specific Skindex-29 to evaluate the QoL of AGA patients. Results: The results of the Hair Specific Skindex-29 on patients with AGA were as follows: symptom scale: 26.3±19.5, function scale: 24.0±20.1, emotion scale: 32.1±21.8, and global score: 27.3±19.1. According to this assessment, QoL was more damaged if the patient had severe alopecia, a longer duration of AGA, younger age, had received previous non-medical hair care, and visited the hospital for AGA treatment. Conclusion: This study showed that AGA could harmfully affect the patients` QoL. These findings indicate that dermatologists should address these QoL issues when treating patients with alopecia.