http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
정홍 啓明大學校 産業技術硏究所 1997 産業技術硏究所 論文報告集 Vol.20 No.-
This paper studies a method for making efficient classification rules in the ID3 using the rough set theory. Decision tree technique of the ID3 always uses all the attributes in a table of examples for making a new decision tree, but rough set technique can in advance eleminate dispensable attributes. And the former generates only one type of classification rules, but the latter generates all the possibles types of them. The rules generated by the rough set technique are the simplist form as proved by the rough set theory. Therefore, ID3, applying the rough set technique, can reduct the size of the table of examples, generate the simplist form of the classification rules, and also implement an effective classification system.
Generation of Approximation Rules Using Information Gain
정홍 啓明大學校 産業技術硏究所 1999 産業技術硏究所 論文報告集 Vol.22 No.2
This paper suggests a method for generating approximation rules by the information gain used in the machine learning by decision trees. We studied that these rules are better than other approximation rules induced by using x²goodness of fittest of dependency of attributes in rough set theory by an experiment using neural network.
음경해면체신경 절단 백서의 음경 조직에서 단백 발현 변화의 단백체학적 접근
정홍,이창권,김보경,김홍섭,김동욱,백성현,전현수,양상국 대한비뇨의학회 2009 Investigative and Clinical Urology Vol.50 No.5
Purpose: Cavernous nerve resection (CNR) in rats is a standard model of animal experiments on erectile dysfunction (ED) that occurs after radical prostatectomy (RP). Injured cavernous nerves after surgery can cause fibrosis and apoptosis that lead to penile structural changes that may be accompanied by alterations of protein expression. This study aimed to analyze the changes in protein after CNR in Wistar Kyoto rats. Materials and Methods: Using 8-week-old male Wistar Kyoto rats, sham and CNR operation under a microscope were performed. Two and 8 weeks after surgery, we applied 2-DE and MALDI-TOF/TOF (AB 4700) to identify differently expressed penile proteins after CNR. 2-DE gels were stained with silver nitrate and were analyzed with PDQuest. After in-gel digestion, peptide mass spectra were obtained by MALDI-TOF/TOF mass spectrometry in the positive ion reflector mode. The obtained data were screened with a rat database from both the NCBI and the Swiss-Prot/TrFMBL home page. Results: The proteins that were changed more than 1.5-fold compared with the sham group were annexin A4 and pyruvate kinase (PK). Annexin A4 was increased by 1.75-fold after 2 weeks, whereas PK was decreased by 4.16 after 8 weeks. These results were confirmed by immunohistochemistry. Conclusions: Annexin A4 in the CNR group was increased, which may be related to emiocytosis during apoptosis. The decrease in PK of the CNR group is assumed to be related to a decrease in efficacy during glycolysis. Further study will be needed to elucidate the molecular pathophysiology of ED after cavernous nerve injury. Purpose: Cavernous nerve resection (CNR) in rats is a standard model of animal experiments on erectile dysfunction (ED) that occurs after radical prostatectomy (RP). Injured cavernous nerves after surgery can cause fibrosis and apoptosis that lead to penile structural changes that may be accompanied by alterations of protein expression. This study aimed to analyze the changes in protein after CNR in Wistar Kyoto rats. Materials and Methods: Using 8-week-old male Wistar Kyoto rats, sham and CNR operation under a microscope were performed. Two and 8 weeks after surgery, we applied 2-DE and MALDI-TOF/TOF (AB 4700) to identify differently expressed penile proteins after CNR. 2-DE gels were stained with silver nitrate and were analyzed with PDQuest. After in-gel digestion, peptide mass spectra were obtained by MALDI-TOF/TOF mass spectrometry in the positive ion reflector mode. The obtained data were screened with a rat database from both the NCBI and the Swiss-Prot/TrFMBL home page. Results: The proteins that were changed more than 1.5-fold compared with the sham group were annexin A4 and pyruvate kinase (PK). Annexin A4 was increased by 1.75-fold after 2 weeks, whereas PK was decreased by 4.16 after 8 weeks. These results were confirmed by immunohistochemistry. Conclusions: Annexin A4 in the CNR group was increased, which may be related to emiocytosis during apoptosis. The decrease in PK of the CNR group is assumed to be related to a decrease in efficacy during glycolysis. Further study will be needed to elucidate the molecular pathophysiology of ED after cavernous nerve injury.
연속어 인식 및 형태소 태깅을 위한 재귀적 알고리즘에 관한 연구
정홍,박기윤,허희수 한국뇌학회 2003 한국뇌학회지 Vol.3 No.1
자모의 삽입/탈락/치환, 맞춤법에 어긋난 어절 경계 등으로 나타나는 입력오류에 강인한 형태소 분석과 태깅을 통합한 방법에 관한 방법을 제시하며 음성 인식시스템과 연동 방안을 모색한다. We present a integrated approach for morphological analysis and tagging for Korean spoken language. The method is robust for alphabet insertion, deletion, substitution and wrong boundary of eojeol according to spelling rules. It can link with speech recognition system for post-processing.
鄭洪 啓明大學校 産業技術硏究所 1983 産業技術硏究所 論文報告集 Vol.5 No.-
In accordance with the expansion of the computer utilization for effective administration in various organizations, centralization of data at the computer site is inevitable. Therefore, there occur some important problems on how to maintain data file prevention or data security protection ,and these problems are directly related to the operation of the computer center. This paper deals with the management activities for data file prevention or data security protection.
정홍 啓明大學校 産業技術硏究所 1997 産業技術硏究所 論文報告集 Vol.19 No.2
This study adopts the rough set theory to analysis the attribute dependency among decision rules, and to induce the reducts of minimal subsets of the decision rules. The methods eliminates dispensable attributes in the simplification process and removes irrelevant attributes in the reduction process. The rules generated are concise, expressive and strong because it is in the most reduced form. The paper, using the real data of a chemical process, first finds reducts of condition attributes. removes duplicate decision rules and then finds value-reducts of condition attributes and again, if necessary, eleminates redundant decision rules. This method leads to a simple algorithm for decision table simplification or generation of minimal decision algorithm , which outperform other methods in term of achievable degree in the number of conditions, and it gives all possible solution to the problem.