http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
오재응(Jae-Eung Oh),유국현(Kook Hyun Yoo),정운창(Un-Chang Jeong),김진수(Jin-Su Kim),노정준(Jeong-Joon Roh) 한국소음진동공학회 2015 한국소음진동공학회 학술대회논문집 Vol.2015 No.10
In the present study, refrigerant noise generated from an air-conditioning unit in operation was characterized as water, sha or gas sound by jury testing and was diagnosed and classified by logistic regression which was performed with objective sound quality parameters. A chi-square test was conducted to find parameters that influenced the probability for refrigerant noise to occur, and with the coefficient for each parameter, the study diagnosed the effects of objective sound quality parameters on the probability of refrigerant noise occurring. The normalization of units was carried out to identify the relative influence of each parameter on the probability of such noise occurring. Further, re-logistic regression was performed with parameters selected by the chi-square test. Probability-based optimal cutoff values were determined for the classification of water, sha and gas sounds. Air-conditioning refrigerant noise was definitely classified by taking into account the logistic regression and cutoffs. New experiments on the generation of refrigerant noise were conducted to validate the logistic regression classification. Data obtained from the experiments was classified at an accuracy level of 93.9 percent.
감성분석과 Word2vec을 이용한 비정형 품질 데이터 분석
이진욱 ( Chinuk Lee ),유국현 ( Kook Hyun Yoo ),문병민 ( Byeong Min Mun ),배석주 ( Suk Joo Bae ) 한국품질경영학회 2017 품질경영학회지 Vol.45 No.1
Purpose: This study analyzes automobile quality review data to develop alternative analytical method of informal data. Existing methods to analyze informal data are based mainly on the frequency of informal data, however, this research tries to use correlation information of each informal data. Method: After sentimental analysis to acquire the user information for automobile products, three classification methods, that is, naive Bayes, random forest, and support vector machine, were employed to accurately classify the informal user opinions with respect to automobile qualities. Additionally, Word2vec was applied to discover correlated information about informal data. Result: As applicative results of three classification methods, random forest method shows most effective results compared to the other classification methods. Word2vec method manages to discover closest relevant data with automobile components. Conclusion: The proposed method shows its effectiveness in terms of accuracy and sensitivity on the analysis of informal quality data, however, only two sentiments (positive or negative) can be categorized due to human errors. Further studies are required to derive more sentiments to accurately classify informal quality data. Word2vec method also shows comparative results to discover the relevance of components precisely.