• The Journal of the Petrological Society of Korea
• /
• v.6 no.1
• /
• pp.19-33
• /
• 1997

This study reports the results about the petrography and geochemical characteristics of 10 representative volacanic rocks. The Cretaceous volcanic rocks distributed in the vicinity of the Kageo island composed of andesitic rocks, dacitic welded tuff, and rhyolitic rocks in ascending order. Sedimentary rock is the basement in the study area covered with volcanic rocks. Andesitic rocks composed of pyroclastic volcanic breccia, lithic lapilli tuff and cryptocrystallin lava-flow. Most dacitic rocks are lapilli ash-flow welded tuff. Rhyolitic rocks consists of rhyolite tuff and rhyolite lava flow. Rhyolite tuff are lithic crystal ash-flow tuff and crystal vitric ash-flow tuff with somewhat accidental fragments of andesitic rocks, but dacitic rocks. The variation of major and trace element of the volcanic rocks show that contents of $Al_2O_3$, FeO, CaO, MgO, $TiO_2$ decrease with increasing of $SiO_2$. On the basis of Variation diagrams such as $Al_2O_3$ vs. CaO, Th/Yb vs. Ta/Yb, and $Ce_N/YB_N$ vs. $Ce_N$, these rocks represent mainly differentiation trend of calc-alkaline rock series. On the discriminant diagrams such as Ba/La and La/Th ratio, Rb vs. Y + Nb, the volcanic rocks in study area belongs to high-K Orogenic suites, with abundances of trace element and ternary diagram of K, Na, Ca. According to the tectonic discriminant diagram by Wood, these rocks falls into the diestructructive continental margin. K-Ar ages of whole rocks are from andesite to rhyolite $97.0{\pm}6.8~94.5{\pm}6.6,\68.9{\pm}4.8,\61.5{\pm}4.9~60.7{\pm}4.2$ Ma, repectively. Volcanic rocks in study area show well correlation to the Yucheon Group in terms of rock age dating and geochemcial data, and derived from andesitic calc-alkaline magma that undergone low pressure fractional crystallization dominated plagioclase at <30km.

• Journal of Chest Surgery
• /
• v.32 no.3
• /
• pp.281-286
• /
• 1999

Background: The purpose of this study is to analyze the types of complications, the incidences of complications, and preoperative and postoperative risk factors affecting the incidence of the complication. Material and Method: Between August 1990 and August 1997 in Asan Medical Center, 42 patients(24 men and 18 women) underwent surgical resection for pulmonary aspergilloma. The mean age was 46.6${\pm}$11.5 years(range 29 to 69 years). Hemoptysis(90%) was the most common presentation. Pulmonary tuberculosis was the most common predisposing cause(81%). The associated diseases were bronchiectasis(n=11), active puolmonary tuberculosis(n=9), diabetes mellitus(n=8), lung carcinoid(n=1), and acute myeloblastic leukemia(n=1). Lobectomy was done in 32 cases(76%), segmentectomy or wedge resection in 4, pneumonectomy in 2, and lobectomy combined with segmentectomy in 4. Result: Operative mortality was 2%. The most common postoperative complication was persistent air leakage(n=6). The variables such as age, sex, pulmonary function test, amount and duration of hemoptysis, associated diseases(diabetes mellitus, active pulmonary tuberculosis), mode of preoperative management(steroid, antifungal agent, bronchial arterial embolization), and modes of operative procedures were statistically insignificant. The radiologic extent of infiltration to normal lung parenchyme was statistically significant(p=0.04). Conclusion: We conclude that the extent of the infiltration to normal lung parenchyme in preoperative radiologic studies should be carefully evaluated to reduce the postoperative complications in surgery for pulmonary aspergilloma.

• Journal of the Korean Applied Science and Technology
• /
• v.18 no.3
• /
• pp.197-213
• /
• 2001

Conjugate trienoic acids (CTA) occurred in triacylglycerols (TGs) of the seed oils of Trichosanthes kirilowii, Momordica charantia and Aleurites fordii, and they were easily converted to their methyl esters in a mixture of sodium methoxide-methanol without any structural destruction. The main fatty acids in triacylglycerol (TG) fraction of the seed oils of Trichosanthes kirilowii are $C_{18:2{\omega}6}$ (32.2 mol %), $C_{18:3{\;}9c.11t,13c}$ (38.0 mol %) and $C_{18:1{\omega}9}$ (11.8 mol %), followed with $C_{16:0}$ (4.8 mol %) and $C_{18:0}$ (3.1 mol %). The TG fraction was resolved into 20 TG molecular species according to the partition number (PN) by reversed-phase (RP)-HPLC. The main TG species were $DT_{c2}$, $MDT_{c}$ and $D_{2}T_{c}$, of which amounts reached 63 mol % of total TG molecular species. The TG sample was fractionated into 11 fractions according to the number of double bond in the molecule by $Ag^{+}-HPLC$ and the species of $DT_{c2}$, $MDT_{c}$ and $D_{2}T_{c}$ were also eluted as main components. The TG species containing CTA showed unusual behaviours in the order of elution by HPLC ; first, TG moleular species of $DT_{c2}$ (D; dienoic acid, $T_{c}$; punicic acid, $T_{ci}$; ${\alpha}-eleostearic$ acid, M ; monoenoic acid, $S_{t}$; stearic acid) was eluted earlier than $Mt_{c2}$, although they have the same PN number of 40, and, secondly, the species of $DT_{ci2}$ with eight double bonds was eluted earlier than that of $D_2T_{ci}$ with seven double bonds. Intact TG of the seed oils of Momordica charantia contained mainly fatty acids such as $C_{18:3{\omega}9c,11t,13t}$ (57.7 mol %), $C_{18:1{\omega}9}$ (17.4 mol %), $C_{18:0}$ (12.3 mol %) and $C_{18:2{\omega}6}$ (10.6 mol %), and was classified into 13 fractions by RP-HPLC. The main TG species were as follows ; $MT_{ci2}$ [$(C_{18:1{\omega}9})(C_{18:3\;9c,11t,13t})_{2}$, 39.1 mol %] and $S_{t}T_{ci2}$ [$(C_{18:0})(C_{18:3\;9c,11t,13t})_2$, 33.9 mol %] comprising about 73 mol % of total TG species, accompanied by $DT_{ci2}$ [$(C_{18:2{\omega}6})(C_{18:3\;9c,11t,13t})_{2}$, 7.3 mol %], $D_{2}T_{ci}$ [$ (C_{18:2{\omega}6})_{2}(C_{18:3\;9c,11t,13t})$, 3.6 mol %] and $MDT_{ci}$ [$(C_{18:1{\omega}9})(C_{18:2{\omega}6})(C_{18:3\;9c,11t,13t})$, 3.5 mol %]. Simple TG species of $T_{ci3}$ [$(C_{18:3\;9c,11t,13t})_3]$ was present in a small amount of 1.4 mol %, but other simple TG species were not detected. The TG was also resolved into 11 fractions according to the number of double bond by $Ag^{+}-HPLC$, and the species were mainly occupied by $MT_{ci2}$ [$(C_{18:1{\omega}9})(C_{18:3\;9c,11t,13t})_{2}$, 39.4 mol %] and $S_tT-{ci2}$ [$(C_{18:0})(C_{18:3\;9c,11t,13t})_{2}$, 35.4 mol %] $DT_{ci2}$ species with eight double bonds was also developed faster than $D_2T_{ci}$ one with seven double bonds as indicated in the analysis of TG of the seed oils of T. kirilowii, and $MT_{ci2}$ species with cis, trans, trans-configurated double bond was eluted earlier than $MT_{c2}$ species with cis, trans, cis-configurated double bond. The main components of fatty acid in total TG fraction isolated from the seed oils of of Aleurites fordii were in the following order ; $C_{18:3\;9c,11t,13t}$ (81.2 mol %)> $C_{18:2{\omega}6}$ (8.5 mol %)> $C_{18:1{\omega}9}$ (5.4 mol %)$. With resolution of the TG by RP-HPLC, eight fractions such as $T_{ci3}$, $Dt_{ci2}$, $D_{2}T_{ci}$, $MT_{ci2}$, $PT_{ci2}$ (P; palmitic acid), $PMT_{ci}$, $PDT_{ci}$ and $S_{t}T_{ci2}$ ($S_{t}$; stearic acid) were isolated, respectively. TG species of $T_{ci3}$ [$(C_{18:3\;9c,11t,13t})_{3}$, 54.2 mol %], $DT_{ci2}$ [$(C_{18:2{\omega}6})(C_{18:3\;9c,11t,13t})_{2}$, 15.0 mol %] and $MT_{ci2}$ [$(C_{18:1{\omega}9})(C_{18:3 9c,11t,13t})_{2}$, 14.8 mol %] were present as main species.

• Journal of Intelligence and Information Systems
• /
• v.25 no.4
• /
• pp.141-154
• /
• 2019

Rapid growth of internet technology and social media is progressing. Data mining technology has evolved to enable unstructured document representations in a variety of applications. Sentiment analysis is an important technology that can distinguish poor or high-quality content through text data of products, and it has proliferated during text mining. Sentiment analysis mainly analyzes people's opinions in text data by assigning predefined data categories as positive and negative. This has been studied in various directions in terms of accuracy from simple rule-based to dictionary-based approaches using predefined labels. In fact, sentiment analysis is one of the most active researches in natural language processing and is widely studied in text mining. When real online reviews aren't available for others, it's not only easy to openly collect information, but it also affects your business. In marketing, real-world information from customers is gathered on websites, not surveys. Depending on whether the website's posts are positive or negative, the customer response is reflected in the sales and tries to identify the information. However, many reviews on a website are not always good, and difficult to identify. The earlier studies in this research area used the reviews data of the Amazon.com shopping mal, but the research data used in the recent studies uses the data for stock market trends, blogs, news articles, weather forecasts, IMDB, and facebook etc. However, the lack of accuracy is recognized because sentiment calculations are changed according to the subject, paragraph, sentiment lexicon direction, and sentence strength. This study aims to classify the polarity analysis of sentiment analysis into positive and negative categories and increase the prediction accuracy of the polarity analysis using the pretrained IMDB review data set. First, the text classification algorithm related to sentiment analysis adopts the popular machine learning algorithms such as NB (naive bayes), SVM (support vector machines), XGboost, RF (random forests), and Gradient Boost as comparative models. Second, deep learning has demonstrated discriminative features that can extract complex features of data. Representative algorithms are CNN (convolution neural networks), RNN (recurrent neural networks), LSTM (long-short term memory). CNN can be used similarly to BoW when processing a sentence in vector format, but does not consider sequential data attributes. RNN can handle well in order because it takes into account the time information of the data, but there is a long-term dependency on memory. To solve the problem of long-term dependence, LSTM is used. For the comparison, CNN and LSTM were chosen as simple deep learning models. In addition to classical machine learning algorithms, CNN, LSTM, and the integrated models were analyzed. Although there are many parameters for the algorithms, we examined the relationship between numerical value and precision to find the optimal combination. And, we tried to figure out how the models work well for sentiment analysis and how these models work. This study proposes integrated CNN and LSTM algorithms to extract the positive and negative features of text analysis. The reasons for mixing these two algorithms are as follows. CNN can extract features for the classification automatically by applying convolution layer and massively parallel processing. LSTM is not capable of highly parallel processing. Like faucets, the LSTM has input, output, and forget gates that can be moved and controlled at a desired time. These gates have the advantage of placing memory blocks on hidden nodes. The memory block of the LSTM may not store all the data, but it can solve the CNN's long-term dependency problem. Furthermore, when LSTM is used in CNN's pooling layer, it has an end-to-end structure, so that spatial and temporal features can be designed simultaneously. In combination with CNN-LSTM, 90.33% accuracy was measured. This is slower than CNN, but faster than LSTM. The presented model was more accurate than other models. In addition, each word embedding layer can be improved when training the kernel step by step. CNN-LSTM can improve the weakness of each model, and there is an advantage of improving the learning by layer using the end-to-end structure of LSTM. Based on these reasons, this study tries to enhance the classification accuracy of movie reviews using the integrated CNN-LSTM model.