• Journal of Intelligence and Information Systems
• /
• v.20 no.4
• /
• pp.89-105
• /
• 2014

After emergence of Internet, social media with highly interactive Web 2.0 applications has provided very user friendly means for consumers and companies to communicate with each other. Users have routinely published contents involving their opinions and interests in social media such as blogs, forums, chatting rooms, and discussion boards, and the contents are released real-time in the Internet. For that reason, many researchers and marketers regard social media contents as the source of information for business analytics to develop business insights, and many studies have reported results on mining business intelligence from Social media content. In particular, opinion mining and sentiment analysis, as a technique to extract, classify, understand, and assess the opinions implicit in text contents, are frequently applied into social media content analysis because it emphasizes determining sentiment polarity and extracting authors' opinions. A number of frameworks, methods, techniques and tools have been presented by these researchers. However, we have found some weaknesses from their methods which are often technically complicated and are not sufficiently user-friendly for helping business decisions and planning. In this study, we attempted to formulate a more comprehensive and practical approach to conduct opinion mining with visual deliverables. First, we described the entire cycle of practical opinion mining using Social media content from the initial data gathering stage to the final presentation session. Our proposed approach to opinion mining consists of four phases: collecting, qualifying, analyzing, and visualizing. In the first phase, analysts have to choose target social media. Each target media requires different ways for analysts to gain access. There are open-API, searching tools, DB2DB interface, purchasing contents, and so son. Second phase is pre-processing to generate useful materials for meaningful analysis. If we do not remove garbage data, results of social media analysis will not provide meaningful and useful business insights. To clean social media data, natural language processing techniques should be applied. The next step is the opinion mining phase where the cleansed social media content set is to be analyzed. The qualified data set includes not only user-generated contents but also content identification information such as creation date, author name, user id, content id, hit counts, review or reply, favorite, etc. Depending on the purpose of the analysis, researchers or data analysts can select a suitable mining tool. Topic extraction and buzz analysis are usually related to market trends analysis, while sentiment analysis is utilized to conduct reputation analysis. There are also various applications, such as stock prediction, product recommendation, sales forecasting, and so on. The last phase is visualization and presentation of analysis results. The major focus and purpose of this phase are to explain results of analysis and help users to comprehend its meaning. Therefore, to the extent possible, deliverables from this phase should be made simple, clear and easy to understand, rather than complex and flashy. To illustrate our approach, we conducted a case study on a leading Korean instant noodle company. We targeted the leading company, NS Food, with 66.5% of market share; the firm has kept No. 1 position in the Korean "Ramen" business for several decades. We collected a total of 11,869 pieces of contents including blogs, forum contents and news articles. After collecting social media content data, we generated instant noodle business specific language resources for data manipulation and analysis using natural language processing. In addition, we tried to classify contents in more detail categories such as marketing features, environment, reputation, etc. In those phase, we used free ware software programs such as TM, KoNLP, ggplot2 and plyr packages in R project. As the result, we presented several useful visualization outputs like domain specific lexicons, volume and sentiment graphs, topic word cloud, heat maps, valence tree map, and other visualized images to provide vivid, full-colored examples using open library software packages of the R project. Business actors can quickly detect areas by a swift glance that are weak, strong, positive, negative, quiet or loud. Heat map is able to explain movement of sentiment or volume in categories and time matrix which shows density of color on time periods. Valence tree map, one of the most comprehensive and holistic visualization models, should be very helpful for analysts and decision makers to quickly understand the "big picture" business situation with a hierarchical structure since tree-map can present buzz volume and sentiment with a visualized result in a certain period. This case study offers real-world business insights from market sensing which would demonstrate to practical-minded business users how they can use these types of results for timely decision making in response to on-going changes in the market. We believe our approach can provide practical and reliable guide to opinion mining with visualized results that are immediately useful, not just in food industry but in other industries as well.

• Journal of the Korean Academy of Child and Adolescent Psychiatry
• /
• v.12 no.1
• /
• pp.103-114
• /
• 2001

Tourette's disorder is a disease which manifests one or more motor tics and vocal tics for more than a year. Chronic motor tic or vocal tic disorders are characterized by only one kind of tics for more than a year. We intended to investigate the clinical characteristics of the patients with chronic motor tic disorders or Tourette's disorders who had admitted from May 1, 1998 to May 1, 1999 to Seoul National University Hospital Child and Adolescent Psychiatry ward. In addition, we compared the clinical characteristics of the patients in order to elucidate the relationship between the two disorders. The patients with learning disabilities were selected as controls. There was no statistically significant difference between the onsets of the patients with chronic motor tic disorders(n=13, $7.3{\pm}2.5$ years), and Tourette's disorder(n=39, $7.2{\pm}2.2$ years), but with learning disability($4.2{\pm}1.9$ years). Also, the patients with chronic motor tic disorder and Tourette's disorder showed similar age at admission($11.7{\pm}2.7$ versus $11.5{\pm}2.6$ years), duration of admission($5.7{\pm}5.4$ versus $11.0{\pm}8.7$ weeks), mothers' ages at child birth($27.3{\pm}2.9$ versus $28.3{\pm}6.7$ years old),and fathers' age at child birth($32.2{\pm}3.2$ versus $33.3{\pm}5.2$ years old). We observed that those who had learning disabilities were alike in those aspects, except for age at visit to clinic($9.8{\pm}3.2$ years old). Family history of psychiatric illnesses(24.1% versus 46.2%), recognized precipitating factors(11.1% versus 35.7%) and response to pharmacological treatments(77.8% versus 76.9%) of the patients with chronic motor tic disorders and Tourette's disorders were observed and no differences were found. Comorbid patterns of diseases were noted. Intrafamilial conflicts were more common in the patients with learning disabilities than those with chronic tic disorders or Tourette's disorders. Precipitating factors were observed more frequent in chronic tic disorder and Tourette's disorder than learning disability. Neurocognitive profiles were investigated, and verbal IQs of the patients with chronic motor tic disorder, Tourette's disorder and learning disability were $92.3{\pm}10.7$, $94.7{\pm}14.9$, $94.3{\pm}13.8$, performance IQs $93.0{\pm}20.5$, $97.5{\pm}13.0$, $95.0{\pm}16.9$ and full-scale IQs $91.9{\pm}20.1$, $95.8{\pm}14.5$, $93.9{\pm}15.1$, respectively, which were found to be not significantly different. No difference was found in structural neurological abnormalities and EEG profiles. The patients with learning disabilities showed more common Bender-Gestalt test abnormalities. In conclusion, we have not found any affirmative clues for the division of chronic motor tic disorder and Tourette's disorder in clinical perspective.

• Journal of Intelligence and Information Systems
• /
• v.24 no.2
• /
• pp.59-83
• /
• 2018

With the increasing importance of sentiment analysis to grasp the needs of customers and the public, various types of deep learning models have been actively applied to English texts. In the sentiment analysis of English texts by deep learning, natural language sentences included in training and test datasets are usually converted into sequences of word vectors before being entered into the deep learning models. In this case, word vectors generally refer to vector representations of words obtained through splitting a sentence by space characters. There are several ways to derive word vectors, one of which is Word2Vec used for producing the 300 dimensional Google word vectors from about 100 billion words of Google News data. They have been widely used in the studies of sentiment analysis of reviews from various fields such as restaurants, movies, laptops, cameras, etc. Unlike English, morpheme plays an essential role in sentiment analysis and sentence structure analysis in Korean, which is a typical agglutinative language with developed postpositions and endings. A morpheme can be defined as the smallest meaningful unit of a language, and a word consists of one or more morphemes. For example, for a word '예쁘고', the morphemes are '예쁘(= adjective)' and '고(=connective ending)'. Reflecting the significance of Korean morphemes, it seems reasonable to adopt the morphemes as a basic unit in Korean sentiment analysis. Therefore, in this study, we use 'morpheme vector' as an input to a deep learning model rather than 'word vector' which is mainly used in English text. The morpheme vector refers to a vector representation for the morpheme and can be derived by applying an existent word vector derivation mechanism to the sentences divided into constituent morphemes. By the way, here come some questions as follows. What is the desirable range of POS(Part-Of-Speech) tags when deriving morpheme vectors for improving the classification accuracy of a deep learning model? Is it proper to apply a typical word vector model which primarily relies on the form of words to Korean with a high homonym ratio? Will the text preprocessing such as correcting spelling or spacing errors affect the classification accuracy, especially when drawing morpheme vectors from Korean product reviews with a lot of grammatical mistakes and variations? We seek to find empirical answers to these fundamental issues, which may be encountered first when applying various deep learning models to Korean texts. As a starting point, we summarized these issues as three central research questions as follows. First, which is better effective, to use morpheme vectors from grammatically correct texts of other domain than the analysis target, or to use morpheme vectors from considerably ungrammatical texts of the same domain, as the initial input of a deep learning model? Second, what is an appropriate morpheme vector derivation method for Korean regarding the range of POS tags, homonym, text preprocessing, minimum frequency? Third, can we get a satisfactory level of classification accuracy when applying deep learning to Korean sentiment analysis? As an approach to these research questions, we generate various types of morpheme vectors reflecting the research questions and then compare the classification accuracy through a non-static CNN(Convolutional Neural Network) model taking in the morpheme vectors. As for training and test datasets, Naver Shopping's 17,260 cosmetics product reviews are used. To derive morpheme vectors, we use data from the same domain as the target one and data from other domain; Naver shopping's about 2 million cosmetics product reviews and 520,000 Naver News data arguably corresponding to Google's News data. The six primary sets of morpheme vectors constructed in this study differ in terms of the following three criteria. First, they come from two types of data source; Naver news of high grammatical correctness and Naver shopping's cosmetics product reviews of low grammatical correctness. Second, they are distinguished in the degree of data preprocessing, namely, only splitting sentences or up to additional spelling and spacing corrections after sentence separation. Third, they vary concerning the form of input fed into a word vector model; whether the morphemes themselves are entered into a word vector model or with their POS tags attached. The morpheme vectors further vary depending on the consideration range of POS tags, the minimum frequency of morphemes included, and the random initialization range. All morpheme vectors are derived through CBOW(Continuous Bag-Of-Words) model with the context window 5 and the vector dimension 300. It seems that utilizing the same domain text even with a lower degree of grammatical correctness, performing spelling and spacing corrections as well as sentence splitting, and incorporating morphemes of any POS tags including incomprehensible category lead to the better classification accuracy. The POS tag attachment, which is devised for the high proportion of homonyms in Korean, and the minimum frequency standard for the morpheme to be included seem not to have any definite influence on the classification accuracy.

• Journal of the Korean Academy of Child and Adolescent Psychiatry
• /
• v.15 no.2
• /
• pp.160-167
• /
• 2004

Objectives : There have been very few studies on the compliance of methylphenidate-immediate releasing form(MPH-IR), which is the most frequently used drug in Korea, in Attention Deficit Hyperactivity Disorder(ADHD). This study was conducted to investigate the compliance rate and the related factors in the one year pharmacotherapy process via OPD for children with ADHD. Method : Total 100 ADHD patients were selected randomly among patients who have been treated with MPH-IR from September in 2002 to December in 2002. All the selected patients were diagnosed with DSM-IV-ADHD criteria and fulfilled the inclusion criteria. In March, 2003(at the time of 6 month treatment), all the patients and parents received the questionnaire for the compliance and satisfaction for MPH-IR treatment. In October 2003(at time of 1 year treatment), we, investigators evaluated the socio-demographic variables, developmental data, medical data, family data, comorbid disorders, treatment variables, and compliance rate. Through these very comprehensive data, The compliance rate at the time of mean 1 year treatment and the related factors were investigated. Result : 1) In the questionnaire for compliance and satisfaction for MPND treatment, the 60% of respondents(parents) reported more than moderate degree of satisfaction in the effectiveness of MPND. Their compliance rate for the morning prescription was 81%, but the rate of afternoon prescription was 43%. 2) In the evaluation at the time of 1 year treatment(October 2003), the 38% of parents were dropped out from the OPD treatment. The mean compliance rate for the 1 year treatment was 62%. the 38% of parents were dropped out from the OPD treatment. The mean compliance rate for the 1year treatment was 62%. 3) Compared with the noncompliant group(drop-out group), compliant group showed higher total, verbal and performance IQ scores. In the treatment variables, higher reposponder rate(clinician rating), higher medication dosage and more compliance rate in afternoon prescription were found in the compliant group compared with the noncompliant group. There were no statistical differences in the demographic variables(age, sex, SES, parental education level), medical data, developmental profiles and academic function. Conclusion : To our knowledge, this is the first report about the compliance rate of the MPH-IR treatment for the children with ADHD. The compliance rate at the time of mean 1year treatment was 62%, which was comparable with other studies performed in foreign countries, especially States. In this study, the compliance related factors were IQ score, clinical treatment response, dosage of MPH-IR, and early compliance for the afternoon prescription. These results suggest that clinician plan the strategies for the promotion of the early compliance for the after prescription and enhancement of overall treatment response.

• Korean Journal of Biological Psychiatry
• /
• v.13 no.2
• /
• pp.70-81
• /
• 2006

Objective : The purpose of this study was to investigate the differences of the cognitive performance, emotional and behavioral problems among the attention-deficit/hyperactivity disorder(ADHD) groups that show the difference between visual and auditory attention. Method : Using 'ADHD Diagnostic System(ADS)', visual attention and auditory attention of 98 children diagnosed as ADHD were measured. According to the omission and commission error of ADS, they were divided into three groups ; 1) the group whose each visual omission and commission error scores were higher than each auditory omission and commission error scores(VV group), 2) the group whose each auditory omission and commission error scores were higher than each visual omission and commission error scores(AA group), 3) the group that was the rest of VV and AA group(M group). And the results of both the subscales of Korean Educational Development Institute-Wechsler Intelligence Scale for Children(KEDI-WISC) and the subscales of Korean Child Behavior Checklist(K-CBCL) among three groups were compared. Finally, the correlation between the visual omission, visual commission, auditory omission, auditory commission error and the results of KEDI-WISC, K-CBCL were investigated. Results : The results were as follows ; 1) In 98 ADHD children, the number of VV group(N=56) was higher than that of AA (N=10) and M group (N=32). 2) All mean scores of the subscales of KEDI-WISC of VV group were higher than those of M and AA group. The score of verbal IQ(p=.039) of VV group was significantly higher than that of AA group and the scores of block design(p=.015), Kaufman's factor 2(p=.045), performance IQ(p=.004) were significantly higher than those of M group. The score of full IQ(p=.004) were significantly higher than that of M and AA group. 3) The mean scores of all K-CBCL subscales of VV group were higher than those of M and AA group, except the score of Somatic complaint subscale. The score of Social subscale(p=.041) of VV group was significantly higher than that of AA group. The score of Withdrawn subscale(p=.021) of AA group was significantly higher than that of VV group. 4) There were no significant correlation between the scores of visual omission/commission error and those of each subscale of KEDI-WISC. But, there were many significant correlations between the scores of auditory omission/commission error and those of each subscale of KEDI-WISC. 5) There were significant correlation between the score of the visual omission error and that of Thought problem subscale(r=.205, p=.043) of K-CBCL. There were significant correlation between the scores of the auditory omission error and those of Social subscale(r=-.319, p=.001), Social problems subscale(r=.206, p=.042), Thought problem subscale(r=.235, p=.021). Finally, there were significant correlation between the scores of auditory commission error and those of Social subscale(r=-.241, p=.017), Thought problem subscale(r=.235, p=.020). Conclusion : The ADHD children whose auditory attention ability were higher than visual attention ability had relatively better cognitive performance and less emotional/behavioral problems than the others. The more comprehensive experiment will be needed about the cognitive performance, emotion and behavior problems of the ADHD children showing the difference between visual and auditory attention.

• The Journal of Korean Philosophical History
• /
• no.32
• /
• pp.215-249
• /
• 2011

This essay reflects on the modernity of Korea by examining the transformation of the traditional knowledge system from a historico-semantic perspective with its focus on the opposition and collision of the old and the new conception occurred in the early period(1890~1910) of the acceptance of the Western modern civilization. With scientific success, trick of reason, Christianity and evolutionary view of history, the Western modernity regarded itself as a peak of civilization and forced the non-Western societies into the world system in which they came to be considered as 'barbarism(野蠻)' or 'half-enlightened(半開).' The East Asian civilization, which had its own history for several centuries, became degraded as kind of delusion and old-fashioned customs from which it ought to free itself. The Western civilization presented itself as exemplary future which East Asian people should achieve, while East Asian past traditions came to be conceived as just unnecessary vestiges which it was better to wipe out. It can be said that East Asian modernization was established through the propagation and acceptance of the modern products of the Western civilization rather than through the preservation of its past experience and pursuit of the new at the same time. Accordingly, it is difficult to apply directly to East Asian societies Koselleck's hypothesis; while mapping out his Basic Concept of History, he assumed that, in the so-called 'age of saddle,' semantic struggle over concepts becomes active between the past experience and the horizon of expectation on the future, and concepts undergoes 'temporalization', 'democratization', 'ideologization', 'politicization.'The struggle over the old and new conceptions in Korea was most noticeable in the opposition of the Neo-Confucian scholars of Hwangseongsinmun and the theorists of civilization of Doknipsinmun. The opposition and struggle demanded the change of understanding in every field, but there was difference of opinion over the conception of the past traditional knowledge system. For the theorists of civilization, 'the old(舊)' was not just 'past' and 'old-fashioned' things, but rather an obstacle to the building of new civilization. On the other hand, it contained the possibility of regeneration(新) for the Neo-Confucian scholars; that is, they suggested finding a guide into tomorrow by taking lessons from the past. The traditional knowledge system lost their holy status of learning(聖學) in the process of its change into a 'new learning(新學),' and religion and religious tradition also weakened. The traditional knowledge system could change itself into modern learning by accepting scientific methodology which pursues objectivity and rationality. This transformation of the traditional knowledge system and 'the formation of the new learning from the old learning' was accompanied by the intersection between the old and new conceptions. It is necessary to pay attention to the role played by the concept of Sil(hak)(實學) or Practical Learning in the intersection of the old and new conceptions. Various modern media published before and after the 20th century show clearly the multi-layered development of the old and new conceptions, and it is noticeable that 'Sil(hak)' as conceptual frame of reference contributed to the transformation of the traditional knowledge system into the new learning. Although Silhak often designated, or was even considered equivalent to, the Western learning, Neo-Confucian scholars reinterpreted the concept of 'Silhak' which the theorists of civilization had monopolized until then, and opened the way to change the traditional knowledge system into the new learning. They re-appropriated the concept of Silhak, and enabled it to be invested with values, which were losing their own status due to the overwhelming scientific technology. With Japanese occupation of Korea by force, the attempt to transform the traditional knowledge system independently was obliged to reach its own limit, but its theory of 'making new learning from old one' can be considered to get over both the contradiction of Dondoseogi(東道西器: principle of preserving Eastern philosophy while accepting Western technology) and the de-subjectivity of the theory of civilization. While developing its own logic, the theory of Dongdoseogi was compelled to bring in the contradiction of considering the indivisible(道and 器) as divisible, though it tried to cope with the reality where the principle of morality and that of competition were opposed each other and the ideologies of 'evolution' and 'progress' prevailed. On the other hand, the theory of civilization was not free from the criticism that it brought about a crack in subjectivity due to its internalization of the West, cutting itself off from the traditional knowledge system.