• Journal of Intelligence and Information Systems
• /
• v.26 no.1
• /
• pp.1-21
• /
• 2020

With the rapid development of artificial intelligence technology, various techniques have been developed to extract meaningful information from unstructured text data which constitutes a large portion of big data. Over the past decades, text mining technologies have been utilized in various industries for practical applications. In the field of business intelligence, it has been employed to discover new market and/or technology opportunities and support rational decision making of business participants. The market information such as market size, market growth rate, and market share is essential for setting companies' business strategies. There has been a continuous demand in various fields for specific product level-market information. However, the information has been generally provided at industry level or broad categories based on classification standards, making it difficult to obtain specific and proper information. In this regard, we propose a new methodology that can estimate the market sizes of product groups at more detailed levels than that of previously offered. We applied Word2Vec algorithm, a neural network based semantic word embedding model, to enable automatic market size estimation from individual companies' product information in a bottom-up manner. The overall process is as follows: First, the data related to product information is collected, refined, and restructured into suitable form for applying Word2Vec model. Next, the preprocessed data is embedded into vector space by Word2Vec and then the product groups are derived by extracting similar products names based on cosine similarity calculation. Finally, the sales data on the extracted products is summated to estimate the market size of the product groups. As an experimental data, text data of product names from Statistics Korea's microdata (345,103 cases) were mapped in multidimensional vector space by Word2Vec training. We performed parameters optimization for training and then applied vector dimension of 300 and window size of 15 as optimized parameters for further experiments. We employed index words of Korean Standard Industry Classification (KSIC) as a product name dataset to more efficiently cluster product groups. The product names which are similar to KSIC indexes were extracted based on cosine similarity. The market size of extracted products as one product category was calculated from individual companies' sales data. The market sizes of 11,654 specific product lines were automatically estimated by the proposed model. For the performance verification, the results were compared with actual market size of some items. The Pearson's correlation coefficient was 0.513. Our approach has several advantages differing from the previous studies. First, text mining and machine learning techniques were applied for the first time on market size estimation, overcoming the limitations of traditional sampling based- or multiple assumption required-methods. In addition, the level of market category can be easily and efficiently adjusted according to the purpose of information use by changing cosine similarity threshold. Furthermore, it has a high potential of practical applications since it can resolve unmet needs for detailed market size information in public and private sectors. Specifically, it can be utilized in technology evaluation and technology commercialization support program conducted by governmental institutions, as well as business strategies consulting and market analysis report publishing by private firms. The limitation of our study is that the presented model needs to be improved in terms of accuracy and reliability. The semantic-based word embedding module can be advanced by giving a proper order in the preprocessed dataset or by combining another algorithm such as Jaccard similarity with Word2Vec. Also, the methods of product group clustering can be changed to other types of unsupervised machine learning algorithm. Our group is currently working on subsequent studies and we expect that it can further improve the performance of the conceptually proposed basic model in this study.

• Journal of Intelligence and Information Systems
• /
• v.16 no.3
• /
• pp.77-97
• /
• 2010

Market timing is an investment strategy which is used for obtaining excessive return from financial market. In general, detection of market timing means determining when to buy and sell to get excess return from trading. In many market timing systems, trading rules have been used as an engine to generate signals for trade. On the other hand, some researchers proposed the rough set analysis as a proper tool for market timing because it does not generate a signal for trade when the pattern of the market is uncertain by using the control function. The data for the rough set analysis should be discretized of numeric value because the rough set only accepts categorical data for analysis. Discretization searches for proper "cuts" for numeric data that determine intervals. All values that lie within each interval are transformed into same value. In general, there are four methods for data discretization in rough set analysis including equal frequency scaling, expert's knowledge-based discretization, minimum entropy scaling, and na$\ddot{i}$ve and Boolean reasoning-based discretization. Equal frequency scaling fixes a number of intervals and examines the histogram of each variable, then determines cuts so that approximately the same number of samples fall into each of the intervals. Expert's knowledge-based discretization determines cuts according to knowledge of domain experts through literature review or interview with experts. Minimum entropy scaling implements the algorithm based on recursively partitioning the value set of each variable so that a local measure of entropy is optimized. Na$\ddot{i}$ve and Booleanreasoning-based discretization searches categorical values by using Na$\ddot{i}$ve scaling the data, then finds the optimized dicretization thresholds through Boolean reasoning. Although the rough set analysis is promising for market timing, there is little research on the impact of the various data discretization methods on performance from trading using the rough set analysis. In this study, we compare stock market timing models using rough set analysis with various data discretization methods. The research data used in this study are the KOSPI 200 from May 1996 to October 1998. KOSPI 200 is the underlying index of the KOSPI 200 futures which is the first derivative instrument in the Korean stock market. The KOSPI 200 is a market value weighted index which consists of 200 stocks selected by criteria on liquidity and their status in corresponding industry including manufacturing, construction, communication, electricity and gas, distribution and services, and financing. The total number of samples is 660 trading days. In addition, this study uses popular technical indicators as independent variables. The experimental results show that the most profitable method for the training sample is the na$\ddot{i}$ve and Boolean reasoning but the expert's knowledge-based discretization is the most profitable method for the validation sample. In addition, the expert's knowledge-based discretization produced robust performance for both of training and validation sample. We also compared rough set analysis and decision tree. This study experimented C4.5 for the comparison purpose. The results show that rough set analysis with expert's knowledge-based discretization produced more profitable rules than C4.5.

• The Journal of Information Systems
• /
• v.24 no.2
• /
• pp.23-52
• /
• 2015

Purpose This study aims to understand the role of digital knowledge in accepting the green technology. This study combined digital option theory with the second version of the Unified Theory of Acceptance and Use of Technology (UTAUT2). Contrary to other studies in which the UTAUT2 is used to explain IT adoption behavior, we look at the relationship between IT and the UTAUT2 from a new angle, incorporating an important aspect of IT, that is, digitized knowledge richness, as a determinant of the UTAUT2. Design/methodology/approach Grounded in the UTAUT2, a content analysis was conducted to investigate novel constructs dedicated to explaining green technology adoption. In this study, an amended version of the UTAUT2 specific to green technology is offered that better explains the green technology adoption behavior of consumers. Using the items identified by content analysis, we developed a questionnaire with 36 survey items. We measured all the items on a seven-point Likert-type scale. We randomly selected 402 survey respondents from a set of panel data. After a pilot study, we analyzed the main survey data by using PLS 2.0M3 and SPSS 20.0, and employed structural equation modeling to test the hypotheses. Findings The results suggest that the UTAUT2 was found to be extendable to technologies other than conventional IT. Social influence is more significant than conventional utilitarian and hedonic-based constructs such as those utilized in the UTAUT and UTAUT2 in explaining adoption behavior in the context of green technologies. The hypothesized connection between digitized knowledge richness and adoption intention was supported by the results of studies on the role of IT in formation of attitudes toward eco-friendly production. The results also indicate that digital knowledge can also encourage people to try green technology when they learn that their peers are already using the technology successfully.

• Journal of the Korean Academy of Child and Adolescent Psychiatry
• /
• v.11 no.2
• /
• pp.252-261
• /
• 2000

Objectivity：This study was conducted to examine the depression, self-concept, perception of stress & coping strategy in children with chronic physical illness. Methods：Two groups of participants were recruited for this study, 13 children with chronic illness in outpatient or inpatient treatment at Seoul National University Children's Hospital, and 13 nonpatient children. They were assessed using Korean form of the Piers-Harris Self-Concept Scale (PHSCS), Kovacs' Children's Depression Inventory(CDI) and three subscaleds('color how you feel' 'color how others make you feel' 'A children in the rain' of Children's Self-Report and Projective Inventory(CSRPI). Result：There were significant differences between the chronic ill children and the healthy children in scores of depression and self-concept. The chronicity ill children were more depressive and had very negative self-concept, and obtained significantly lower scores than the healthy children in the subscales of PHSCS, 'intellectual/school status' and 'popularity' Among three scales of CSRPI, there was no difference in 'color how you feel' and 'color how others make you feel' But there were significant differences in all items of 'A child in the rain'(quantity of raining, duration of raining, tool, and effectiveness). 'Duration of raining' correlated most negatively with PHSCS scores, and correlated positively with CDI scores. Conclusion：The children with chronic illness are more depressive and have very negative selfconcept. And they feel that the stresses are more permanent, but have no appropriate coping strategy. The results suggest that the chronic illness strongly affects the psychological and emotional adjustment of children(i.e. depression, peer relation, stress coping strategy).

• Journal of the Korean Institute of Landscape Architecture
• /
• v.37 no.2
• /
• pp.1-13
• /
• 2009

The purpose of this study was to look into the feasibility of site suitability focused on the potential for environmentally- and water-friendly recreation area development in a wide area(Nakdong River 35km) and to study new methods for providing basic data in regard to the recreation planning over a wide area as well as in application to other sites. The results of this study are as follows. Through classification by mesh method, the site of this study was classified into 42 grids, and by means of the analysis of evaluation indicators, 20 indicators were established and sorted into 4 types of significant recreation activity. According to the results of the analysis for each recreation activity type, there were 8 essentials for water-friendly recreation activity types based on water use while water-friendly recreation types for static activity included 12 sub-essentials. As a result of the first evaluation(the minimum required evaluation) by each classified grid, 32 of the 42 total grids were implemented by the minimum requirements. These grids were usually distributed evenly through the whole site. In terms of the second evaluation(specific site evaluation) results, 6 grids were highly suitable for recreational nature experiences and landscape ecological learning, 4 grids for developing water-friendly recreation for exercise, 1 grid for building water-friendly recreation based on water use, and 4 grids for planning water-friendly recreation for static activity. The results of the grid evaluation of this study could be extended to contiguous grids or reduced. Actual planning for a water-friendly recreation area must change the grid shape or size through boundary adjustments.

• Clinical and Experimental Pediatrics
• /
• v.48 no.1
• /
• pp.21-26
• /
• 2005

Purpose : The aim of this study is to find the prevalence of obesity, the serum lipid levels and the age at menarche in adolescent female athletes and to examine the effects of exercise in adolescent stage. Methods : With a questionnaire regarding their age at menarche, physical measurement, body fat, and blood samples of the serum lipid levels to evaluate the hyperlipidemia were obtained from adolescent female athletes(n=107) and general adolescent students(n=650) who didn't exercise at regular intervals, aged 12 to 18 years. Results : The mean weight in the athletes' group was $53.3{\pm}7.3kg$ which was similar with $54.3{\pm}8.0kg$ in the control group. The mean height in the athletes' group was $161.4{\pm}5.4cm$, which was taller than $158.9{\pm}5.3cm$ in the control group. The prevalence of obesity by obesity index, body fat, and BMI in the athletes' group were significantly lower than in control group. There was no significant difference in age at menarche between two groups($12.6{\pm}1.3$, $12.9{\pm}1.2$). The levels of cholesterol, LDL cholesterol, and HDL cholesterol in the athletes' group were higher than in the control group. The levels of triglyceride in the athletes' group was lower than in control group. Conclusion : These data suggest the importance of exercise in adolescents for the prevention of obesity since it may reduce body fat and increase the height. There was no negative effect of exercise on the age at menarche. We think that more controlled assessment of nutrition, diet habit, hormonal effect and height are warranted to find the correlation with hyperlipidemia and exercise at the adolescent stage.

• Clinical and Experimental Pediatrics
• /
• v.52 no.7
• /
• pp.824-831
• /
• 2009

Purpose : We aimed to investigate the efficacy of and functional recovery after intracerebral transplantation of different doses of mouse mesenchymal stem cells (mMSCs) in immature rat brain with hypoxic-ischemic encephalopathy (HIE). Methods : Postnatal 7-days-old Sprague-Dawley rats, which had undergone unilateral HI operation, were given stereotaxic intracerebral injections of either vehicle or mMSCs and then tested for locomotory activity in the 2nd, 4th, 6th, and 8th week of the stem cell injection. In the 8th week, Morris water maze test was performed to evaluate the learning and memory dysfunction for a week. Results : In the open field test, no differences were observed in the total distance/the total duration (F=0.412, P=0.745) among the 4 study groups. In the invisible-platform Morris water maze test, significant differences were observed in escape latency (F=380.319, P<0.01) among the 4 groups. The escape latency in the control group significantly differed from that in the high-dose mMSC and/or sham group on training days 2-5 (Scheffe's test, P<0.05) and became prominent with time progression (F=6.034, P<0.01). In spatial probe trial and visible-platform Morris water maze test, no significant improvement was observed in the rats that had undergone transplantation. Conclusion : Although the rats that received a high dose of mMSCs showed significant recovery in the learning-related behavioral test only, our data support that mMSCs may be used as a valuable source to improve outcome in HIE. Further study is necessary to identify the optimal dose that shows maximal efficacy for HIE treatment.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.40 no.3
• /
• pp.273-283
• /
• 2020

Because of climate change, the occurrence of localized and heavy rainfall is increasing. It is important to predict floods in urban areas that have suffered inundation in the past. For flood prediction, not only numerical analysis models but also machine learning-based models can be applied. The LSTM (Long Short-Term Memory) neural network used in this study is appropriate for sequence data, but it demands a lot of data. However, rainfall that causes flooding does not appear every year in a single urban basin, meaning it is difficult to collect enough data for deep learning. Therefore, in addition to the rainfall observed in the study area, the observed rainfall in another urban basin was applied in the predictive model. The LSTM neural network was used for predicting the total overflow, and the result of the SWMM (Storm Water Management Model) was applied as target data. The prediction of the inundation map was performed by using logistic regression; the independent variable was the total overflow and the dependent variable was the presence or absence of flooding in each grid. The dependent variable of logistic regression was collected through the simulation results of a two-dimensional flood model. The input data of the two-dimensional flood model were the overflow at each manhole calculated by the SWMM. According to the LSTM neural network parameters, the prediction results of total overflow were compared. Four predictive models were used in this study depending on the parameter of the LSTM. The average RMSE (Root Mean Square Error) for verification and testing was 1.4279 ㎥/s, 1.0079 ㎥/s for the four LSTM models. The minimum RMSE of the verification and testing was calculated as 1.1655 ㎥/s and 0.8797 ㎥/s. It was confirmed that the total overflow can be predicted similarly to the SWMM simulation results. The prediction of inundation extent was performed by linking the logistic regression with the results of the LSTM neural network, and the maximum area fitness was 97.33 % when more than 0.5 m depth was considered. The methodology presented in this study would be helpful in improving urban flood response based on deep learning methodology.

• Clinical and Experimental Pediatrics
• /
• v.48 no.7
• /
• pp.701-705
• /
• 2005

Purpose : Parents' genetic information plays an important role in their children's genetic expression. Human chromosome has 23-paternal chromosomes and 23-maternal chromosomes. Parental chromosomal translocation can induce clinical problems in their children because of imbalance in genetic information. We intent to analyze the cytogenentic and clinical features about children with maternal balanced translocation between chromosome 15 and 18. Methods : We detected by one family's FISH study of chromosome 15. We have evaluated children born to clinically normal parents about peripheral bood analysis, endocrine, metabolic, radiologic study, electroencephalogram and social & intelligence scale. and We analysis their clinical manifestation by hospital records. Results : Patient's father and elder sister are normal clinically and genetically. Her mother's chromosome show balanced translocation, 46, XX, t(15;18)(p11.2;p11.3). One child has 46, XX, der(18) t(15;18)(p11.2;p11.3), mental retardation, growth retardation, speech & social developmental delay, recurrent infection and mild mitochondria dysfunction. Her young brother has 46, XY, der(15) t(15;18) (p11.2;p11.3), mental retardation, aggressive behavior, obesity and speech developmental delay. Conclusion : In this study we observed the children with developmental delay, dysmorphic facial features, mental retardation, growth retardation associated with growth hormone deficiency and aggressive behavior due to unbalanced translocation between chromosome 15 and 18.

• Clinical and Experimental Pediatrics
• /
• v.48 no.7
• /
• pp.706-710
• /
• 2005

Purpose : Early diagnosis of congenital hearing loss through the neonatal hearing screening test minimizes language defect. This research intends to identify frequency of congenital hearing loss in infants through neonatal hearing screening test with the aim of communicating the importance of hearing test for infants. Methods : From May 20, 2003 to May 19, 2004, infants were subjected to Automated Auditory Brainstem Response test during one month of birth to conduct the test with 35 dB sound. Infants who passed the 1st round of hearing test, were classified into 'pass' group whereas those who did not were classified into 'refer' group. Infants who did not 'pass' in the hearing test conducted within one month of birth were subjected to re-test one month later, and if classified as 'refer' during the re-test, they were subjected to the diagnosis for validation of hearing loss by requesting test to the hearing loss clinic. Results : There was no difference among the 'pass' and 'refer' group in terms of form of childbirth, weight at birth and gestational age. In the 1st test, total of 45 infants were classified into 'refer' group. Six among 35 who were subjected to re-test(17%) did not pass the re-test, and all were diagnosed with congenital hearing loss. This corresponds to 0.35%(3.5 per 1,000) among total number of 1,718 subjects. Conclusion : In our study the congenital hearing loss tends to be considerably more frequently than congenital metabolic disorder. Accordingly, newly born infants are strongly recommended to undergo neonatal hearing screening test.