• Journal of Intelligence and Information Systems
• /
• v.18 no.4
• /
• pp.117-127
• /
• 2012

Due to the recent expansion of the Web 2.0 -based services, along with the widespread of smartphones, online social network services are being popularized among users. Online social network services are the online community services which enable users to communicate each other, share information and expand human relationships. In the social network services, each relation between users is represented by a graph consisting of nodes and links. As the users of online social network services are increasing rapidly, the SNS are actively utilized in enterprise marketing, analysis of social phenomenon and so on. Social Network Analysis (SNA) is the systematic way to analyze social relationships among the members of the social network using the network theory. In general social network theory consists of nodes and arcs, and it is often depicted in a social network diagram. In a social network diagram, nodes represent individual actors within the network and arcs represent relationships between the nodes. With SNA, we can measure relationships among the people such as degree of intimacy, intensity of connection and classification of the groups. Ever since Social Networking Services (SNS) have drawn increasing attention from millions of users, numerous researches have made to analyze their user relationships and messages. There are typical representative SNA methods: degree centrality, betweenness centrality and closeness centrality. In the degree of centrality analysis, the shortest path between nodes is not considered. However, it is used as a crucial factor in betweenness centrality, closeness centrality and other SNA methods. In previous researches in SNA, the computation time was not too expensive since the size of social network was small. Unfortunately, most SNA methods require significant time to process relevant data, and it makes difficult to apply the ever increasing SNS data in social network studies. For instance, if the number of nodes in online social network is n, the maximum number of link in social network is n(n-1)/2. It means that it is too expensive to analyze the social network, for example, if the number of nodes is 10,000 the number of links is 49,995,000. Therefore, we propose a heuristic-based method for finding the shortest path among users in the SNS user graph. Through the shortest path finding method, we will show how efficient our proposed approach may be by conducting betweenness centrality analysis and closeness centrality analysis, both of which are widely used in social network studies. Moreover, we devised an enhanced method with addition of best-first-search method and preprocessing step for the reduction of computation time and rapid search of the shortest paths in a huge size of online social network. Best-first-search method finds the shortest path heuristically, which generalizes human experiences. As large number of links is shared by only a few nodes in online social networks, most nods have relatively few connections. As a result, a node with multiple connections functions as a hub node. When searching for a particular node, looking for users with numerous links instead of searching all users indiscriminately has a better chance of finding the desired node more quickly. In this paper, we employ the degree of user node vn as heuristic evaluation function in a graph G = (N, E), where N is a set of vertices, and E is a set of links between two different nodes. As the heuristic evaluation function is used, the worst case could happen when the target node is situated in the bottom of skewed tree. In order to remove such a target node, the preprocessing step is conducted. Next, we find the shortest path between two nodes in social network efficiently and then analyze the social network. For the verification of the proposed method, we crawled 160,000 people from online and then constructed social network. Then we compared with previous methods, which are best-first-search and breath-first-search, in time for searching and analyzing. The suggested method takes 240 seconds to search nodes where breath-first-search based method takes 1,781 seconds (7.4 times faster). Moreover, for social network analysis, the suggested method is 6.8 times and 1.8 times faster than betweenness centrality analysis and closeness centrality analysis, respectively. The proposed method in this paper shows the possibility to analyze a large size of social network with the better performance in time. As a result, our method would improve the efficiency of social network analysis, making it particularly useful in studying social trends or phenomena.

• Journal of Intelligence and Information Systems
• /
• v.26 no.2
• /
• pp.105-129
• /
• 2020

This study uses corporate data from 2012 to 2018 when K-IFRS was applied in earnest to predict default risks. The data used in the analysis totaled 10,545 rows, consisting of 160 columns including 38 in the statement of financial position, 26 in the statement of comprehensive income, 11 in the statement of cash flows, and 76 in the index of financial ratios. Unlike most previous prior studies used the default event as the basis for learning about default risk, this study calculated default risk using the market capitalization and stock price volatility of each company based on the Merton model. Through this, it was able to solve the problem of data imbalance due to the scarcity of default events, which had been pointed out as the limitation of the existing methodology, and the problem of reflecting the difference in default risk that exists within ordinary companies. Because learning was conducted only by using corporate information available to unlisted companies, default risks of unlisted companies without stock price information can be appropriately derived. Through this, it can provide stable default risk assessment services to unlisted companies that are difficult to determine proper default risk with traditional credit rating models such as small and medium-sized companies and startups. Although there has been an active study of predicting corporate default risks using machine learning recently, model bias issues exist because most studies are making predictions based on a single model. Stable and reliable valuation methodology is required for the calculation of default risk, given that the entity's default risk information is very widely utilized in the market and the sensitivity to the difference in default risk is high. Also, Strict standards are also required for methods of calculation. The credit rating method stipulated by the Financial Services Commission in the Financial Investment Regulations calls for the preparation of evaluation methods, including verification of the adequacy of evaluation methods, in consideration of past statistical data and experiences on credit ratings and changes in future market conditions. This study allowed the reduction of individual models' bias by utilizing stacking ensemble techniques that synthesize various machine learning models. This allows us to capture complex nonlinear relationships between default risk and various corporate information and maximize the advantages of machine learning-based default risk prediction models that take less time to calculate. To calculate forecasts by sub model to be used as input data for the Stacking Ensemble model, training data were divided into seven pieces, and sub-models were trained in a divided set to produce forecasts. To compare the predictive power of the Stacking Ensemble model, Random Forest, MLP, and CNN models were trained with full training data, then the predictive power of each model was verified on the test set. The analysis showed that the Stacking Ensemble model exceeded the predictive power of the Random Forest model, which had the best performance on a single model. Next, to check for statistically significant differences between the Stacking Ensemble model and the forecasts for each individual model, the Pair between the Stacking Ensemble model and each individual model was constructed. Because the results of the Shapiro-wilk normality test also showed that all Pair did not follow normality, Using the nonparametric method wilcoxon rank sum test, we checked whether the two model forecasts that make up the Pair showed statistically significant differences. The analysis showed that the forecasts of the Staging Ensemble model showed statistically significant differences from those of the MLP model and CNN model. In addition, this study can provide a methodology that allows existing credit rating agencies to apply machine learning-based bankruptcy risk prediction methodologies, given that traditional credit rating models can also be reflected as sub-models to calculate the final default probability. Also, the Stacking Ensemble techniques proposed in this study can help design to meet the requirements of the Financial Investment Business Regulations through the combination of various sub-models. We hope that this research will be used as a resource to increase practical use by overcoming and improving the limitations of existing machine learning-based models.

• The Journal of Korean Society for Radiation Therapy
• /
• v.32
• /
• pp.93-109
• /
• 2020

Purpose: The radiochromic film (Gafchromic EBT3, Ashland Advanced Materials, USA) and 3-dimensional analysis system dosimetry checkTM (DC, MathResolutions, USA) were evaluated for patient-specific quality assurance (QA) of helical tomotherapy. Materials and Methods: Depending on the tumors' positions, three types of targets, which are the abdominal tumor (130.6㎤), retroperitoneal tumor (849.0㎤), and the whole abdominal metastasis tumor (3131.0㎤) applied to the humanoid phantom (Anderson Rando Phantom, USA). We established a total of 12 comparative treatment plans by the four geometric conditions of the beam irradiation, which are the different field widths (FW) of 2.5-cm, 5.0-cm, and pitches of 0.287, 0.43. Ionization measurements (1D) with EBT3 by inserting the cheese phantom (2D) were compared to DC measurements of the 3D dose reconstruction on CT images from beam fluence log information. For the clinical feasibility evaluation of the DC, dose reconstruction has been performed using the same cheese phantom with the EBT3 method. Recalculated dose distributions revealed the dose error information during the actual irradiation on the same CT images quantitatively compared to the treatment plan. The Thread effect, which might appear in the Helical Tomotherapy, was analyzed by ripple amplitude (%). We also performed gamma index analysis (DD: 3mm/ DTA: 3%, pass threshold limit: 95%) for pattern check of the dose distribution. Results: Ripple amplitude measurement resulted in the highest average of 23.1% in the peritoneum tumor. In the radiochromic film analysis, the absolute dose was on average 0.9±0.4%, and gamma index analysis was on average 96.4±2.2% (Passing rate: >95%), which could be limited to the large target sizes such as the whole abdominal metastasis tumor. In the DC analysis with the humanoid phantom for FW of 5.0-cm, the three regions' average was 91.8±6.4% in the 2D and 3D plan. The three planes (axial, coronal, and sagittal) and dose profile could be analyzed with the entire peritoneum tumor and the whole abdominal metastasis target, with planned dose distributions. The dose errors based on the dose-volume histogram in the DC evaluations increased depending on FW and pitch. Conclusion: The DC method could implement a dose error analysis on the 3D patient image data by the measured beam fluence log information only without any dosimetry tools for patient-specific quality assurance. Also, there may be no limit to apply for the tumor location and size; therefore, the DC could be useful in patient-specific QAl during the treatment of Helical Tomotherapy of large and irregular tumors.

• Journal of Korean Home Economics Education Association
• /
• v.24 no.2
• /
• pp.117-134
• /
• 2012

The purpose of this study was to identify the stage of concern, the level of use, and the innovation configuration of Home Economics teachers regarding creativity and personality education in Home Economics(HE) classes. The survey questionnaires were sent through mails and e-mails to middle-school HE teachers in the whole country selected by systematic sampling and convenience sampling. Questionnaires of the stages of concern and the levels of use developed by Hall(1987) were used in this study. 187 data were used for the final analysis by using SPSS/window(12.0) program. The results of the study were as following: First, for the stage of concerns of HE teachers on creativity and personality education, the information stage of concerns(85.51) was the one with the highest response rate and the next high in the following order: the management stage of concerns(81.88), the awareness stage of concerns(82.15), the refocusing stage of concerns(68.80), the collaboration stage of concerns(61.97), and the consequence stage of concerns(59.76). Second, the levels of use of HE teachers on creativity and personality education was highest with the mechanical levels(level 3; 21.4%) and the next high in the following order: the orientation levels of use(level 1; 20.9%), the refinement levels(level 5; 17.1%), the non-use levels(level 0; 15.0%), the preparation levels(level 2; 10.2%), the integration levels(level 6; 5.9%), the renewal levels(level 7; 4.8%), the routine levels(level 4; 4.8%). Third, for the innovation configuration of HE teachers on creativity and personality education, more than half of the HE teachers(56.1%) mainly focused on personality education in their HE classes; 31.0% of the HE teachers performed both creativity and personality education; a small number of teachers(6.4%) focused on creativity education; the same number of teachers(6.4%) responded that they do not focus on neither of the two. Examining the level and type of performance HE teachers applied, the average score on the performance of creativity and personality education was 3.76 out of 5.00 and the mean of creativity component was 3.59 and of personality component was 3.94, higher than standard. For the creativity education, openness/sensitivity(3.97) education was performed most and the next most in the following order: problem-solving skill(3.79), curiosity/interest(3.73), critical thinking(3.63), problem-finding skill(3.61), originality(3.57), analogy(3.47), fluency/adaptability(3.46), precision(3.46), imagination(3.37), and focus/sympathy(3.37). For the personality education, the following components were performed in order from most to least: power of execution(4.07), cooperation/consideration/just(4.06), self-management skill(4.04), civic consciousness(4.04), career development ability(4.03), environment adaptability(3.95), responsibility/ownership(3.94), decision making(3.89), trust/honesty/promise(3.88), autonomy(3.86), and global competency(3.55). Regarding what makes performing creativity and personality education difficult, most HE teachers(64.71%) chose the lack of instructional materials and 40.11% of participants chose the lack of seminar and workshop opportunity. 38.5% chose the difficulty of developing an evaluation criteria or an evaluation tool while 25.67% responded that they do not know any means of performing creativity and personality education. Regarding the better way to support for creativity and personality education, the HE teachers chose in order from most to least: 'expansion of hands-on activities for students related to education on creativity and personality'(4.34), 'development of HE classroom culture putting emphasis on creativity and personality'(4.29), 'a proper curriculum on creativity and personality education that goes along with students' developmental stages'(4.27), 'securing enough human resource and number of professors who will conduct creativity and personality education'(4.21), 'establishment of the concept and value of the education on creativity and personality'(4.09), and 'educational promotion on creativity and personality education supported by local communities and companies'(3.94).

• Tuberculosis and Respiratory Diseases
• /
• v.42 no.6
• /
• pp.900-912
• /
• 1995

Background: Although graded exercise stress tests are widely used for the evaluation of cardiorespiratory performance, normal standards on respiratory gas exchange and ventilatory functions at maximal exercise in Koreans have not been well established. The purpose of this study is to provide reference values on these by sex and age, along with derivation of some of their prediction equations. Method: Symptom-limited maximal exercise test was carried out by Bruce protocol in 1,000 healthy adults consisting of 603 males and 397 females, aged 20~66 years. Among them VC, $FEV_1$ and MVV were also determined in 885 cases. All the subjects were members of a health center, excluding athletes. During the exercise, subjects were allowed to hold on to front hand rail of the treadmill for safety purpose. Results: The $VO_2\;max/m^2$, $VCO_2\;max/m^2$ and $V_E\;max/m^2$ were greater in males than in females and decreased with age. The RR max in men and women was similar but decreased slightly with age. The $V_T$ max was markedly greater in men but showed no significant changes with age in either gender. The mean of $V_T$ max/VC, $V_E$ max/MVV and BR revealed that there were considerable ventilatory reserves at maximal exercise even in older females. The regression equations of the cardinal parameters obtained using exercise time(ET, min), age(A, yr), height(Ht, cm), weight(W, kg), sex(S, 0=male; 1=female), VC(L), $FEV_1$(L) and $V_E$ max(L) as variables are as follows: $VO_2\;max/m^2$(L/min)=1.449+0.073 ET-0.007A+0.010W-0.006Ht-0.209S, $VCO_2\;max/m^2$(L/min)=1.672+0.063ET-0.008A+0.010W-0.005Ht-0.319S, VE max/$m^2$(L/min)=58.161+1.503ET-0.315A-9.871S or VE max/$m^2$(L/min)=47.873+6.548 $FEV_1$-5.715 S, and VT max(L)=1.497+0.223VC-0.493S. Conclusion: Respiratory gas exchange and ventilatory variables at maximal exercise were studied in 1,000 non-athletes by Bruce protocol. During exercise, the subjects were allowed to hold on to hand rail of the treadmill for safety purpose. We feel that our results would provide ideal target values for patients and healthy individuals to be achieved, since our study subjects were members of a health center whose physical fitness levels were presumably higher than ordinary population.

• The Korean Journal of Air & Space Law and Policy
• /
• v.30 no.2
• /
• pp.403-425
• /
• 2015

The Chicago Convention and Annexes have become the basis of aviation safety regulations for every contracting state. Generally, the State's aviation safety regulations refer to the Standards and Recommended Practices(SARPs) provided in the Annexes of the Chicago Convention. In order to properly reflect international aviation safety regulations, constant studies of the aviation fields are of paramount importance. This Paper is intended to identify the main differences between korean and foreign regulation and suggest a few amendment proposals on Mixed-fleet Flying(at or more two aircraft type operation) of flight crew. Comparing with these regulations, the korean regulations and implementations have some insufficiency points. I suggest some amendment proposals of korean regulations concerning Mixed-fleet Flying that flight crew operate aircraft of different types. Basically an operator shall not assign a pilot-in-command or a co-pilot to operate at the flight controls of a type of airplane during take-off and landing unless that pilot has operated the flight controls during at least three take-offs and landings within the preceding 90 days on the same type of airplane or in a flight simulator. Also, flight crew members are familiarized with the significant differences in equipment and/or procedures between concurrently operated types. An operator shall ensure that piloting technique and the ability to execute emergency procedures is checked in such a way as to demonstrate the pilot's competence on each type or variant of a type of airplane. Proficiency check shall be performed periodically. When an operator schedules flight crew on different types of airplanes with similar characteristics in terms of operating procedures, systems and handling, the State shall decide the requirements for each type of airplane can be combined. In conclusion, it is necessary for flight crew members to remain concurrently qualified to operate multiple types. The operator shall have a program to include, as a minimum, required differences training between types and qualification to maintain currency on each type. If the Operator utilizes flight crew members to concurrently operate aircraft of different types, the operator shall have qualification processes approved or accepted by the State. If applicable, the qualification curriculum as defined in the operator's Advanced Qualification Program could be applied. Flight crew members are familiarized with the significant differences in equipment and/or procedures between concurrently operated types. The difference among different types of airpcrafts decrease and standards for these airpcrafts can be applied increasingly because function and performance have been improved by aircraft manufacture company in accordance to basic aircraft system in terms of developing new aircrafts for flight standard procedure and safety of flight. Also, it becomes more necessary for flight crews to control multi aircraft types due to various aviation business and activation of leisure business. Nevertheless, in terms of flight crew training and qualification program, there are no regulations in Korea to be applied to new aircraft types differently in accordance with different levels. In addition, it has no choice different programs based on different levels because there are not provisions to restrict or limit and specific standards to operate at or more than two aircraft types for flight safety. Therefore the aviation authority introduce Flight Standardization and/or Operational Evaluation Board in order to analysis differences among aircraft types. In addition to that, the aviation authority should also improve standard flight evaluation and qualification system among different aircraft types for flight crews to apply reasonable training and qualification efficiently. For all the issue mentioned above, I have studied the ICAO SARPs and some state's regulation concerning operating aircraft of different types(Mixed-fleet flying), and suggested some proposals on the different aircraft type operation as an example of comprehensive problem solving. I hope that this paper is 1) to help understanding about the international issue, 2) to help the improvement of korean aviation regulations, 3) to help compliance with international standards and to contribute to the promotion of aviation safety, in addition.

• Korean Journal of Agricultural Science
• /
• v.2 no.2
• /
• pp.341-356
• /
• 1975

This study was carried out to evaluate three winter active synthetic varieties in a succeeding generations of improvement and polycross progenies of seven genotypes selected at the cool and wet climate of the Western Oregon, in their performance of the polycross progeny test comparing with a control variety, high yielding 'Fawn', at Daejon, Korea. Various plant and leaf characteristics, especially related to photosynthesis, and forage production during the first summer after the establishment were examined. The important conclusions of this study are summarized as follows: 1. The differences of leaf fresh weight among groups and control exhibit genetic differences. The a verage of leaf fresh weight of polycross progeny group was the heaviest and those of winter active synthetic varieties in the succeeding generations of improvement was heavier than variety 'fawn'. Within polycross progeny group the genotypes exhibit genetic differences for leaf dry weight. 2. The leaf area exhibited genetic differences among groups and control. The average of winter active synthetic varieties in a succeeding generation was larger than variety 'Fawn'. Those oi the polycross progeny group was the largest among groups and control. 3. Differences of specific leaf weight(S. L. W.) among and within varieties, genotypes and control were not significant. Further investigation in this respect is necessary through the study of the diurnal change in S. L. W. 4. Differences of leaf width among groups and control exhibited genetic differences. The average leaf width of winter active varieties was larger than those of 'Fawn' variety. And those of polycross progenies of genotypes was the largest. 5. Plant height of 'fawn' variety in the first measurement was higher than those of winter active tall fescue varieties and genotypes. The deviation in plant height among polyeross progenies of seven genotypes gave a great deviation. The regrowth ability of plant height was not different suggesting that this characteristics was about the same among and within groups and control. 6. Plant width, spreading ability, improved through the succeeding generations of the improvement of the winter active synthetic varieties for the first measurement. Differences of plant width at the second measurement among genotypes within polycross progeny group were big enough to show the genetic difference. 7. Tiller number of the winter active synthetic varieties and the average of genotypes in polycross progeny was more than those of the control 'Fawn' in the first measurement. On the second measurement, the differences of tiller number appeared among three synthetic varieties indicating improvement, and there were genetic differences among seven genotypes in polycross progeny test. 8. Forage yield on the first cutting showed a considerble improvement of forage yield in the more advanced generation of synthetic varieties and genetic differences among seven genotypes in the polycross progeny test. The average of polycross progeny group was higher than those of the control or three winter active varieties. It was suggested that we could make a further improvement for the forage yield. 9. The regrowth ability of these winter active varieties and genotypes was about the same capacity at least on the measurement of the regrowth in forage yield and plant height during summer. 10. On the whole, the averages of the polycross progeny group was in the highest value and those of synthetic varieties were higher than the control variety, 'Fawn', for the most characteristics except S. L. W. and the plant height on the first measurement even though the differences were not always significant. And there were genetic differences among seven gentypes in their performance of the polycross progeny. 11. Although it was not always sgnificant, the most advanced winter active variety, '1002', had in the highest value for all plant characteristics and forage yield measurements than the other two varieties, '1001'. 12. The results of the association study among various characteristics were quite agreeable and would be useful in the selection of desirable genotypes for the development of a better variety.

• Journal of Intelligence and Information Systems
• /
• v.25 no.3
• /
• pp.117-140
• /
• 2019

As the aging phenomenon accelerates and various social problems related to the elderly of the vulnerable are raised, the need for effective elderly care solutions to protect the health and safety of the elderly generation is growing. Recently, more and more people are using Smart Toys equipped with ICT technology for care for elderly. In particular, log data collected through smart toys is highly valuable to be used as a quantitative and objective indicator in areas such as policy-making and service planning. However, research related to smart toys is limited, such as the development of smart toys and the validation of smart toy effectiveness. In other words, there is a dearth of research to derive insights based on log data collected through smart toys and to use them for decision making. This study will analyze log data collected from smart toy and derive effective insights to improve the quality of life for elderly users. Specifically, the user profiling-based analysis and elicitation of a change in quality of life mechanism based on behavior were performed. First, in the user profiling analysis, two important dimensions of classifying the type of elderly group from five factors of elderly user's living management were derived: 'Routine Activities' and 'Work-out Activities'. Based on the dimensions derived, a hierarchical cluster analysis and K-Means clustering were performed to classify the entire elderly user into three groups. Through a profiling analysis, the demographic characteristics of each group of elderlies and the behavior of using smart toy were identified. Second, stepwise regression was performed in eliciting the mechanism of change in quality of life. The effects of interaction, content usage, and indoor activity have been identified on the improvement of depression and lifestyle for the elderly. In addition, it identified the role of user performance evaluation and satisfaction with smart toy as a parameter that mediated the relationship between usage behavior and quality of life change. Specific mechanisms are as follows. First, the interaction between smart toy and elderly was found to have an effect of improving the depression by mediating attitudes to smart toy. The 'Satisfaction toward Smart Toy,' a variable that affects the improvement of the elderly's depression, changes how users evaluate smart toy performance. At this time, it has been identified that it is the interaction with smart toy that has a positive effect on smart toy These results can be interpreted as an elderly with a desire to meet emotional stability interact actively with smart toy, and a positive assessment of smart toy, greatly appreciating the effectiveness of smart toy. Second, the content usage has been confirmed to have a direct effect on improving lifestyle without going through other variables. Elderly who use a lot of the content provided by smart toy have improved their lifestyle. However, this effect has occurred regardless of the attitude the user has toward smart toy. Third, log data show that a high degree of indoor activity improves both the lifestyle and depression of the elderly. The more indoor activity, the better the lifestyle of the elderly, and these effects occur regardless of the user's attitude toward smart toy. In addition, elderly with a high degree of indoor activity are satisfied with smart toys, which cause improvement in the elderly's depression. However, it can be interpreted that elderly who prefer outdoor activities than indoor activities, or those who are less active due to health problems, are hard to satisfied with smart toys, and are not able to get the effects of improving depression. In summary, based on the activities of the elderly, three groups of elderly were identified and the important characteristics of each type were identified. In addition, this study sought to identify the mechanism by which the behavior of the elderly on smart toy affects the lives of the actual elderly, and to derive user needs and insights.

• 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.26 no.3
• /
• pp.109-125
• /
• 2020

The Ministry of National Defense is pushing for the Defense Acquisition Program to build strong defense capabilities, and it spends more than 10 trillion won annually on defense improvement. As the Defense Acquisition Program is directly related to the security of the nation as well as the lives and property of the people, it must be carried out very transparently and efficiently by experts. However, the excessive diversification of laws and regulations related to the Defense Acquisition Program has made it challenging for many working-level officials to carry out the Defense Acquisition Program smoothly. It is even known that many people realize that there are related regulations that they were unaware of until they push ahead with their work. In addition, the statutory statements related to the Defense Acquisition Program have the tendency to cause serious issues even if only a single expression is wrong within the sentence. Despite this, efforts to establish a sentence comparison system to correct this issue in real time have been minimal. Therefore, this paper tries to propose a "Comparison System between the Statement of Military Reports and Related Laws" implementation plan that uses the Siamese Network-based artificial neural network, a model in the field of natural language processing (NLP), to observe the similarity between sentences that are likely to appear in the Defense Acquisition Program related documents and those from related statutory provisions to determine and classify the risk of illegality and to make users aware of the consequences. Various artificial neural network models (Bi-LSTM, Self-Attention, D_Bi-LSTM) were studied using 3,442 pairs of "Original Sentence"(described in actual statutes) and "Edited Sentence"(edited sentences derived from "Original Sentence"). Among many Defense Acquisition Program related statutes, DEFENSE ACQUISITION PROGRAM ACT, ENFORCEMENT RULE OF THE DEFENSE ACQUISITION PROGRAM ACT, and ENFORCEMENT DECREE OF THE DEFENSE ACQUISITION PROGRAM ACT were selected. Furthermore, "Original Sentence" has the 83 provisions that actually appear in the Act. "Original Sentence" has the main 83 clauses most accessible to working-level officials in their work. "Edited Sentence" is comprised of 30 to 50 similar sentences that are likely to appear modified in the county report for each clause("Original Sentence"). During the creation of the edited sentences, the original sentences were modified using 12 certain rules, and these sentences were produced in proportion to the number of such rules, as it was the case for the original sentences. After conducting 1 : 1 sentence similarity performance evaluation experiments, it was possible to classify each "Edited Sentence" as legal or illegal with considerable accuracy. In addition, the "Edited Sentence" dataset used to train the neural network models contains a variety of actual statutory statements("Original Sentence"), which are characterized by the 12 rules. On the other hand, the models are not able to effectively classify other sentences, which appear in actual military reports, when only the "Original Sentence" and "Edited Sentence" dataset have been fed to them. The dataset is not ample enough for the model to recognize other incoming new sentences. Hence, the performance of the model was reassessed by writing an additional 120 new sentences that have better resemblance to those in the actual military report and still have association with the original sentences. Thereafter, we were able to check that the models' performances surpassed a certain level even when they were trained merely with "Original Sentence" and "Edited Sentence" data. If sufficient model learning is achieved through the improvement and expansion of the full set of learning data with the addition of the actual report appearance sentences, the models will be able to better classify other sentences coming from military reports as legal or illegal. Based on the experimental results, this study confirms the possibility and value of building "Real-Time Automated Comparison System Between Military Documents and Related Laws". The research conducted in this experiment can verify which specific clause, of several that appear in related law clause is most similar to the sentence that appears in the Defense Acquisition Program-related military reports. This helps determine whether the contents in the military report sentences are at the risk of illegality when they are compared with those in the law clauses.