• Journal of Intelligence and Information Systems
• /
• v.23 no.2
• /
• pp.71-88
• /
• 2017

Language models were originally developed for speech recognition and language processing. Using a set of example sentences, a language model predicts the next word or character based on sequential input data. N-gram models have been widely used but this model cannot model the correlation between the input units efficiently since it is a probabilistic model which are based on the frequency of each unit in the training set. Recently, as the deep learning algorithm has been developed, a recurrent neural network (RNN) model and a long short-term memory (LSTM) model have been widely used for the neural language model (Ahn, 2016; Kim et al., 2016; Lee et al., 2016). These models can reflect dependency between the objects that are entered sequentially into the model (Gers and Schmidhuber, 2001; Mikolov et al., 2010; Sundermeyer et al., 2012). In order to learning the neural language model, texts need to be decomposed into words or morphemes. Since, however, a training set of sentences includes a huge number of words or morphemes in general, the size of dictionary is very large and so it increases model complexity. In addition, word-level or morpheme-level models are able to generate vocabularies only which are contained in the training set. Furthermore, with highly morphological languages such as Turkish, Hungarian, Russian, Finnish or Korean, morpheme analyzers have more chance to cause errors in decomposition process (Lankinen et al., 2016). Therefore, this paper proposes a phoneme-level language model for Korean language based on LSTM models. A phoneme such as a vowel or a consonant is the smallest unit that comprises Korean texts. We construct the language model using three or four LSTM layers. Each model was trained using Stochastic Gradient Algorithm and more advanced optimization algorithms such as Adagrad, RMSprop, Adadelta, Adam, Adamax, and Nadam. Simulation study was done with Old Testament texts using a deep learning package Keras based the Theano. After pre-processing the texts, the dataset included 74 of unique characters including vowels, consonants, and punctuation marks. Then we constructed an input vector with 20 consecutive characters and an output with a following 21st character. Finally, total 1,023,411 sets of input-output vectors were included in the dataset and we divided them into training, validation, testsets with proportion 70:15:15. All the simulation were conducted on a system equipped with an Intel Xeon CPU (16 cores) and a NVIDIA GeForce GTX 1080 GPU. We compared the loss function evaluated for the validation set, the perplexity evaluated for the test set, and the time to be taken for training each model. As a result, all the optimization algorithms but the stochastic gradient algorithm showed similar validation loss and perplexity, which are clearly superior to those of the stochastic gradient algorithm. The stochastic gradient algorithm took the longest time to be trained for both 3- and 4-LSTM models. On average, the 4-LSTM layer model took 69% longer training time than the 3-LSTM layer model. However, the validation loss and perplexity were not improved significantly or became even worse for specific conditions. On the other hand, when comparing the automatically generated sentences, the 4-LSTM layer model tended to generate the sentences which are closer to the natural language than the 3-LSTM model. Although there were slight differences in the completeness of the generated sentences between the models, the sentence generation performance was quite satisfactory in any simulation conditions: they generated only legitimate Korean letters and the use of postposition and the conjugation of verbs were almost perfect in the sense of grammar. The results of this study are expected to be widely used for the processing of Korean language in the field of language processing and speech recognition, which are the basis of artificial intelligence systems.

• 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.

• Information Systems Review
• /
• v.16 no.3
• /
• pp.161-177
• /
• 2014

Corporate credit rating assessment consists of complicated processes in which various factors describing a company are taken into consideration. Such assessment is known to be very expensive since domain experts should be employed to assess the ratings. As a result, the data-driven corporate credit rating prediction using statistical and artificial intelligence (AI) techniques has received considerable attention from researchers and practitioners. In particular, statistical methods such as multiple discriminant analysis (MDA) and multinomial logistic regression analysis (MLOGIT), and AI methods including case-based reasoning (CBR), artificial neural network (ANN), and multiclass support vector machine (MSVM) have been applied to corporate credit rating.2) Among them, MSVM has recently become popular because of its robustness and high prediction accuracy. In this study, we propose a novel optimized MSVM model, and appy it to corporate credit rating prediction in order to enhance the accuracy. Our model, named 'GAMSVM (Genetic Algorithm-optimized Multiclass Support Vector Machine),' is designed to simultaneously optimize the kernel parameters and the feature subset selection. Prior studies like Lorena and de Carvalho (2008), and Chatterjee (2013) show that proper kernel parameters may improve the performance of MSVMs. Also, the results from the studies such as Shieh and Yang (2008) and Chatterjee (2013) imply that appropriate feature selection may lead to higher prediction accuracy. Based on these prior studies, we propose to apply GAMSVM to corporate credit rating prediction. As a tool for optimizing the kernel parameters and the feature subset selection, we suggest genetic algorithm (GA). GA is known as an efficient and effective search method that attempts to simulate the biological evolution phenomenon. By applying genetic operations such as selection, crossover, and mutation, it is designed to gradually improve the search results. Especially, mutation operator prevents GA from falling into the local optima, thus we can find the globally optimal or near-optimal solution using it. GA has popularly been applied to search optimal parameters or feature subset selections of AI techniques including MSVM. With these reasons, we also adopt GA as an optimization tool. To empirically validate the usefulness of GAMSVM, we applied it to a real-world case of credit rating in Korea. Our application is in bond rating, which is the most frequently studied area of credit rating for specific debt issues or other financial obligations. The experimental dataset was collected from a large credit rating company in South Korea. It contained 39 financial ratios of 1,295 companies in the manufacturing industry, and their credit ratings. Using various statistical methods including the one-way ANOVA and the stepwise MDA, we selected 14 financial ratios as the candidate independent variables. The dependent variable, i.e. credit rating, was labeled as four classes: 1(A1); 2(A2); 3(A3); 4(B and C). 80 percent of total data for each class was used for training, and remaining 20 percent was used for validation. And, to overcome small sample size, we applied five-fold cross validation to our dataset. In order to examine the competitiveness of the proposed model, we also experimented several comparative models including MDA, MLOGIT, CBR, ANN and MSVM. In case of MSVM, we adopted One-Against-One (OAO) and DAGSVM (Directed Acyclic Graph SVM) approaches because they are known to be the most accurate approaches among various MSVM approaches. GAMSVM was implemented using LIBSVM-an open-source software, and Evolver 5.5-a commercial software enables GA. Other comparative models were experimented using various statistical and AI packages such as SPSS for Windows, Neuroshell, and Microsoft Excel VBA (Visual Basic for Applications). Experimental results showed that the proposed model-GAMSVM-outperformed all the competitive models. In addition, the model was found to use less independent variables, but to show higher accuracy. In our experiments, five variables such as X7 (total debt), X9 (sales per employee), X13 (years after founded), X15 (accumulated earning to total asset), and X39 (the index related to the cash flows from operating activity) were found to be the most important factors in predicting the corporate credit ratings. However, the values of the finally selected kernel parameters were found to be almost same among the data subsets. To examine whether the predictive performance of GAMSVM was significantly greater than those of other models, we used the McNemar test. As a result, we found that GAMSVM was better than MDA, MLOGIT, CBR, and ANN at the 1% significance level, and better than OAO and DAGSVM at the 5% significance level.

• Journal of Intelligence and Information Systems
• /
• v.22 no.1
• /
• pp.1-18
• /
• 2016

Owing to the extensive use of Web media and the development of the IT industry, a large amount of data has been generated, shared, and stored. Nowadays, various types of unstructured data such as image, sound, video, and text are distributed through Web media. Therefore, many attempts have been made in recent years to discover new value through an analysis of these unstructured data. Among these types of unstructured data, text is recognized as the most representative method for users to express and share their opinions on the Web. In this sense, demand for obtaining new insights through text analysis is steadily increasing. Accordingly, text mining is increasingly being used for different purposes in various fields. In particular, issue tracking is being widely studied not only in the academic world but also in industries because it can be used to extract various issues from text such as news, (SocialNetworkServices) to analyze the trends of these issues. Conventionally, issue tracking is used to identify major issues sustained over a long period of time through topic modeling and to analyze the detailed distribution of documents involved in each issue. However, because conventional issue tracking assumes that the content composing each issue does not change throughout the entire tracking period, it cannot represent the dynamic mutation process of detailed issues that can be created, merged, divided, and deleted between these periods. Moreover, because only keywords that appear consistently throughout the entire period can be derived as issue keywords, concrete issue keywords such as "nuclear test" and "separated families" may be concealed by more general issue keywords such as "North Korea" in an analysis over a long period of time. This implies that many meaningful but short-lived issues cannot be discovered by conventional issue tracking. Note that detailed keywords are preferable to general keywords because the former can be clues for providing actionable strategies. To overcome these limitations, we performed an independent analysis on the documents of each detailed period. We generated an issue flow diagram based on the similarity of each issue between two consecutive periods. The issue transition pattern among categories was analyzed by using the category information of each document. In this study, we then applied the proposed methodology to a real case of 53,739 news articles. We derived an issue flow diagram from the articles. We then proposed the following useful application scenarios for the issue flow diagram presented in the experiment section. First, we can identify an issue that actively appears during a certain period and promptly disappears in the next period. Second, the preceding and following issues of a particular issue can be easily discovered from the issue flow diagram. This implies that our methodology can be used to discover the association between inter-period issues. Finally, an interesting pattern of one-way and two-way transitions was discovered by analyzing the transition patterns of issues through category analysis. Thus, we discovered that a pair of mutually similar categories induces two-way transitions. In contrast, one-way transitions can be recognized as an indicator that issues in a certain category tend to be influenced by other issues in another category. For practical application of the proposed methodology, high-quality word and stop word dictionaries need to be constructed. In addition, not only the number of documents but also additional meta-information such as the read counts, written time, and comments of documents should be analyzed. A rigorous performance evaluation or validation of the proposed methodology should be performed in future works.

• Cartoon and Animation Studies
• /
• s.33
• /
• pp.597-616
• /
• 2013

Character industry is not only knowledge-based industry chiefly with copyrights but also motive power for creative economy to take a role functionally over the fields of industries because it has industrial characteristic as complement product to promote sale value in manufacturing industry and service industry and increase profit on sales. Since 2003, the national policy related to character has aimed to maximize effect among connected industries, extend its business abroad, enforce copyrights through the improvement of marketing system, develop industrial infrastructure through raising quality of character products. With the result of this policy, the successful cases of connected contents have been crystallized and domestic character industry has stepped up methodically since 2007. It is needed to reset the scales of character industry and industrial stats because there are more know-how of self industry promotion and more related characters through strategy of market departmentalization starting with cartoon, animation, games, novels, movies and musicals. Especially, The Korea government set our target for 'Global Top Five Character Power' since 2009 and has started to carry out to find global star characters, support to establish network among connected industries, diversify promotion channels, and develop licensing business. Particularly, since 2013, There have been prospered the indoor character theme park with time management just like character experimental marketing or Kids cafes using characters, the demand market of digital character focusing on SNS emoticon, and the performance market for character musical consistently. Moreover, The domestic and foreign illegal black markets on off-line have been enlarged, so we need another policy alternative. To prepare for the era of exploding character demand market and diversifying platform, it is needed to set up a solid strategy that is required the elements of policy changes in character industry to vitalize character industry and support new character design and connected contents. the following shows that the elements of policy changes related to the existing policy, the current position of market. Nowadays, the elements of policy changes in domestic character industry are that variety of consumers in the digital character market according to platform diversification, Convergence contents of character goods for the Korean waves, legalization of the illegal black contents market, and controling the tendency of consumers in departmentalized market. This can help find the policy issue entirely deferent with the existing character powers like US, Japan or Europe. In its final analysis, the alternatives are the promotion of models with contract copyrights of domestic and foreign connected contents, the diversification of profit models of platform economy, the additive development of target market related to enlarging the Korean waves, and the strategy of character market for the age-specific tendency according to developing character demand market.

• Korean Security Journal
• /
• no.33
• /
• pp.51-80
• /
• 2012

This study is to summarize main contents which was mentioned by Beomseok Kim' doctoral dissertation. The purpose of this study focuses on presenting the outplacement countermeasure and retention level examination analysis about outplacement competency of special security government official through implement of questionnaire method. The questionnaire for retention level examination including four groups of outplacement competency and twenty subcategories was implemented in the object of six hundered persons relevant to outplacement more than forty age and five grade administration official of special security government officials, who have outplacement experiences as outplacement successors, outplacement losers, and outplacement expectants, in order to achieve this research purpose effectively. The questionnaire examination items are four groups of outplacement competency and twenty subcategories which are the group of knowledge competency & four subcategories including expert knowledge, outplacement knowledge, self comprehension, and organization comprehension, the group of skill competency & nine subcategories including job skill competency, job performance skill, problem-solving skill, reforming skill, communication skill, organization management skill, crisis management skill, career development skill, and human network application skill, the group of attitude-emotion competency & seven subcategories including positive attitude, active attitude, responsibility, professionalism, devoting-sacrificing attitude, affinity, and self-controlling ability, and the group of value-ethics competency & two subcategories including ethical consciousness and morality. The respondents highly regard twenty-two outplacement competency and they consider themselves well-qualified for the subcategories valued over 4.0 such as the professional knowledge, active attitude, responsibility, ethics and morality while they mark the other subcategories below average still need to be improved. Thus, the following is suggestions for successful outplacement. First, individual effort is essential to strengthen their capabilities based on accurate self evaluation, for which the awareness and concept need to be redefined to help them face up to the reality by readjusting career goal to a realistic level. Second, active career development plan to improve shortcoming in terms of outplacement competency is required. Third, it is necessary to establish the infrastructure related to outplacement training such as ON-OFF Line training system and facilities for learning to reinforce user-oriented outplacement training as a regular training course before during after the retirement.

• Journal of Internet Computing and Services
• /
• v.21 no.4
• /
• pp.17-23
• /
• 2020

Biometric information indicating measurement items related to human characteristics has attracted great attention as security technology with high reliability since there is no fear of theft or loss. Among these biometric information, fingerprints are mainly used in fields such as identity verification and identification. If there is a problem such as a wound, wrinkle, or moisture that is difficult to authenticate to the fingerprint image when identifying the identity, the fingerprint expert can identify the problem with the fingerprint directly through the preprocessing step, and apply the image processing algorithm appropriate to the problem. Solve the problem. In this case, by implementing artificial intelligence software that distinguishes fingerprint images with cuts and wrinkles on the fingerprint, it is easy to check whether there are cuts or wrinkles, and by selecting an appropriate algorithm, the fingerprint image can be easily improved. In this study, we developed a total of 17,080 fingerprint databases by acquiring all finger prints of 1,010 students from the Royal University of Cambodia, 600 Sokoto open data sets, and 98 Korean students. In order to determine if there are any injuries or wrinkles in the built database, criteria were established, and the data were validated by experts. The training and test datasets consisted of Cambodian data and Sokoto data, and the ratio was set to 8: 2. The data of 98 Korean students were set up as a validation data set. Using the constructed data set, five CNN-based architectures such as Classic CNN, AlexNet, VGG-16, Resnet50, and Yolo v3 were implemented. A study was conducted to find the model that performed best on the readings. Among the five architectures, ResNet50 showed the best performance with 81.51%.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.3
• /
• pp.599-610
• /
• 2016

This paper reports the results of a systematic review (SR) and meta-analysis research to compare the effect of Kangaroo care, targeting mothers and premature infants. A randomized clinical trial study was performed until February 2015. The domestic literature contained the non-randomized clinical trial research without restriction according to the level of the study design. A search of the Ovid-Medline, CINAHL, PubMed and KoreaMed, the National Library of KOREA, the National Assembly Library, NDSL, KISS and RISS. Through the KMbase we searched and combined the main term ((kangaroo OR KC OR skin-to-skin) AND (care OR contact)) AND (infant OR preterm OR Low Birth Weight OR LBW), ((kangaroo OR kangaroo OR kangaroo) AND (care OR nursing care OR management OR skin contact)) was made; these were all combined with a keywords search through the selection process. They were excluded in the final 25 studies (n=3051). A methodology checklist for randomized controlled trials (RCTs) designed by SIGN (Scottish Intercollegiate Guidelines Network) was utilized to assess the risk of bias. The overall risk of bias was regarded as low. In 16 studies that were evaluated as a grade of "++", 9 studies were evaluated as a grade of "+". As a result of meta-analysis, kangaroo care regarding the effects of premature mortality, severe infection/sepsis had an insignificant effect. Hyperthermia incidence, growth and development (height and weight), mother-infant attachment, hypothermia incidence, length of hospital days, breast feeding rate, sleeping, anxiety, confidence, and gratification of mothering role were considered significant. In satisfaction of the role performance, depression and stress presented contradictory research results for individual studies showing overall significant difference. This study has some limitations due to the few RCTs comparing kangaroo care in the country. Therefore, further RCTs comparing kangaroo care should be conducted.

• Archives of design research
• /
• v.18 no.2 s.60
• /
• pp.37-46
• /
• 2005

Today's advertisement is dominated by visual images, which deliver messages in the most efficient and engaging way. Vivid and live images, through multimedia, attract the viewer to consume the products being advertised. In this course, the effect of 'Verfremdung' or alienation is often detected. This terminology was at first used in a epic play of Bertolt Brecht to signify a specific effect to block the empathy of the audience by reminding them of their position as the audience separate from the play. 'Verfremdung' helps them to have a sense of objectivity and critical attitude toward the performance. Multimedia commercials nowadays exhibit such a quality together with 'viewer participation:' criticism on reality: and 'speed.' In an advertisement for Shinsegi Communications' mobile system titled <017-I-Touch series, the hand transformed into a web>, the copy, 'my blood type is 'i', ' is accompanied by six unrelated fictitious scenes in which the same ocean appears as a common denominator. Because there is no connection between the scenes, free imagination of the viewer has to be involved and thus plays a significant role in making them into a context. This fact dearly exhibit some characteristics of post- modern advertisement. Momentary 'Gap' and 'Difference' between scenes contribute to 'Verfremdung' or alienation' that makes it hard for the consumers to comprehend the content on the spot. Such an uneasy situation, however, keeps the viewers thinking about the advertisement itself. While repeatedly exposed to the alienated images, the viewers come to get involved in the advertisement, trying to make the fragmented images into a coherent context. In addition, the leaps between the scenes produce a sense of 'speed' in a context, which adds more impact to the way of delivering messages using multimedia. With the help of multimedia, 'Verfremdung: which was originally intended to bring about objective and critical altitude of the audience in a play, plays a crucial role in attracting the viewer's attention and conveying a specific message in a moment in contemporary advertisement.

• Journal of KIISE:Computing Practices and Letters
• /
• v.13 no.5
• /
• pp.293-299
• /
• 2007

Intelligent pipeline inspection gauges (PIGs) are inspection vehicles that move along within a (gas or oil) pipeline and acquire signals (also called sensor data) from their surrounding rings of sensors. By analyzing the signals captured in intelligent PIGs, we can detect pipeline defects, such as holes and curvatures and other potential causes of gas explosions. There are two major data access patterns apparent when an analyzer accesses the pipeline signal data. The first is a sequential pattern where an analyst reads the sensor data one time only in a sequential fashion. The second is the repetitive pattern where an analyzer repeatedly reads the signal data within a fixed range; this is the dominant pattern in analyzing the signal data. The existing PIG software reads signal data directly from the server at every user#s request, requiring network transfer and disk access cost. It works well only for the sequential pattern, but not for the more dominant repetitive pattern. This problem becomes very serious in a client/server environment where several analysts analyze the signal data concurrently. To tackle this problem, we devise a fast in-memory cache manager, called T-Cache, by considering pipeline sensor data as multiple time-series data and by efficiently caching the time-series data at T-Cache. To the best of the authors# knowledge, this is the first research on caching pipeline signals on the client-side. We propose a new concept of the signal cache line as a caching unit, which is a set of time-series signal data for a fixed distance. We also provide the various data structures including smart cursors and algorithms used in T-Cache. Experimental results show that T-Cache performs much better for the repetitive pattern in terms of disk I/Os and the elapsed time. Even with the sequential pattern, T-Cache shows almost the same performance as a system that does not use any caching, indicating the caching overhead in T-Cache is negligible.