• Journal of Intelligence and Information Systems
• /
• v.24 no.3
• /
• pp.21-44
• /
• 2018

In recent years, the rapid development of internet technology and the popularization of smart devices have resulted in massive amounts of text data. Those text data were produced and distributed through various media platforms such as World Wide Web, Internet news feeds, microblog, and social media. However, this enormous amount of easily obtained information is lack of organization. Therefore, this problem has raised the interest of many researchers in order to manage this huge amount of information. Further, this problem also required professionals that are capable of classifying relevant information and hence text classification is introduced. Text classification is a challenging task in modern data analysis, which it needs to assign a text document into one or more predefined categories or classes. In text classification field, there are different kinds of techniques available such as K-Nearest Neighbor, Naïve Bayes Algorithm, Support Vector Machine, Decision Tree, and Artificial Neural Network. However, while dealing with huge amount of text data, model performance and accuracy becomes a challenge. According to the type of words used in the corpus and type of features created for classification, the performance of a text classification model can be varied. Most of the attempts are been made based on proposing a new algorithm or modifying an existing algorithm. This kind of research can be said already reached their certain limitations for further improvements. In this study, aside from proposing a new algorithm or modifying the algorithm, we focus on searching a way to modify the use of data. It is widely known that classifier performance is influenced by the quality of training data upon which this classifier is built. The real world datasets in most of the time contain noise, or in other words noisy data, these can actually affect the decision made by the classifiers built from these data. In this study, we consider that the data from different domains, which is heterogeneous data might have the characteristics of noise which can be utilized in the classification process. In order to build the classifier, machine learning algorithm is performed based on the assumption that the characteristics of training data and target data are the same or very similar to each other. However, in the case of unstructured data such as text, the features are determined according to the vocabularies included in the document. If the viewpoints of the learning data and target data are different, the features may be appearing different between these two data. In this study, we attempt to improve the classification accuracy by strengthening the robustness of the document classifier through artificially injecting the noise into the process of constructing the document classifier. With data coming from various kind of sources, these data are likely formatted differently. These cause difficulties for traditional machine learning algorithms because they are not developed to recognize different type of data representation at one time and to put them together in same generalization. Therefore, in order to utilize heterogeneous data in the learning process of document classifier, we apply semi-supervised learning in our study. However, unlabeled data might have the possibility to degrade the performance of the document classifier. Therefore, we further proposed a method called Rule Selection-Based Ensemble Semi-Supervised Learning Algorithm (RSESLA) to select only the documents that contributing to the accuracy improvement of the classifier. RSESLA creates multiple views by manipulating the features using different types of classification models and different types of heterogeneous data. The most confident classification rules will be selected and applied for the final decision making. In this paper, three different types of real-world data sources were used, which are news, twitter and blogs.

• Journal of dental hygiene science
• /
• v.18 no.2
• /
• pp.97-104
• /
• 2018

The purpose of this study was to classify risk groups according to Caries Risk Assessment (CRA) and to investigate the effect of caries prevention program after 1 year of caries prevention intervention program in 6-year-old infants with high caries risk. The subjects were selected based on responses to CRA questionnaires. At the first visit, oral examination, Cariview, bacteria and saliva flow test were performed. The caries risk group was classified accordingly. The subjects were given fluoride application and oral health education every four months and evaluated the same as the first visit after 1 year. As a result of classifying the risk level according to CRA, more than 80% of the subjects were in the high or extreme high risk. The dft index was increased in all risk groups after the intervention. There was a significant difference between the before and after intervention (p<0.05). The Cariview score showed a slight decrease after the intervention in the moderate and high risk groups. As a result of the evaluation of bacteria test, Streptococcus mutans were decreased to ${\geq}10^5CFU/ml$ saliva after intervention in all groups. Lactobacilli were decreased after intervention in high risk and extreme high risk groups. As a result of saliva flow, there was significant difference between caries risk groups before and after intervention (p<0.05). In conclusion, regular caries management has been shown to influence caries risk factors in high-caries risk children. Also, it is necessary to find out periodical dental risk management system which is suitable for domestic situation through the related studies.

• Journal of Intelligence and Information Systems
• /
• v.21 no.1
• /
• pp.29-45
• /
• 2015

Response modeling is a well-known research issue for those who have tried to get more superior performance in the capability of predicting the customers' response for the marketing promotion. The response model for customers would reduce the marketing cost by identifying prospective customers from very large customer database and predicting the purchasing intention of the selected customers while the promotion which is derived from an undifferentiated marketing strategy results in unnecessary cost. In addition, the big data environment has accelerated developing the response model with data mining techniques such as CBR, neural networks and support vector machines. And CBR is one of the most major tools in business because it is known as simple and robust to apply to the response model. However, CBR is an attractive data mining technique for data mining applications in business even though it hasn't shown high performance compared to other machine learning techniques. Thus many studies have tried to improve CBR and utilized in business data mining with the enhanced algorithms or the support of other techniques such as genetic algorithm, decision tree and AHP (Analytic Process Hierarchy). Ahn and Kim(2008) utilized logit, neural networks, CBR to predict that which customers would purchase the items promoted by marketing department and tried to optimized the number of k for k-nearest neighbor with genetic algorithm for the purpose of improving the performance of the integrated model. Hong and Park(2009) noted that the integrated approach with CBR for logit, neural networks, and Support Vector Machine (SVM) showed more improved prediction ability for response of customers to marketing promotion than each data mining models such as logit, neural networks, and SVM. This paper presented an approach to predict customers' response of marketing promotion with Case Based Reasoning. The proposed model was developed by applying different weights to each feature. We deployed logit model with a database including the promotion and the purchasing data of bath soap. After that, the coefficients were used to give different weights of CBR. We analyzed the performance of proposed weighted CBR based model compared to neural networks and pure CBR based model empirically and found that the proposed weighted CBR based model showed more superior performance than pure CBR model. Imbalanced data is a common problem to build data mining model to classify a class with real data such as bankruptcy prediction, intrusion detection, fraud detection, churn management, and response modeling. Imbalanced data means that the number of instance in one class is remarkably small or large compared to the number of instance in other classes. The classification model such as response modeling has a lot of trouble to recognize the pattern from data through learning because the model tends to ignore a small number of classes while classifying a large number of classes correctly. To resolve the problem caused from imbalanced data distribution, sampling method is one of the most representative approach. The sampling method could be categorized to under sampling and over sampling. However, CBR is not sensitive to data distribution because it doesn't learn from data unlike machine learning algorithm. In this study, we investigated the robustness of our proposed model while changing the ratio of response customers and nonresponse customers to the promotion program because the response customers for the suggested promotion is always a small part of nonresponse customers in the real world. We simulated the proposed model 100 times to validate the robustness with different ratio of response customers to response customers under the imbalanced data distribution. Finally, we found that our proposed CBR based model showed superior performance than compared models under the imbalanced data sets. Our study is expected to improve the performance of response model for the promotion program with CBR under imbalanced data distribution in the real world.

• Journal of Intelligence and Information Systems
• /
• v.25 no.4
• /
• pp.67-88
• /
• 2019

Webtoon is a Korean-style digital comics platform that distributes comics content produced using the characteristic elements of the Internet in a form that can be consumed online. With the recent rapid growth of the webtoon industry and the exponential increase in the supply of webtoon content, the need for effective webtoon content recommendation measures is growing. Webtoons are digital content products that combine pictorial, literary and digital elements. Therefore, webtoons stimulate consumer sentiment by making readers have fun and engaging and empathizing with the situations in which webtoons are produced. In this context, it can be expected that the sentiment that webtoons evoke to consumers will serve as an important criterion for consumers' choice of webtoons. However, there is a lack of research to improve webtoons' recommendation performance by utilizing consumer sentiment. This study is aimed at developing consumer sentiment pattern maps that can support effective recommendations of webtoon content, focusing on consumer sentiments that have not been fully discussed previously. Metadata and consumer sentiments data were collected for 200 works serviced on the Korean webtoon platform 'Naver Webtoon' to conduct this study. 488 sentiment terms were collected for 127 works, excluding those that did not meet the purpose of the analysis. Next, similar or duplicate terms were combined or abstracted in accordance with the bottom-up approach. As a result, we have built webtoons specialized sentiment-index, which are reduced to a total of 63 emotive adjectives. By performing exploratory factor analysis on the constructed sentiment-index, we have derived three important dimensions for classifying webtoon types. The exploratory factor analysis was performed through the Principal Component Analysis (PCA) using varimax factor rotation. The three dimensions were named 'Immersion', 'Touch' and 'Irritant' respectively. Based on this, K-Means clustering was performed and the entire webtoons were classified into four types. Each type was named 'Snack', 'Drama', 'Irritant', and 'Romance'. For each type of webtoon, we wrote webtoon-sentiment 2-Mode network graphs and looked at the characteristics of the sentiment pattern appearing for each type. In addition, through profiling analysis, we were able to derive meaningful strategic implications for each type of webtoon. First, The 'Snack' cluster is a collection of webtoons that are fast-paced and highly entertaining. Many consumers are interested in these webtoons, but they don't rate them well. Also, consumers mostly use simple expressions of sentiment when talking about these webtoons. Webtoons belonging to 'Snack' are expected to appeal to modern people who want to consume content easily and quickly during short travel time, such as commuting time. Secondly, webtoons belonging to 'Drama' are expected to evoke realistic and everyday sentiments rather than exaggerated and light comic ones. When consumers talk about webtoons belonging to a 'Drama' cluster in online, they are found to express a variety of sentiments. It is appropriate to establish an OSMU(One source multi-use) strategy to extend these webtoons to other content such as movies and TV series. Third, the sentiment pattern map of 'Irritant' shows the sentiments that discourage customer interest by stimulating discomfort. Webtoons that evoke these sentiments are hard to get public attention. Artists should pay attention to these sentiments that cause inconvenience to consumers in creating webtoons. Finally, Webtoons belonging to 'Romance' do not evoke a variety of consumer sentiments, but they are interpreted as touching consumers. They are expected to be consumed as 'healing content' targeted at consumers with high levels of stress or mental fatigue in their lives. The results of this study are meaningful in that it identifies the applicability of consumer sentiment in the areas of recommendation and classification of webtoons, and provides guidelines to help members of webtoons' ecosystem better understand consumers and formulate strategies.

• Korean Security Journal
• /
• no.28
• /
• pp.153-179
• /
• 2011

Korea has been a victim of State supporting terrorism by North Korea even before international society realize the terrorism threats because of 9.11 in US. However, state supporting terrorism against South Korea by North Korea went along with East/West Cold War System by US and the Soviet Union. It is because socialism that Kim Il-sung who established a separate government in North Korea with the political, economic, social and military support of the Soviet Union selected as his political ideology justifies terrorism as the tool to complete the proletariat revolution. North Korea's state supporting terrorism is being operated systematically and efficiently by military of North Korea. It gives big worries to international society not only by performing terrorism against Korea but also by dispatching terrorists and exporting terrorism strategies to the third world countries. In this situation, terrorism against Korea has met a new transition point at 9${\cdot}$11 in US. As South Korea is confronting North Korea and the war has not ended but suspended, the alliance between US and Korea is more important than anything else. Because of this Korea decided to support the anti-terrorism wars against Afghanistan and Iraq of US and other western countries and send military force there. The preface of the anti-terrorism war has begun as such. On October 7, 2001, US and UK started to attack Afghanistan and Taleban government in Afghanistan was dethroned on December 7, 2001. US and western countries started a war against Iraq on March 20, 2003. On April 9, 2003 Baghdad, the capital of Iraq fell, and Saddam Hussein al-Majid al-Awja government was expelled. During the process, the terrorism threat against South Korea has expanded to Arab terrorists and terrorism organizations as well as North Korea. Consequently, although Korean government, scholars and working level public servants made discussions and tried to seek countermeasures, the damages are extending. Accordingly, terrorism against Korean companies in overseas after 9${\cdot}$11 were analyzed focusing on Nation, Region, Victimology, and Weapons used for the attacks. Especially, the trend of terrorism against the Korean companies in overseas was discussed by classifying them chronologically such as initiation and termination of anti-terrorism wars against Afghanistan and Iraq, and from the execution of Iraqi President, Saddam Hussein al-Majid al-Awja to December 2010. Through this, possible terrorism incidents after the execution of Osama bin Laden, the leader of Al-Qaeda, on May 2, 2011 were projected and proposals were made for the countermeasures.

• Fire Science and Engineering
• /
• v.31 no.4
• /
• pp.95-102
• /
• 2017

The fires in the traditional markets often occur recently with the most of them expanded into great fires so that the damage is very serious. The status of traditional markets handling the distribution for ordinary people is greatly shrunk with the aggressive marketing of the local large companies and the foreign large distribution companies after the overall opening of the local distribution market. Most of the traditional markets have the history and tradition from decades to centuries and have grown steadily with the joys and sorrows of ordinary people and the development of the local economy. The fire developing to the large fire has the characteristics of the problem that the fire possibility is high since all products can be flammable due to the deterioration of facilities, the arbitrary modification of equipment, and the crowding of the goods for sale. Furthermore, most of the stores are petty with their small sizes so that the passage is narrow affecting the passage of pedestrians. Accordingly, the traditional markets are vulnerable to fire due to the initial unplanned structural problem so that the large scale fire damage occurs. The study is concerned with systematically classifying and analyzing the result by applying the TRIZ tool to the fire risk factors to extract the fundamental problem with the fire of the traditional market and make the active response. The study was done for preventing the fire on the basis of it and the expansion to the large fire in case of fire to prepare the specific measure to minimize the fire damage. On the basis of the fire expansion risk factor of the derived traditional market, the study presented the passive measures such as the improvement of the fire resisting capacity, the fire safety island, etc. and the active and institutional measures such as the obligation of the fire breaking news facilities, the application of the extra-high pressure pump system, the divided use of the electric line, etc.

• Journal of Intelligence and Information Systems
• /
• v.22 no.2
• /
• pp.127-142
• /
• 2016

Deep learning model is a kind of neural networks that allows multiple hidden layers. There are various deep learning architectures such as convolutional neural networks, deep belief networks and recurrent neural networks. Those have been applied to fields like computer vision, automatic speech recognition, natural language processing, audio recognition and bioinformatics where they have been shown to produce state-of-the-art results on various tasks. Among those architectures, convolutional neural networks and recurrent neural networks are classified as the supervised learning model. And in recent years, those supervised learning models have gained more popularity than unsupervised learning models such as deep belief networks, because supervised learning models have shown fashionable applications in such fields mentioned above. Deep learning models can be trained with backpropagation algorithm. Backpropagation is an abbreviation for "backward propagation of errors" and a common method of training artificial neural networks used in conjunction with an optimization method such as gradient descent. The method calculates the gradient of an error function with respect to all the weights in the network. The gradient is fed to the optimization method which in turn uses it to update the weights, in an attempt to minimize the error function. Convolutional neural networks use a special architecture which is particularly well-adapted to classify images. Using this architecture makes convolutional networks fast to train. This, in turn, helps us train deep, muti-layer networks, which are very good at classifying images. These days, deep convolutional networks are used in most neural networks for image recognition. Convolutional neural networks use three basic ideas: local receptive fields, shared weights, and pooling. By local receptive fields, we mean that each neuron in the first(or any) hidden layer will be connected to a small region of the input(or previous layer's) neurons. Shared weights mean that we're going to use the same weights and bias for each of the local receptive field. This means that all the neurons in the hidden layer detect exactly the same feature, just at different locations in the input image. In addition to the convolutional layers just described, convolutional neural networks also contain pooling layers. Pooling layers are usually used immediately after convolutional layers. What the pooling layers do is to simplify the information in the output from the convolutional layer. Recent convolutional network architectures have 10 to 20 hidden layers and billions of connections between units. Training deep learning networks has taken weeks several years ago, but thanks to progress in GPU and algorithm enhancement, training time has reduced to several hours. Neural networks with time-varying behavior are known as recurrent neural networks or RNNs. A recurrent neural network is a class of artificial neural network where connections between units form a directed cycle. This creates an internal state of the network which allows it to exhibit dynamic temporal behavior. Unlike feedforward neural networks, RNNs can use their internal memory to process arbitrary sequences of inputs. Early RNN models turned out to be very difficult to train, harder even than deep feedforward networks. The reason is the unstable gradient problem such as vanishing gradient and exploding gradient. The gradient can get smaller and smaller as it is propagated back through layers. This makes learning in early layers extremely slow. The problem actually gets worse in RNNs, since gradients aren't just propagated backward through layers, they're propagated backward through time. If the network runs for a long time, that can make the gradient extremely unstable and hard to learn from. It has been possible to incorporate an idea known as long short-term memory units (LSTMs) into RNNs. LSTMs make it much easier to get good results when training RNNs, and many recent papers make use of LSTMs or related ideas.

• Korean Journal of Social Welfare Studies
• /
• v.48 no.3
• /
• pp.171-208
• /
• 2017

The purpose of this study is to classify the underlying and parsimonious types of child maltreatment and examine whether the effects of risk factors on child maltreatment recurrence differ by type of maltreatment. We utilized the multiyear national administrative data from the National Child Maltreatment Information System collected by Child Protection Agency in Korea. Of 26,921 child maltreatment victims reported and substantiated on or after January 1, 2012, 1,447 children who had recurrence of child maltreatment until December 31, 2015 were selected as maltreatment recurrence group and 4,580 children who had not experienced maltreatment since first substantiation were assigned as maltreatment non-recurrence group. Latent class analysis(LCA) and latent transition analysis(LTA) were used to group children with similar maltreatment subtypes into discrete classes of child maltreatment recurrence. Logistic regression is employed to examine the association between the child maltreatment predominant types and risk factors for recurrence. Results of LCA and LTA showed four latent classes representing predominant type of child maltreatment: 'physical abuse predominant type', 'emotional abuse predominant type', 'sexual abuse predominant type', and 'neglect type'. Significant differences in the effect of risk factors among latent classes were found in child's age and gender, perpetrator's gender, family poverty, biological parent as the perpetrator, domestic violence toward partner, perpetrator's alcoholic problem, insufficient parenting skills, and out-of-home care service, Based on these findings, results suggested how the typology can be used to guide decision about who to target in prevention and intervention programs, and which features of risk factors to target. Practice and policy implications as well as further research tasks were discussed in the lights of searching for useful and important strategies to prevent recurrence of child maltreatment.

• Sijohaknonchong
• /
• v.24
• /
• pp.223-258
• /
• 2006

This Research is about the classification of 'Sijochang', or the Korean ode narrative song, in terms of music. Contrary to the literature classification by the number of letters, sijochang varies with the melody. Literally, the classification is generally made as Dansijo(or short ode) Jungsijo(or medium ode) and Jangsijo(or lengthy ode) but the sijochang is normally divided into 'Pyongsijo' (or plain ode), 'Jirumsijo and Saseolsijo'. As while the same Sijochang is called under the different names, the different type of sijochang is also called as the same name, it needs the discussion about its name. Some Korean classical musicians have attempted to define it but they are trying to do it without the specific reasoning about its concept. As a result, the systematic research is required. This study designs to streamline the currently confusing and complex names and set up the sijo's classification system. After reviewing the ancient music note, current sijo score and the traditional theory, I largely classified it into 3 types: Pyongsijo, Jirumsijo and Saseolsijo. And then, 1 analyzed on to which type the sijochang which is presently called belongs, based on several principles. The 67 names of the sijo which I have investigated about are classified with them sharpy reduced into 16. Among the current sijo names. there are some which are of same type yet of different phonetics and there are others which are of different phonetics yet of same type. To avoid such complex and troublesome names, I have orchestrated them as follows, taking the literary and music concept into account. 1) Pyongsijo type : Pyongsijo, Joongherisijo, Wujosijo and Payeonkok 2) Jirumsijo type: Jirumsiro, Namchangjirumsijo(it refers to Jirumsijo sung by male ), Yeochangjirumsijo (it refers to Jirumsijo sung by female), Banjirumsijo(it refers to half the Jiumsiro), Onjirumsijo (it refers to the whole Jirumsijo), Wujojr\irumsijo, Saseoljirumsijo and Whimorisijo) 3) Saseolsijo type : Saseolsijo, Bansaseolsijo(it refers to half the Saseolsijo, Gaksijo or Pyongsiro There are still lots of things to musically streamline, in the fields of disposition of Sijo letters, its form, musical scale and influences on other genre. etc. and as such. the accumulation of theory on them is urgently required. Those musical elements need an in-depth review and study by the experts and the Korean traditional musicians. Later research is expected to play a role of exploring it.

• The Korean Journal of Archival Studies
• /
• no.16
• /
• pp.39-96
• /
• 2007

Traditionally, the concept of transfer is that physical records such as paper documents, videos, photos are made a delivery to Archives or Records centers on the basis of transfer guidelines. But, with the automation of records management environment and spreading new records creation and management applications, we can create records and manage them in the cyberspace. In these reasons, the existing transfer system is that we move filed records to Archives or Records centers by paper boxes, needs to be changed. Under the needing conditions of a new transfer paradigm, the fact that the revision of Records Act that include some provisions about electronic records management and transfer, is desirable and proper. Nevertheless, the electronic transfer provisions are too conceptional to apply records management practice, so we have to develop detailed methods and processes. In this context, this paper suggest that a electronic records transfer process model on the basis of international standard and foreign countries' cases. Doing transfer records is one of the records management courses to use valuable records in the future. So, both producer and archive have to transfer records itself and context information to long-term preservation repository according to the transfer guidelines. In the long run, transfer comes to be the conclusion that records are moved to archive by a formal transfer process with taking a proper records protection steps. To accomplish these purposes, I analyzed the 'OAIS Reference Model' and 'Producer-Archive Interface Methodology Abstract Standard-CCSDS Blue Book' which is made by CCSDS(Consultative committee for Space Data Systems). but from both the words of 'Reference Model' and 'Standard', we can understand that these standard are not suitable for applying business practice directly. To solve this problem, I also analyzed foreign countries' transfer cases. Through the analysis of theory and case, I suggest that an Electronic Records Transfer Process Model which is consist of five sub-process that are 'Ingest prepare ${\rightarrow}$ Ingest ${\rightarrow}$ Validation ${\rightarrow}$ Preservation ${\rightarrow}$ Archival storage' and each sub-process also have some transfer elements. Especially, to confirm the new process model's feasibility, after classifying two types - one is from Public Records center to Public Archive, the other is from Civil Records center to Public or Civil Archive - of Korean Transfer, I made the new Transfer Model applied to the two types of transfer cases.