• Journal of Intelligence and Information Systems
• /
• v.24 no.1
• /
• pp.167-181
• /
• 2018

Over the past decade, deep learning has been in spotlight among various machine learning algorithms. In particular, CNN(Convolutional Neural Network), which is known as the effective solution for recognizing and classifying images or voices, has been popularly applied to classification and prediction problems. In this study, we investigate the way to apply CNN in business problem solving. Specifically, this study propose to apply CNN to stock market prediction, one of the most challenging tasks in the machine learning research. As mentioned, CNN has strength in interpreting images. Thus, the model proposed in this study adopts CNN as the binary classifier that predicts stock market direction (upward or downward) by using time series graphs as its inputs. That is, our proposal is to build a machine learning algorithm that mimics an experts called 'technical analysts' who examine the graph of past price movement, and predict future financial price movements. Our proposed model named 'CNN-FG(Convolutional Neural Network using Fluctuation Graph)' consists of five steps. In the first step, it divides the dataset into the intervals of 5 days. And then, it creates time series graphs for the divided dataset in step 2. The size of the image in which the graph is drawn is $40(pixels){\times}40(pixels)$, and the graph of each independent variable was drawn using different colors. In step 3, the model converts the images into the matrices. Each image is converted into the combination of three matrices in order to express the value of the color using R(red), G(green), and B(blue) scale. In the next step, it splits the dataset of the graph images into training and validation datasets. We used 80% of the total dataset as the training dataset, and the remaining 20% as the validation dataset. And then, CNN classifiers are trained using the images of training dataset in the final step. Regarding the parameters of CNN-FG, we adopted two convolution filters ($5{\times}5{\times}6$ and $5{\times}5{\times}9$) in the convolution layer. In the pooling layer, $2{\times}2$ max pooling filter was used. The numbers of the nodes in two hidden layers were set to, respectively, 900 and 32, and the number of the nodes in the output layer was set to 2(one is for the prediction of upward trend, and the other one is for downward trend). Activation functions for the convolution layer and the hidden layer were set to ReLU(Rectified Linear Unit), and one for the output layer set to Softmax function. To validate our model - CNN-FG, we applied it to the prediction of KOSPI200 for 2,026 days in eight years (from 2009 to 2016). To match the proportions of the two groups in the independent variable (i.e. tomorrow's stock market movement), we selected 1,950 samples by applying random sampling. Finally, we built the training dataset using 80% of the total dataset (1,560 samples), and the validation dataset using 20% (390 samples). The dependent variables of the experimental dataset included twelve technical indicators popularly been used in the previous studies. They include Stochastic %K, Stochastic %D, Momentum, ROC(rate of change), LW %R(Larry William's %R), A/D oscillator(accumulation/distribution oscillator), OSCP(price oscillator), CCI(commodity channel index), and so on. To confirm the superiority of CNN-FG, we compared its prediction accuracy with the ones of other classification models. Experimental results showed that CNN-FG outperforms LOGIT(logistic regression), ANN(artificial neural network), and SVM(support vector machine) with the statistical significance. These empirical results imply that converting time series business data into graphs and building CNN-based classification models using these graphs can be effective from the perspective of prediction accuracy. Thus, this paper sheds a light on how to apply deep learning techniques to the domain of business problem solving.

• Journal of Intelligence and Information Systems
• /
• v.26 no.3
• /
• pp.149-169
• /
• 2020

"The Urban Renewal New Deal project", one of the government's major national projects, is about developing underdeveloped areas by investing 50 trillion won in 100 locations on the first year and 500 over the next four years. This project is drawing keen attention from the media and local governments. However, the project model which fails to reflect the original characteristics of the area as it divides project area into five categories: "Our Neighborhood Restoration, Housing Maintenance Support Type, General Neighborhood Type, Central Urban Type, and Economic Base Type," According to keywords for successful urban regeneration in Korea, "resident participation," "regional specialization," "ministerial cooperation" and "public-private cooperation", when local governments propose urban regeneration projects to the government, they can see that it is most important to accurately understand the characteristics of the city and push ahead with the projects in a way that suits the characteristics of the city with the help of local residents and private companies. In addition, considering the gentrification problem, which is one of the side effects of urban regeneration projects, it is important to select and implement urban regeneration types suitable for the characteristics of the area. In order to supplement the limitations of the 'Urban Regeneration New Deal Project' methodology, this study aims to propose a system that recommends urban regeneration types suitable for urban regeneration sites by utilizing various machine learning algorithms, referring to the urban regeneration types of the '2025 Seoul Metropolitan Government Urban Regeneration Strategy Plan' promoted based on regional characteristics. There are four types of urban regeneration in Seoul: "Low-use Low-Level Development, Abandonment, Deteriorated Housing, and Specialization of Historical and Cultural Resources" (Shon and Park, 2017). In order to identify regional characteristics, approximately 100,000 text data were collected for 22 regions where the project was carried out for a total of four types of urban regeneration. Using the collected data, we drew key keywords for each region according to the type of urban regeneration and conducted topic modeling to explore whether there were differences between types. As a result, it was confirmed that a number of topics related to real estate and economy appeared in old residential areas, and in the case of declining and underdeveloped areas, topics reflecting the characteristics of areas where industrial activities were active in the past appeared. In the case of the historical and cultural resource area, since it is an area that contains traces of the past, many keywords related to the government appeared. Therefore, it was possible to confirm political topics and cultural topics resulting from various events. Finally, in the case of low-use and under-developed areas, many topics on real estate and accessibility are emerging, so accessibility is good. It mainly had the characteristics of a region where development is planned or is likely to be developed. Furthermore, a model was implemented that proposes urban regeneration types tailored to regional characteristics for regions other than Seoul. Machine learning technology was used to implement the model, and training data and test data were randomly extracted at an 8:2 ratio and used. In order to compare the performance between various models, the input variables are set in two ways: Count Vector and TF-IDF Vector, and as Classifier, there are 5 types of SVM (Support Vector Machine), Decision Tree, Random Forest, Logistic Regression, and Gradient Boosting. By applying it, performance comparison for a total of 10 models was conducted. The model with the highest performance was the Gradient Boosting method using TF-IDF Vector input data, and the accuracy was 97%. Therefore, the recommendation system proposed in this study is expected to recommend urban regeneration types based on the regional characteristics of new business sites in the process of carrying out urban regeneration projects."

• Journal of Intelligence and Information Systems
• /
• v.19 no.2
• /
• pp.125-140
• /
• 2013

We call a data set in which the number of records belonging to a certain class far outnumbers the number of records belonging to the other class, 'imbalanced data set'. Most of the classification techniques perform poorly on imbalanced data sets. When we evaluate the performance of a certain classification technique, we need to measure not only 'accuracy' but also 'sensitivity' and 'specificity'. In a customer churn prediction problem, 'retention' records account for the majority class, and 'churn' records account for the minority class. Sensitivity measures the proportion of actual retentions which are correctly identified as such. Specificity measures the proportion of churns which are correctly identified as such. The poor performance of the classification techniques on imbalanced data sets is due to the low value of specificity. Many previous researches on imbalanced data sets employed 'oversampling' technique where members of the minority class are sampled more than those of the majority class in order to make a relatively balanced data set. When a classification model is constructed using this oversampled balanced data set, specificity can be improved but sensitivity will be decreased. In this research, we developed a hybrid model of support vector machine (SVM), artificial neural network (ANN) and decision tree, that improves specificity while maintaining sensitivity. We named this hybrid model 'hybrid SVM model.' The process of construction and prediction of our hybrid SVM model is as follows. By oversampling from the original imbalanced data set, a balanced data set is prepared. SVM_I model and ANN_I model are constructed using the imbalanced data set, and SVM_B model is constructed using the balanced data set. SVM_I model is superior in sensitivity and SVM_B model is superior in specificity. For a record on which both SVM_I model and SVM_B model make the same prediction, that prediction becomes the final solution. If they make different prediction, the final solution is determined by the discrimination rules obtained by ANN and decision tree. For a record on which SVM_I model and SVM_B model make different predictions, a decision tree model is constructed using ANN_I output value as input and actual retention or churn as target. We obtained the following two discrimination rules: 'IF ANN_I output value <0.285, THEN Final Solution = Retention' and 'IF ANN_I output value ${\geq}0.285$, THEN Final Solution = Churn.' The threshold 0.285 is the value optimized for the data used in this research. The result we present in this research is the structure or framework of our hybrid SVM model, not a specific threshold value such as 0.285. Therefore, the threshold value in the above discrimination rules can be changed to any value depending on the data. In order to evaluate the performance of our hybrid SVM model, we used the 'churn data set' in UCI Machine Learning Repository, that consists of 85% retention customers and 15% churn customers. Accuracy of the hybrid SVM model is 91.08% that is better than that of SVM_I model or SVM_B model. The points worth noticing here are its sensitivity, 95.02%, and specificity, 69.24%. The sensitivity of SVM_I model is 94.65%, and the specificity of SVM_B model is 67.00%. Therefore the hybrid SVM model developed in this research improves the specificity of SVM_B model while maintaining the sensitivity of SVM_I model.

• Journal of Intelligence and Information Systems
• /
• v.24 no.4
• /
• pp.137-154
• /
• 2018

Animal infectious diseases, such as avian influenza and foot and mouth disease, occur almost every year and cause huge economic and social damage to the country. In order to prevent this, the anti-quarantine authorities have tried various human and material endeavors, but the infectious diseases have continued to occur. Avian influenza is known to be developed in 1878 and it rose as a national issue due to its high lethality. Food and mouth disease is considered as most critical animal infectious disease internationally. In a nation where this disease has not been spread, food and mouth disease is recognized as economic disease or political disease because it restricts international trade by making it complex to import processed and non-processed live stock, and also quarantine is costly. In a society where whole nation is connected by zone of life, there is no way to prevent the spread of infectious disease fully. Hence, there is a need to be aware of occurrence of the disease and to take action before it is distributed. Epidemiological investigation on definite diagnosis target is implemented and measures are taken to prevent the spread of disease according to the investigation results, simultaneously with the confirmation of both human infectious disease and animal infectious disease. The foundation of epidemiological investigation is figuring out to where one has been, and whom he or she has met. In a data perspective, this can be defined as an action taken to predict the cause of disease outbreak, outbreak location, and future infection, by collecting and analyzing geographic data and relation data. Recently, an attempt has been made to develop a prediction model of infectious disease by using Big Data and deep learning technology, but there is no active research on model building studies and case reports. KT and the Ministry of Science and ICT have been carrying out big data projects since 2014 as part of national R &D projects to analyze and predict the route of livestock related vehicles. To prevent animal infectious diseases, the researchers first developed a prediction model based on a regression analysis using vehicle movement data. After that, more accurate prediction model was constructed using machine learning algorithms such as Logistic Regression, Lasso, Support Vector Machine and Random Forest. In particular, the prediction model for 2017 added the risk of diffusion to the facilities, and the performance of the model was improved by considering the hyper-parameters of the modeling in various ways. Confusion Matrix and ROC Curve show that the model constructed in 2017 is superior to the machine learning model. The difference between the2016 model and the 2017 model is that visiting information on facilities such as feed factory and slaughter house, and information on bird livestock, which was limited to chicken and duck but now expanded to goose and quail, has been used for analysis in the later model. In addition, an explanation of the results was added to help the authorities in making decisions and to establish a basis for persuading stakeholders in 2017. This study reports an animal infectious disease prevention system which is constructed on the basis of hazardous vehicle movement, farm and environment Big Data. The significance of this study is that it describes the evolution process of the prediction model using Big Data which is used in the field and the model is expected to be more complete if the form of viruses is put into consideration. This will contribute to data utilization and analysis model development in related field. In addition, we expect that the system constructed in this study will provide more preventive and effective prevention.

• Journal of Intelligence and Information Systems
• /
• v.27 no.1
• /
• pp.83-102
• /
• 2021

The government recently announced various policies for developing big-data and artificial intelligence fields to provide a great opportunity to the public with respect to disclosure of high-quality data within public institutions. KSURE(Korea Trade Insurance Corporation) is a major public institution for financial policy in Korea, and thus the company is strongly committed to backing export companies with various systems. Nevertheless, there are still fewer cases of realized business model based on big-data analyses. In this situation, this paper aims to develop a new business model which can be applied to an ex-ante prediction for the likelihood of the insurance accident of credit guarantee. We utilize internal data from KSURE which supports export companies in Korea and apply machine learning models. Then, we conduct performance comparison among the predictive models including Logistic Regression, Random Forest, XGBoost, LightGBM, and DNN(Deep Neural Network). For decades, many researchers have tried to find better models which can help to predict bankruptcy since the ex-ante prediction is crucial for corporate managers, investors, creditors, and other stakeholders. The development of the prediction for financial distress or bankruptcy was originated from Smith(1930), Fitzpatrick(1932), or Merwin(1942). One of the most famous models is the Altman's Z-score model(Altman, 1968) which was based on the multiple discriminant analysis. This model is widely used in both research and practice by this time. The author suggests the score model that utilizes five key financial ratios to predict the probability of bankruptcy in the next two years. Ohlson(1980) introduces logit model to complement some limitations of previous models. Furthermore, Elmer and Borowski(1988) develop and examine a rule-based, automated system which conducts the financial analysis of savings and loans. Since the 1980s, researchers in Korea have started to examine analyses on the prediction of financial distress or bankruptcy. Kim(1987) analyzes financial ratios and develops the prediction model. Also, Han et al.(1995, 1996, 1997, 2003, 2005, 2006) construct the prediction model using various techniques including artificial neural network. Yang(1996) introduces multiple discriminant analysis and logit model. Besides, Kim and Kim(2001) utilize artificial neural network techniques for ex-ante prediction of insolvent enterprises. After that, many scholars have been trying to predict financial distress or bankruptcy more precisely based on diverse models such as Random Forest or SVM. One major distinction of our research from the previous research is that we focus on examining the predicted probability of default for each sample case, not only on investigating the classification accuracy of each model for the entire sample. Most predictive models in this paper show that the level of the accuracy of classification is about 70% based on the entire sample. To be specific, LightGBM model shows the highest accuracy of 71.1% and Logit model indicates the lowest accuracy of 69%. However, we confirm that there are open to multiple interpretations. In the context of the business, we have to put more emphasis on efforts to minimize type 2 error which causes more harmful operating losses for the guaranty company. Thus, we also compare the classification accuracy by splitting predicted probability of the default into ten equal intervals. When we examine the classification accuracy for each interval, Logit model has the highest accuracy of 100% for 0~10% of the predicted probability of the default, however, Logit model has a relatively lower accuracy of 61.5% for 90~100% of the predicted probability of the default. On the other hand, Random Forest, XGBoost, LightGBM, and DNN indicate more desirable results since they indicate a higher level of accuracy for both 0~10% and 90~100% of the predicted probability of the default but have a lower level of accuracy around 50% of the predicted probability of the default. When it comes to the distribution of samples for each predicted probability of the default, both LightGBM and XGBoost models have a relatively large number of samples for both 0~10% and 90~100% of the predicted probability of the default. Although Random Forest model has an advantage with regard to the perspective of classification accuracy with small number of cases, LightGBM or XGBoost could become a more desirable model since they classify large number of cases into the two extreme intervals of the predicted probability of the default, even allowing for their relatively low classification accuracy. Considering the importance of type 2 error and total prediction accuracy, XGBoost and DNN show superior performance. Next, Random Forest and LightGBM show good results, but logistic regression shows the worst performance. However, each predictive model has a comparative advantage in terms of various evaluation standards. For instance, Random Forest model shows almost 100% accuracy for samples which are expected to have a high level of the probability of default. Collectively, we can construct more comprehensive ensemble models which contain multiple classification machine learning models and conduct majority voting for maximizing its overall performance.

• Journal of Intelligence and Information Systems
• /
• v.18 no.3
• /
• pp.185-202
• /
• 2012

.Since the value of information has been realized in the information society, the usage and collection of information has become important. A facial expression that contains thousands of information as an artistic painting can be described in thousands of words. Followed by the idea, there has recently been a number of attempts to provide customers and companies with an intelligent service, which enables the perception of human emotions through one's facial expressions. For example, MIT Media Lab, the leading organization in this research area, has developed the human emotion prediction model, and has applied their studies to the commercial business. In the academic area, a number of the conventional methods such as Multiple Regression Analysis (MRA) or Artificial Neural Networks (ANN) have been applied to predict human emotion in prior studies. However, MRA is generally criticized because of its low prediction accuracy. This is inevitable since MRA can only explain the linear relationship between the dependent variables and the independent variable. To mitigate the limitations of MRA, some studies like Jung and Kim (2012) have used ANN as the alternative, and they reported that ANN generated more accurate prediction than the statistical methods like MRA. However, it has also been criticized due to over fitting and the difficulty of the network design (e.g. setting the number of the layers and the number of the nodes in the hidden layers). Under this background, we propose a novel model using Support Vector Regression (SVR) in order to increase the prediction accuracy. SVR is an extensive version of Support Vector Machine (SVM) designated to solve the regression problems. The model produced by SVR only depends on a subset of the training data, because the cost function for building the model ignores any training data that is close (within a threshold ${\varepsilon}$) to the model prediction. Using SVR, we tried to build a model that can measure the level of arousal and valence from the facial features. To validate the usefulness of the proposed model, we collected the data of facial reactions when providing appropriate visual stimulating contents, and extracted the features from the data. Next, the steps of the preprocessing were taken to choose statistically significant variables. In total, 297 cases were used for the experiment. As the comparative models, we also applied MRA and ANN to the same data set. For SVR, we adopted '${\varepsilon}$-insensitive loss function', and 'grid search' technique to find the optimal values of the parameters like C, d, ${\sigma}^2$, and ${\varepsilon}$. In the case of ANN, we adopted a standard three-layer backpropagation network, which has a single hidden layer. The learning rate and momentum rate of ANN were set to 10%, and we used sigmoid function as the transfer function of hidden and output nodes. We performed the experiments repeatedly by varying the number of nodes in the hidden layer to n/2, n, 3n/2, and 2n, where n is the number of the input variables. The stopping condition for ANN was set to 50,000 learning events. And, we used MAE (Mean Absolute Error) as the measure for performance comparison. From the experiment, we found that SVR achieved the highest prediction accuracy for the hold-out data set compared to MRA and ANN. Regardless of the target variables (the level of arousal, or the level of positive / negative valence), SVR showed the best performance for the hold-out data set. ANN also outperformed MRA, however, it showed the considerably lower prediction accuracy than SVR for both target variables. The findings of our research are expected to be useful to the researchers or practitioners who are willing to build the models for recognizing human emotions.

• Journal of Intelligence and Information Systems
• /
• v.24 no.4
• /
• pp.111-136
• /
• 2018

In this paper, we propose a methodology to extract answer information about queries from various types of unstructured documents collected from multi-sources existing on web in order to expand knowledge base. The proposed methodology is divided into the following steps. 1) Collect relevant documents from Wikipedia, Naver encyclopedia, and Naver news sources for "subject-predicate" separated queries and classify the proper documents. 2) Determine whether the sentence is suitable for extracting information and derive the confidence. 3) Based on the predicate feature, extract the information in the proper sentence and derive the overall confidence of the information extraction result. In order to evaluate the performance of the information extraction system, we selected 400 queries from the artificial intelligence speaker of SK-Telecom. Compared with the baseline model, it is confirmed that it shows higher performance index than the existing model. The contribution of this study is that we develop a sequence tagging model based on bi-directional LSTM-CRF using the predicate feature of the query, with this we developed a robust model that can maintain high recall performance even in various types of unstructured documents collected from multiple sources. The problem of information extraction for knowledge base extension should take into account heterogeneous characteristics of source-specific document types. The proposed methodology proved to extract information effectively from various types of unstructured documents compared to the baseline model. There is a limitation in previous research that the performance is poor when extracting information about the document type that is different from the training data. In addition, this study can prevent unnecessary information extraction attempts from the documents that do not include the answer information through the process for predicting the suitability of information extraction of documents and sentences before the information extraction step. It is meaningful that we provided a method that precision performance can be maintained even in actual web environment. The information extraction problem for the knowledge base expansion has the characteristic that it can not guarantee whether the document includes the correct answer because it is aimed at the unstructured document existing in the real web. When the question answering is performed on a real web, previous machine reading comprehension studies has a limitation that it shows a low level of precision because it frequently attempts to extract an answer even in a document in which there is no correct answer. The policy that predicts the suitability of document and sentence information extraction is meaningful in that it contributes to maintaining the performance of information extraction even in real web environment. The limitations of this study and future research directions are as follows. First, it is a problem related to data preprocessing. In this study, the unit of knowledge extraction is classified through the morphological analysis based on the open source Konlpy python package, and the information extraction result can be improperly performed because morphological analysis is not performed properly. To enhance the performance of information extraction results, it is necessary to develop an advanced morpheme analyzer. Second, it is a problem of entity ambiguity. The information extraction system of this study can not distinguish the same name that has different intention. If several people with the same name appear in the news, the system may not extract information about the intended query. In future research, it is necessary to take measures to identify the person with the same name. Third, it is a problem of evaluation query data. In this study, we selected 400 of user queries collected from SK Telecom 's interactive artificial intelligent speaker to evaluate the performance of the information extraction system. n this study, we developed evaluation data set using 800 documents (400 questions * 7 articles per question (1 Wikipedia, 3 Naver encyclopedia, 3 Naver news) by judging whether a correct answer is included or not. To ensure the external validity of the study, it is desirable to use more queries to determine the performance of the system. This is a costly activity that must be done manually. Future research needs to evaluate the system for more queries. It is also necessary to develop a Korean benchmark data set of information extraction system for queries from multi-source web documents to build an environment that can evaluate the results more objectively.

• Journal of Intelligence and Information Systems
• /
• v.15 no.1
• /
• pp.87-102
• /
• 2009

With the growth of Web 2.0, lots of services allow yours to post their personal information and useful knowledges on networked information spaces such as blogs and online communities etc. As the services are generalized, recent researches related to social network have gained momentum. However, most social network services do not support machine-processable semantic knowledge, so that the information cannot be shared and reused between different domains. Moreover, as explicit definitions of relationships between individual social entities do not be described, it is difficult to analyze social network for inferring unknown semantic relationships. To overcome these limitations, in this paper, we propose a social network analysis system with personal photographic data up-loaded by virtual community users. By using ontology, an informative connectivity between a face entity extracted from photo data and a person entity which already have social relationships was defined clearly and semantic social links were inferred with domain rules. Then the inferred links were provided to yours as a visualized graph. Based on the graph, more efficient social network analysis was achieved in online community.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.19 no.4
• /
• pp.825-840
• /
• 2015

Adaptive bitrate (ABR) streaming technology has become an important and prevalent feature in many multimedia delivery systems, with content providers such as Netflix and Amazon using ABR streaming to increase bandwidth efficiency and provide the maximum user experience when channel conditions are not ideal. Where such systems could see improvement is in the delivery of live video with a closed loop cognitive control of video encoding. In this paper, we present streaming camera system which provides spatially and temporally adaptive video streams, learning the user's preferences in order to make intelligent scaling decisions. The system employs a hardware based H.264/AVC encoder for video compression. The encoding parameters can be configured by the user or by the cognitive system on behalf of the user when the bandwidth changes. A cognitive video client developed in this study learns the user's preferences(i.e. video size over frame rate) over time and intelligently adapts encoding parameters when the channel conditions change. It has been demonstrated that the cognitive decision system developed has the ability to control video bandwidth by altering the spatial and temporal resolution, as well as the ability to make scaling decisions.

• Journal of Broadcast Engineering
• /
• v.17 no.2
• /
• pp.270-283
• /
• 2012

With the advent of multi-channel TV, IPTV and smart TV services, excessive amounts of TV program contents become available at users' sides, which makes it very difficult for TV viewers to easily find and consume their preferred TV programs. Therefore, the service of automatic TV recommendation is an important issue for TV users for future intelligent TV services, which allows to improve access to their preferred TV contents. In this paper, we present a recommendation model based on statistical machine learning using a collaborative filtering concept by taking in account both public and personal preferences on TV program contents. For this, users' preference on TV programs is modeled as a latent topic variable using LDA (Latent Dirichlet Allocation) which is recently applied in various application domains. To apply LDA for TV recommendation appropriately, TV viewers's interested topics is regarded as latent topics in LDA, and asymmetric Dirichlet distribution is applied on the LDA which can reveal the diversity of the TV viewers' interests on topics based on the analysis of the real TV usage history data. The experimental results show that the proposed LDA based TV recommendation method yields average 66.5% with top 5 ranked TV programs in weekly recommendation, average 77.9% precision in bimonthly recommendation with top 5 ranked TV programs for the TV usage history data of similar taste user groups.