• Journal of the Korean Chemical Society
• /
• v.58 no.6
• /
• pp.600-611
• /
• 2014

In this study, we sampled science writing tasks and investigated their frequency of use shown in the chemistry domain from two sets of 18 middle school science textbooks developed under the 2007 Revised National Curriculum(RNC) and the 2009 RNC, respectively. In addition, we categorized the sampled science writing tasks depending on the cognitive process and type of writing and compared with the results obtained from analysis of global issues presented in the science writings. From the textbooks developed under the 2007 RNC, a total of 183 science writing tasks were identified in which 10.17 tasks per textbook and 1.32 tasks per 10 pages were used averagely. A total of 168 were identified from the textbooks for the 2009 RNC. Among them, 9.33 tasks per textbook and 1.23 tasks per 10 pages were used on average. Comparing with these results, the average frequency of use of the tasks per textbook and per ten pages were decreased, respectively. Moreover, the number of science writing tasks were found in each curriculum varied considerably depending on the units and the publishers, and that the writing tasks were mainly arranged in the finale, wrapping up stage. In the analysis of science writing tasks according to the cognitive process, the highest and lowest frequency of use were observed in the category of 'understand' and 'remember', respectively. According to the classification of science writing tasks based on the types of writing, the writings for the information delivery were most used and the highest frequency of use was observed in the category of 'understand' of the cognitive process belonging to 'information delivery'. As for the results of the analysis of global issues, the number of science writing tasks including global issues increased from 21(11.48%) in the 2007 RNC to 33(19.64%) in the 2009 RNC. Furthermore, science writing tasks associated with protection of environment showed the highest frequency of use in the both curriculums, and it was analyzed that the materials of global issues used in the 2009 RNC were much more diverse.

• The KIPS Transactions:PartB
• /
• v.10B no.3
• /
• pp.281-286
• /
• 2003

A large number of features are collected for problem solving in real life, but to utilize ail the features collected would be difficult. It is not so easy to collect of correct data about all features. In case it takes advantage of all collected data to learn, complicated learning model is created and good performance result can't get. Also exist interrelationships or hierarchical relations among the features. We can reduce feature's number analyzing relation among the features using heuristic knowledge or statistical method. Heuristic technique refers to learning through repetitive trial and errors and experience. Experts can approach to relevant problem domain through opinion collection process by experience. These properties can be utilized to reduce the number of feature used in learning. Experts generate a new feature (highly abstract) using raw data. This paper describes machine learning model that reduce the number of features used in learning using heuristic function and use abstracted feature by neural network's input value. We have applied this model to the win/lose prediction in pro-baseball games. The result shows the model mixing two techniques not only reduces the complexity of the neural network model but also significantly improves the classification accuracy than when neural network and heuristic model are used separately.

• Journal of Intelligence and Information Systems
• /
• v.21 no.3
• /
• pp.79-99
• /
• 2015

The prediction of bankruptcy has been extensively studied in the accounting and finance field. It can have an important impact on lending decisions and the profitability of financial institutions in terms of risk management. Many researchers have focused on constructing a more robust bankruptcy prediction model. Early studies primarily used statistical techniques such as multiple discriminant analysis (MDA) and logit analysis for bankruptcy prediction. However, many studies have demonstrated that artificial intelligence (AI) approaches, such as artificial neural networks (ANN), decision trees, case-based reasoning (CBR), and support vector machine (SVM), have been outperforming statistical techniques since 1990s for business classification problems because statistical methods have some rigid assumptions in their application. In previous studies on corporate bankruptcy, many researchers have focused on developing a bankruptcy prediction model using financial ratios. However, there are few studies that suggest the specific types of bankruptcy. Previous bankruptcy prediction models have generally been interested in predicting whether or not firms will become bankrupt. Most of the studies on bankruptcy types have focused on reviewing the previous literature or performing a case study. Thus, this study develops a model using data mining techniques for predicting the specific types of bankruptcy as well as the occurrence of bankruptcy in Korean small- and medium-sized construction firms in terms of profitability, stability, and activity index. Thus, firms will be able to prevent it from occurring in advance. We propose a hybrid approach using two artificial neural networks (ANNs) for the prediction of bankruptcy types. The first is a back-propagation neural network (BPN) model using supervised learning for bankruptcy prediction and the second is a self-organizing map (SOM) model using unsupervised learning to classify bankruptcy data into several types. Based on the constructed model, we predict the bankruptcy of companies by applying the BPN model to a validation set that was not utilized in the development of the model. This allows for identifying the specific types of bankruptcy by using bankruptcy data predicted by the BPN model. We calculated the average of selected input variables through statistical test for each cluster to interpret characteristics of the derived clusters in the SOM model. Each cluster represents bankruptcy type classified through data of bankruptcy firms, and input variables indicate financial ratios in interpreting the meaning of each cluster. The experimental result shows that each of five bankruptcy types has different characteristics according to financial ratios. Type 1 (severe bankruptcy) has inferior financial statements except for EBITDA (earnings before interest, taxes, depreciation, and amortization) to sales based on the clustering results. Type 2 (lack of stability) has a low quick ratio, low stockholder's equity to total assets, and high total borrowings to total assets. Type 3 (lack of activity) has a slightly low total asset turnover and fixed asset turnover. Type 4 (lack of profitability) has low retained earnings to total assets and EBITDA to sales which represent the indices of profitability. Type 5 (recoverable bankruptcy) includes firms that have a relatively good financial condition as compared to other bankruptcy types even though they are bankrupt. Based on the findings, researchers and practitioners engaged in the credit evaluation field can obtain more useful information about the types of corporate bankruptcy. In this paper, we utilized the financial ratios of firms to classify bankruptcy types. It is important to select the input variables that correctly predict bankruptcy and meaningfully classify the type of bankruptcy. In a further study, we will include non-financial factors such as size, industry, and age of the firms. Thus, we can obtain realistic clustering results for bankruptcy types by combining qualitative factors and reflecting the domain knowledge of experts.

• Journal of Intelligence and Information Systems
• /
• v.21 no.3
• /
• pp.53-77
• /
• 2015

Korea has witnessed an increasing number of domestic patent applications, but a majority of them are not utilized to their maximum potential but end up becoming obsolete. According to the 2012 National Congress' Inspection of Administration, about 73% of patents possessed by universities and public-funded research institutions failed to lead to creating social values, but remain latent. One of the main problem of this issue is that patent creators such as individual researcher, university, or research institution lack abilities to commercialize their patents into viable businesses with those enterprises that are in need of them. Also, for enterprises side, it is hard to find the appropriate patents by searching keywords on all such occasions. This system proposes a patent recommendation system that can identify and recommend intellectual rights appropriate to users' interested fields among a rapidly accumulating number of patent assets in a more easy and efficient manner. The proposed system extracts core contents and technology sectors from the existing pool of patents, and combines it with secondary social knowledge, which derives from tags information created by users, in order to find the best patents recommended for users. That is to say, in an early stage where there is no accumulated tag information, the recommendation is done by utilizing content characteristics, which are identified through an analysis of key words contained in such parameters as 'Title of Invention' and 'Claim' among the various patent attributes. In order to do this, the suggested system extracts only nouns from patents and assigns a weight to each noun according to the importance of it in all patents by performing TF-IDF analysis. After that, it finds patents which have similar weights with preferred patents by a user. In this paper, this similarity is called a "Domain Similarity". Next, the suggested system extract technology sector's characteristics from patent document by analyzing the international technology classification code (International Patent Classification, IPC). Every patents have more than one IPC, and each user can attach more than one tag to the patents they like. Thus, each user has a set of IPC codes included in tagged patents. The suggested system manages this IPC set to analyze technology preference of each user and find the well-fitted patents for them. In order to do this, the suggeted system calcuates a 'Technology_Similarity' between a set of IPC codes and IPC codes contained in all other patents. After that, when the tag information of multiple users are accumulated, the system expands the recommendations in consideration of other users' social tag information relating to the patent that is tagged by a concerned user. The similarity between tag information of perferred 'patents by user and other patents are called a 'Social Simialrity' in this paper. Lastly, a 'Total Similarity' are calculated by adding these three differenent similarites and patents having the highest 'Total Similarity' are recommended to each user. The suggested system are applied to a total of 1,638 korean patents obtained from the Korea Industrial Property Rights Information Service (KIPRIS) run by the Korea Intellectual Property Office. However, since this original dataset does not include tag information, we create virtual tag information and utilized this to construct the semi-virtual dataset. The proposed recommendation algorithm was implemented with JAVA, a computer programming language, and a prototype graphic user interface was also designed for this study. As the proposed system did not have dependent variables and uses virtual data, it is impossible to verify the recommendation system with a statistical method. Therefore, the study uses a scenario test method to verify the operational feasibility and recommendation effectiveness of the system. The results of this study are expected to improve the possibility of matching promising patents with the best suitable businesses. It is assumed that users' experiential knowledge can be accumulated, managed, and utilized in the As-Is patent system, which currently only manages standardized patent information.

• 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.24 no.1
• /
• pp.205-225
• /
• 2018

Convolutional Neural Network (ConvNet) is one class of the powerful Deep Neural Network that can analyze and learn hierarchies of visual features. Originally, first neural network (Neocognitron) was introduced in the 80s. At that time, the neural network was not broadly used in both industry and academic field by cause of large-scale dataset shortage and low computational power. However, after a few decades later in 2012, Krizhevsky made a breakthrough on ILSVRC-12 visual recognition competition using Convolutional Neural Network. That breakthrough revived people interest in the neural network. The success of Convolutional Neural Network is achieved with two main factors. First of them is the emergence of advanced hardware (GPUs) for sufficient parallel computation. Second is the availability of large-scale datasets such as ImageNet (ILSVRC) dataset for training. Unfortunately, many new domains are bottlenecked by these factors. For most domains, it is difficult and requires lots of effort to gather large-scale dataset to train a ConvNet. Moreover, even if we have a large-scale dataset, training ConvNet from scratch is required expensive resource and time-consuming. These two obstacles can be solved by using transfer learning. Transfer learning is a method for transferring the knowledge from a source domain to new domain. There are two major Transfer learning cases. First one is ConvNet as fixed feature extractor, and the second one is Fine-tune the ConvNet on a new dataset. In the first case, using pre-trained ConvNet (such as on ImageNet) to compute feed-forward activations of the image into the ConvNet and extract activation features from specific layers. In the second case, replacing and retraining the ConvNet classifier on the new dataset, then fine-tune the weights of the pre-trained network with the backpropagation. In this paper, we focus on using multiple ConvNet layers as a fixed feature extractor only. However, applying features with high dimensional complexity that is directly extracted from multiple ConvNet layers is still a challenging problem. We observe that features extracted from multiple ConvNet layers address the different characteristics of the image which means better representation could be obtained by finding the optimal combination of multiple ConvNet layers. Based on that observation, we propose to employ multiple ConvNet layer representations for transfer learning instead of a single ConvNet layer representation. Overall, our primary pipeline has three steps. Firstly, images from target task are given as input to ConvNet, then that image will be feed-forwarded into pre-trained AlexNet, and the activation features from three fully connected convolutional layers are extracted. Secondly, activation features of three ConvNet layers are concatenated to obtain multiple ConvNet layers representation because it will gain more information about an image. When three fully connected layer features concatenated, the occurring image representation would have 9192 (4096+4096+1000) dimension features. However, features extracted from multiple ConvNet layers are redundant and noisy since they are extracted from the same ConvNet. Thus, a third step, we will use Principal Component Analysis (PCA) to select salient features before the training phase. When salient features are obtained, the classifier can classify image more accurately, and the performance of transfer learning can be improved. To evaluate proposed method, experiments are conducted in three standard datasets (Caltech-256, VOC07, and SUN397) to compare multiple ConvNet layer representations against single ConvNet layer representation by using PCA for feature selection and dimension reduction. Our experiments demonstrated the importance of feature selection for multiple ConvNet layer representation. Moreover, our proposed approach achieved 75.6% accuracy compared to 73.9% accuracy achieved by FC7 layer on the Caltech-256 dataset, 73.1% accuracy compared to 69.2% accuracy achieved by FC8 layer on the VOC07 dataset, 52.2% accuracy compared to 48.7% accuracy achieved by FC7 layer on the SUN397 dataset. We also showed that our proposed approach achieved superior performance, 2.8%, 2.1% and 3.1% accuracy improvement on Caltech-256, VOC07, and SUN397 dataset respectively compare to existing work.

• Asia pacific journal of information systems
• /
• v.21 no.1
• /
• pp.103-122
• /
• 2011

Recently, numerous documents have been made available electronically. Internet search engines and digital libraries commonly return query results containing hundreds or even thousands of documents. In this situation, it is virtually impossible for users to examine complete documents to determine whether they might be useful for them. For this reason, some on-line documents are accompanied by a list of keywords specified by the authors in an effort to guide the users by facilitating the filtering process. In this way, a set of keywords is often considered a condensed version of the whole document and therefore plays an important role for document retrieval, Web page retrieval, document clustering, summarization, text mining, and so on. Since many academic journals ask the authors to provide a list of five or six keywords on the first page of an article, keywords are most familiar in the context of journal articles. However, many other types of documents could not benefit from the use of keywords, including Web pages, email messages, news reports, magazine articles, and business papers. Although the potential benefit is large, the implementation itself is the obstacle; manually assigning keywords to all documents is a daunting task, or even impractical in that it is extremely tedious and time-consuming requiring a certain level of domain knowledge. Therefore, it is highly desirable to automate the keyword generation process. There are mainly two approaches to achieving this aim: keyword assignment approach and keyword extraction approach. Both approaches use machine learning methods and require, for training purposes, a set of documents with keywords already attached. In the former approach, there is a given set of vocabulary, and the aim is to match them to the texts. In other words, the keywords assignment approach seeks to select the words from a controlled vocabulary that best describes a document. Although this approach is domain dependent and is not easy to transfer and expand, it can generate implicit keywords that do not appear in a document. On the other hand, in the latter approach, the aim is to extract keywords with respect to their relevance in the text without prior vocabulary. In this approach, automatic keyword generation is treated as a classification task, and keywords are commonly extracted based on supervised learning techniques. Thus, keyword extraction algorithms classify candidate keywords in a document into positive or negative examples. Several systems such as Extractor and Kea were developed using keyword extraction approach. Most indicative words in a document are selected as keywords for that document and as a result, keywords extraction is limited to terms that appear in the document. Therefore, keywords extraction cannot generate implicit keywords that are not included in a document. According to the experiment results of Turney, about 64% to 90% of keywords assigned by the authors can be found in the full text of an article. Inversely, it also means that 10% to 36% of the keywords assigned by the authors do not appear in the article, which cannot be generated through keyword extraction algorithms. Our preliminary experiment result also shows that 37% of keywords assigned by the authors are not included in the full text. This is the reason why we have decided to adopt the keyword assignment approach. In this paper, we propose a new approach for automatic keyword assignment namely IVSM(Inverse Vector Space Model). The model is based on a vector space model. which is a conventional information retrieval model that represents documents and queries by vectors in a multidimensional space. IVSM generates an appropriate keyword set for a specific document by measuring the distance between the document and the keyword sets. The keyword assignment process of IVSM is as follows: (1) calculating the vector length of each keyword set based on each keyword weight; (2) preprocessing and parsing a target document that does not have keywords; (3) calculating the vector length of the target document based on the term frequency; (4) measuring the cosine similarity between each keyword set and the target document; and (5) generating keywords that have high similarity scores. Two keyword generation systems were implemented applying IVSM: IVSM system for Web-based community service and stand-alone IVSM system. Firstly, the IVSM system is implemented in a community service for sharing knowledge and opinions on current trends such as fashion, movies, social problems, and health information. The stand-alone IVSM system is dedicated to generating keywords for academic papers, and, indeed, it has been tested through a number of academic papers including those published by the Korean Association of Shipping and Logistics, the Korea Research Academy of Distribution Information, the Korea Logistics Society, the Korea Logistics Research Association, and the Korea Port Economic Association. We measured the performance of IVSM by the number of matches between the IVSM-generated keywords and the author-assigned keywords. According to our experiment, the precisions of IVSM applied to Web-based community service and academic journals were 0.75 and 0.71, respectively. The performance of both systems is much better than that of baseline systems that generate keywords based on simple probability. Also, IVSM shows comparable performance to Extractor that is a representative system of keyword extraction approach developed by Turney. As electronic documents increase, we expect that IVSM proposed in this paper can be applied to many electronic documents in Web-based community and digital library.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.10 no.3
• /
• pp.80-92
• /
• 2007

Forest fire is the dominant large-scale disturbance mechanism in the Korean temperate forest, and it strongly influences forest structure and function. Moreover burn severity incorporates both short- and long-term post-fire effects on the local and regional environment. Burn severity is defined by the degree to which an ecosystem has changed owing to the fire. Vegetation rehabilitation may specifically vary according to burn severity after fire. To understand burn severity and process of vegetation rehabilitation at the damaged area after large-fire is required a lot of man powers and budgets. However the analysis of burn severity in the forest area using satellite imagery can acquire rapidly information and more objective results remotely in the large-fire area. Space and airbone sensors have been used to map area burned, assess characteristics of active fires, and characterize post-fire ecological effects. For classifying fire damaged area and analyzing burn severity of Samcheok fire area occurred in 2000, Cheongyang fire in 2002, and Yangyang fire in 2005 we utilized Normalized Burn Ratio(NBR) technique. The NBR is temporally differenced between pre- and post-fire datasets to determine the extent and degree of change detected from burning. In this paper we use pre- and post-fire imagery from the Landsat TM and ETM+ imagery to compute the NBR and evaluate large-scale patterns of burn severity at 30m spatial resolution. 65% in the Samcheok fire area, 91% in the Cheongyang fire area and 65% in the Yangyang fire area were corresponded to burn severity class above 'High'. Therefore the use of a remotely sensed Differenced Normalized Burn Ratio(${\Delta}NBR$) by RS and GIS allows for the burn severity to be quantified spatially by mapping damaged domain and burn severity across large-fire area.

• Industry Promotion Research
• /
• v.7 no.2
• /
• pp.43-49
• /
• 2022

This study aims to study Esping-Anderson's theory of welfare state system, develop a model of welfare state suitable for Korea's situation, and apply it to reality. In this research method, basic research and analysis of ideology is used, focusing on Esping-Anderson's welfare state system theory, and applying it appropriately to the Korean situation. Studies on the model of the welfare state have been studied after the classification of complementary and institutional models asserted by Willensky and Lebo in 1965. In addition, Esping-Andersen asserts three things as a model of the welfare state according to ideology. First, the role of the market is central to the liberal welfare system that best fits the image of classical capitalism, and individualistic solidarity through the market. The role of the state or family, which can be a hindrance, is actually marginalized. In addition, in order to maximize individualistic solidarity through the market, de-commodification in the national domain tends to be minimized. Second, the conservative welfare system has a strong familistic element, so the source of social solidarity is the family, and the state plays a role of supporting and supplementing the characteristics of this family. In the conservative system, de-commodification appears to be high among household heads, or the welfare system takes on a corporatist and nationalistic form, it can be said that these characteristics are reflected. Third, in the social democratic welfare system, the source of social solidarity is the state. Therefore, the role of the state is large, the state has a high possibility of decommodification, and it has the characteristics of substitutes for the family and the market through universalist intervention. This study applies Esping-Anderson's three welfare state models to study a model suitable for the Korean situation. In conclusion, Esping-Anderson's three welfare state models can be classified into a market-oriented model based on a liberal welfare system, a status-oriented model based on a conservative corporatist welfare system, and a solidarity-oriented model based on a social-democratic welfare system, presented a compromise between liberalism and conservatism as a Korean model.

• Korean Journal of Remote Sensing
• /
• v.40 no.3
• /
• pp.295-305
• /
• 2024

This study aims to analyze the micro-sedimentary structures of the Hwangdo tidal flats using ultra-high resolution unmanned aerial vehicle (UAV) data. Tidal flats, located in the transitional area between land and sea, constantly change due to tidal activities and provide a unique environment important for understanding sedimentary processes and environmental conditions. Traditional field observation methods are limited in spatial and temporal coverage, and existing satellite imagery does not provide sufficient resolution to study micro-sedimentary structures. To overcome these limitations, high-resolution images of the Hwangdo tidal flats in Chungcheongnam-do were acquired using UAVs. This area has experienced significant changes in its sedimentary environment due to coastal development projects such as sea wall construction. From May 17 to 18, 2022, sediment samples were collected from 91 points during field surveys and 25 in-situ points were intensively analyzed. UAV data with a spatial resolution of approximately 0.9 mm allowed identifying and extracting parameters related to micro-sedimentary structures. For mud cracks, the length of the major axis of the polygons was extracted, and the wavelength and ripple symmetry index were extracted for ripple marks. The results of the study showed that in areas with mud content above 80%, mud cracks formed at an average major axis length of 37.3 cm. In regions with sand content above 60%, ripples with an average wavelength of 8 cm and a ripple symmetry index of 2.0 were formed. This study demonstrated that micro-sedimentary structures of tidal flats can be effectively analyzed using ultra-high resolution UAV data without field surveys. This highlights the potential of UAV technology as an important tool in environmental monitoring and coastal management and shows its usefulness in the study of sedimentary structures. In addition, the results of this study are expected to serve as baseline data for more accurate sedimentary facies classification.