• Journal of Intelligence and Information Systems
• /
• v.27 no.4
• /
• pp.1-22
• /
• 2021

Aspect Based Sentiment Analysis (ABSA), which analyzes sentiment based on aspects that appear in the text, is drawing attention because it can be used in various business industries. ABSA is a study that analyzes sentiment by aspects for multiple aspects that a text has. It is being studied in various forms depending on the purpose, such as analyzing all targets or just aspects and sentiments. Here, the aspect refers to the property of a target, and the target refers to the text that causes the sentiment. For example, for restaurant reviews, you could set the aspect into food taste, food price, quality of service, mood of the restaurant, etc. Also, if there is a review that says, "The pasta was delicious, but the salad was not," the words "steak" and "salad," which are directly mentioned in the sentence, become the "target." So far, in ABSA, most studies have analyzed sentiment only based on aspects or targets. However, even with the same aspects or targets, sentiment analysis may be inaccurate. Instances would be when aspects or sentiment are divided or when sentiment exists without a target. For example, sentences like, "Pizza and the salad were good, but the steak was disappointing." Although the aspect of this sentence is limited to "food," conflicting sentiments coexist. In addition, in the case of sentences such as "Shrimp was delicious, but the price was extravagant," although the target here is "shrimp," there are opposite sentiments coexisting that are dependent on the aspect. Finally, in sentences like "The food arrived too late and is cold now." there is no target (NULL), but it transmits a negative sentiment toward the aspect "service." Like this, failure to consider both aspects and targets - when sentiment or aspect is divided or when sentiment exists without a target - creates a dual dependency problem. To address this problem, this research analyzes sentiment by considering both aspects and targets (Target-Aspect-Sentiment Detection, hereby TASD). This study detected the limitations of existing research in the field of TASD: local contexts are not fully captured, and the number of epochs and batch size dramatically lowers the F1-score. The current model excels in spotting overall context and relations between each word. However, it struggles with phrases in the local context and is relatively slow when learning. Therefore, this study tries to improve the model's performance. To achieve the objective of this research, we additionally used auxiliary loss in aspect-sentiment classification by constructing CNN(Convolutional Neural Network) layers parallel to existing models. If existing models have analyzed aspect-sentiment through BERT encoding, Pooler, and Linear layers, this research added CNN layer-adaptive average pooling to existing models, and learning was progressed by adding additional loss values for aspect-sentiment to existing loss. In other words, when learning, the auxiliary loss, computed through CNN layers, allowed the local context to be captured more fitted. After learning, the model is designed to do aspect-sentiment analysis through the existing method. To evaluate the performance of this model, two datasets, SemEval-2015 task 12 and SemEval-2016 task 5, were used and the f1-score increased compared to the existing models. When the batch was 8 and epoch was 5, the difference was largest between the F1-score of existing models and this study with 29 and 45, respectively. Even when batch and epoch were adjusted, the F1-scores were higher than the existing models. It can be said that even when the batch and epoch numbers were small, they can be learned effectively compared to the existing models. Therefore, it can be useful in situations where resources are limited. Through this study, aspect-based sentiments can be more accurately analyzed. Through various uses in business, such as development or establishing marketing strategies, both consumers and sellers will be able to make efficient decisions. In addition, it is believed that the model can be fully learned and utilized by small businesses, those that do not have much data, given that they use a pre-training model and recorded a relatively high F1-score even with limited resources.

• Journal of Intelligence and Information Systems
• /
• v.28 no.1
• /
• pp.107-129
• /
• 2022

It is important to preempt new technology because the technology competition is getting much tougher. Stakeholders conduct exploration activities continuously for new technology preoccupancy at the right time. Gartner's Hype Cycle has significant implications for stakeholders. The Hype Cycle is a expectation graph for new technologies which is combining the technology life cycle (S-curve) with the Hype Level. Stakeholders such as R&D investor, CTO(Chef of Technology Officer) and technical personnel are very interested in Gartner's Hype Cycle for new technologies. Because high expectation for new technologies can bring opportunities to maintain investment by securing the legitimacy of R&D investment. However, contrary to the high interest of the industry, the preceding researches faced with limitations aspect of empirical method and source data(news, academic papers, search traffic, patent etc.). In this study, we focused on two research questions. The first research question was 'Is there a difference in the characteristics of the network structure at each stage of the hype cycle?'. To confirm the first research question, the structural characteristics of each stage were confirmed through the component cohesion size. The second research question is 'Is there a pattern of diffusion at each stage of the hype cycle?'. This research question was to be solved through centralization index and network density. The centralization index is a concept of variance, and a higher centralization index means that a small number of nodes are centered in the network. Concentration of a small number of nodes means a star network structure. In the network structure, the star network structure is a centralized structure and shows better diffusion performance than a decentralized network (circle structure). Because the nodes which are the center of information transfer can judge useful information and deliver it to other nodes the fastest. So we confirmed the out-degree centralization index and in-degree centralization index for each stage. For this purpose, we confirmed the structural features of the community and the expectation diffusion patterns using Social Network Serice(SNS) data in 'Gartner Hype Cycle for Artificial Intelligence, 2021'. Twitter data for 30 technologies (excluding four technologies) listed in 'Gartner Hype Cycle for Artificial Intelligence, 2021' were analyzed. Analysis was performed using R program (4.1.1 ver) and Cyram Netminer. From October 31, 2021 to November 9, 2021, 6,766 tweets were searched through the Twitter API, and converting the relationship user's tweet(Source) and user's retweets (Target). As a result, 4,124 edgelists were analyzed. As a reult of the study, we confirmed the structural features and diffusion patterns through analyze the component cohesion size and degree centralization and density. Through this study, we confirmed that the groups of each stage increased number of components as time passed and the density decreased. Also 'Innovation Trigger' which is a group interested in new technologies as a early adopter in the innovation diffusion theory had high out-degree centralization index and the others had higher in-degree centralization index than out-degree. It can be inferred that 'Innovation Trigger' group has the biggest influence, and the diffusion will gradually slow down from the subsequent groups. In this study, network analysis was conducted using social network service data unlike methods of the precedent researches. This is significant in that it provided an idea to expand the method of analysis when analyzing Gartner's hype cycle in the future. In addition, the fact that the innovation diffusion theory was applied to the Gartner's hype cycle's stage in artificial intelligence can be evaluated positively because the Gartner hype cycle has been repeatedly discussed as a theoretical weakness. Also it is expected that this study will provide a new perspective on decision-making on technology investment to stakeholdes.

• Journal of Intelligence and Information Systems
• /
• v.18 no.2
• /
• pp.29-45
• /
• 2012

Bond rating is regarded as an important event for measuring financial risk of companies and for determining the investment returns of investors. As a result, it has been a popular research topic for researchers to predict companies' credit ratings by applying statistical and machine learning techniques. The statistical techniques, including multiple regression, multiple discriminant analysis (MDA), logistic models (LOGIT), and probit analysis, have been traditionally used in bond rating. However, one major drawback is that it should be based on strict assumptions. Such strict assumptions include linearity, normality, independence among predictor variables and pre-existing functional forms relating the criterion variablesand the predictor variables. Those strict assumptions of traditional statistics have limited their application to the real world. Machine learning techniques also used in bond rating prediction models include decision trees (DT), neural networks (NN), and Support Vector Machine (SVM). Especially, SVM is recognized as a new and promising classification and regression analysis method. SVM learns a separating hyperplane that can maximize the margin between two categories. SVM is simple enough to be analyzed mathematical, and leads to high performance in practical applications. SVM implements the structuralrisk minimization principle and searches to minimize an upper bound of the generalization error. In addition, the solution of SVM may be a global optimum and thus, overfitting is unlikely to occur with SVM. In addition, SVM does not require too many data sample for training since it builds prediction models by only using some representative sample near the boundaries called support vectors. A number of experimental researches have indicated that SVM has been successfully applied in a variety of pattern recognition fields. However, there are three major drawbacks that can be potential causes for degrading SVM's performance. First, SVM is originally proposed for solving binary-class classification problems. Methods for combining SVMs for multi-class classification such as One-Against-One, One-Against-All have been proposed, but they do not improve the performance in multi-class classification problem as much as SVM for binary-class classification. Second, approximation algorithms (e.g. decomposition methods, sequential minimal optimization algorithm) could be used for effective multi-class computation to reduce computation time, but it could deteriorate classification performance. Third, the difficulty in multi-class prediction problems is in data imbalance problem that can occur when the number of instances in one class greatly outnumbers the number of instances in the other class. Such data sets often cause a default classifier to be built due to skewed boundary and thus the reduction in the classification accuracy of such a classifier. SVM ensemble learning is one of machine learning methods to cope with the above drawbacks. Ensemble learning is a method for improving the performance of classification and prediction algorithms. AdaBoost is one of the widely used ensemble learning techniques. It constructs a composite classifier by sequentially training classifiers while increasing weight on the misclassified observations through iterations. The observations that are incorrectly predicted by previous classifiers are chosen more often than examples that are correctly predicted. Thus Boosting attempts to produce new classifiers that are better able to predict examples for which the current ensemble's performance is poor. In this way, it can reinforce the training of the misclassified observations of the minority class. This paper proposes a multiclass Geometric Mean-based Boosting (MGM-Boost) to resolve multiclass prediction problem. Since MGM-Boost introduces the notion of geometric mean into AdaBoost, it can perform learning process considering the geometric mean-based accuracy and errors of multiclass. This study applies MGM-Boost to the real-world bond rating case for Korean companies to examine the feasibility of MGM-Boost. 10-fold cross validations for threetimes with different random seeds are performed in order to ensure that the comparison among three different classifiers does not happen by chance. For each of 10-fold cross validation, the entire data set is first partitioned into tenequal-sized sets, and then each set is in turn used as the test set while the classifier trains on the other nine sets. That is, cross-validated folds have been tested independently of each algorithm. Through these steps, we have obtained the results for classifiers on each of the 30 experiments. In the comparison of arithmetic mean-based prediction accuracy between individual classifiers, MGM-Boost (52.95%) shows higher prediction accuracy than both AdaBoost (51.69%) and SVM (49.47%). MGM-Boost (28.12%) also shows the higher prediction accuracy than AdaBoost (24.65%) and SVM (15.42%)in terms of geometric mean-based prediction accuracy. T-test is used to examine whether the performance of each classifiers for 30 folds is significantly different. The results indicate that performance of MGM-Boost is significantly different from AdaBoost and SVM classifiers at 1% level. These results mean that MGM-Boost can provide robust and stable solutions to multi-classproblems such as bond rating.

• Journal of Intelligence and Information Systems
• /
• v.22 no.1
• /
• pp.187-204
• /
• 2016

Document classification based on emotional polarity has become a welcomed emerging task owing to the great explosion of data on the Web. In the big data age, there are too many information sources to refer to when making decisions. For example, when considering travel to a city, a person may search reviews from a search engine such as Google or social networking services (SNSs) such as blogs, Twitter, and Facebook. The emotional polarity of positive and negative reviews helps a user decide on whether or not to make a trip. Sentiment analysis of customer reviews has become an important research topic as datamining technology is widely accepted for text mining of the Web. Sentiment analysis has been used to classify documents through machine learning techniques, such as the decision tree, neural networks, and support vector machines (SVMs). is used to determine the attitude, position, and sensibility of people who write articles about various topics that are published on the Web. Regardless of the polarity of customer reviews, emotional reviews are very helpful materials for analyzing the opinions of customers through their reviews. Sentiment analysis helps with understanding what customers really want instantly through the help of automated text mining techniques. Sensitivity analysis utilizes text mining techniques on text on the Web to extract subjective information in the text for text analysis. Sensitivity analysis is utilized to determine the attitudes or positions of the person who wrote the article and presented their opinion about a particular topic. In this study, we developed a model that selects a hot topic from user posts at China's online stock forum by using the k-means algorithm and self-organizing map (SOM). In addition, we developed a detecting model to predict a hot topic by using machine learning techniques such as logit, the decision tree, and SVM. We employed sensitivity analysis to develop our model for the selection and detection of hot topics from China's online stock forum. The sensitivity analysis calculates a sentimental value from a document based on contrast and classification according to the polarity sentimental dictionary (positive or negative). The online stock forum was an attractive site because of its information about stock investment. Users post numerous texts about stock movement by analyzing the market according to government policy announcements, market reports, reports from research institutes on the economy, and even rumors. We divided the online forum's topics into 21 categories to utilize sentiment analysis. One hundred forty-four topics were selected among 21 categories at online forums about stock. The posts were crawled to build a positive and negative text database. We ultimately obtained 21,141 posts on 88 topics by preprocessing the text from March 2013 to February 2015. The interest index was defined to select the hot topics, and the k-means algorithm and SOM presented equivalent results with this data. We developed a decision tree model to detect hot topics with three algorithms: CHAID, CART, and C4.5. The results of CHAID were subpar compared to the others. We also employed SVM to detect the hot topics from negative data. The SVM models were trained with the radial basis function (RBF) kernel function by a grid search to detect the hot topics. The detection of hot topics by using sentiment analysis provides the latest trends and hot topics in the stock forum for investors so that they no longer need to search the vast amounts of information on the Web. Our proposed model is also helpful to rapidly determine customers' signals or attitudes towards government policy and firms' products and services.

• Applied Biological Chemistry
• /
• v.48 no.1
• /
• pp.23-33
• /
• 2005

Agricultural sciences field in South Korea has many strong points such as numerous researchers, establishment of research infra-structure, excellence in research competitiveness and high technological level. However, there are also many weaknesses including insufficient leadership at related societies and institutes, deficiency of the next generation research group, and insufficiency in research productivity. There are many opportunities including increasing the importance of the biotechnological industry, activating international cooperation researches, and exploring the multitude of possible research areas to be studied. However, some threats still exist, such as pressure from the government of developed countries to open the agricultural market, the decrease of specialized farms, and intensification for researches to gratify economic and social demands. To encourage research activities in the agricultural sciences field in Korea, the following actions and systems are required: 1) formulation of a mid- and a long-term research master plan, 2) development of a database on the man power in related fields, 3) activation of top-down research topics, and associated increase of individual research grants, 4) development of special national programs for basic researches in agricultural sciences, 5) organization of a committee for policy and planning within the related societies, and 6) system development for the fair evaluation of the research results.

• 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 the Korean Association of Geographic Information Studies
• /
• v.22 no.4
• /
• pp.116-130
• /
• 2019

In South Korea, forest fire occurrences are increasing in size and duration due to various factors such as the increase in fuel materials and frequent drying conditions in forests. Therefore, it is necessary to minimize the damage caused by forest fires by appropriately providing the probability of forest fire risk. The purpose of this study is to improve the Daily Weather Index(DWI) provided by the current forest fire forecasting system in South Korea. A new Fire Risk Index(FRI) is proposed in this study, which is provided in a 5km grid through the synergistic use of numerical weather forecast data, satellite-based drought indices, and forest fire-prone areas. The FRI is calculated based on the product of the Fine Fuel Moisture Code(FFMC) optimized for Korea, an integrated drought index, and spatio-temporal weighting approaches. In order to improve the temporal accuracy of forest fire risk, monthly weights were applied based on the forest fire occurrences by month. Similarly, spatial weights were applied using the forest fire density information to improve the spatial accuracy of forest fire risk. In the time series analysis of the number of monthly forest fires and the FRI, the relationship between the two were well simulated. In addition, it was possible to provide more spatially detailed information on forest fire risk when using FRI in the 5km grid than DWI based on administrative units. The research findings from this study can help make appropriate decisions before and after forest fire occurrences.

• Journal of agricultural medicine and community health
• /
• v.24 no.1
• /
• pp.49-63
• /
• 1999

The objective of the study was to examine and compare health behavior between rural area and urban area in Soonchun city. Data were collected through personal interviews from 25, April to 30, May in 1998. Questions were asked to the rural area residents(n=399) and urban area residents(n=149) about their health behaviors, including such as self-recognition of health status, health related behaviors(smoking, drinking, eating habit, and exercising), status of disease and prevention, and utilization of hospital. As we examine the demographic characteristics, rural area residents were more aged(p<0.001) than urban area residents. And the urban residents had higher education(p<0.01), higher income(p<0.01) and higher health care cost(p<0.01) than rural residents. There were difference in health status existed between rural and urban residents. Rural residents had poorer health status(p<0.01) than urban residents, and however urban residents had more anxiety about their health(p<0.01) than rural residents. Comparison of the health related behavior between rural and urban area residents, rural residents were more likely to smoke(p<0.05), less intake of milk(p<0.01), do not exercise(p<0.01), and less try to lose their weight(p<0.01) than urban residents. Rural resident used to suffer from chronic diseases than urban residents(p<0.01). Consideration of health care need for rural residents are required due to the results shown as above. Therefore, the health care center, where most of the rural residents depend on for their treatment and prevention of disease, should make inquiries about resident's health care need and evaluate the important information sources for construction of a health care information system.

• The KIPS Transactions:PartB
• /
• v.19B no.2
• /
• pp.123-126
• /
• 2012

This paper proposed a method to separate a liver into left and right liver lobes for simple and exact volumetry of the river graft at abdominal MDCT(Multi-Detector Computed Tomography) image before the living donor liver transplantation. A medical team can evaluate an accurate river graft with minimized interaction between the team and a system using this algorithm for ensuring donor's and recipient's safe. On the image of segmented liver, 2 points(PMHV: a point in Middle Hepatic Vein and PPV: a point at the beginning of right branch of Portal Vein) are selected to separate a liver into left and right liver lobes. Middle hepatic vein is automatically segmented using PMHV, and the cutting line is decided on the basis of segmented Middle Hepatic Vein. A liver is separated on connecting the cutting line and PPV. The volume and ratio of the river graft are estimated. The volume estimated using 2 points are compared with a manual volume that diagnostic radiologist processed and estimated and the weight measured during surgery to support proof of exact volume. The mean ${\pm}$ standard deviation of the differences between the actual weights and the estimated volumes was $162.38cm^3{\pm}124.39$ in the case of manual segmentation and $107.69cm^3{\pm}97.24$ in the case of 2 points method. The correlation coefficient between the actual weight and the manually estimated volume is 0.79, and the correlation coefficient between the actual weight and the volume estimated using 2 points is 0.87. After selection the 2 points, the time involved in separation a liver into left and right river lobe and volumetry of them is measured for confirmation that the algorithm can be used on real time during surgery. The mean ${\pm}$ standard deviation of the process time is $57.28sec{\pm}32.81$ per 1 data set ($149.17pages{\pm}55.92$).

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.42 no.5 s.305
• /
• pp.29-40
• /
• 2005

According as base of digital hologram has been magnified, discussion of compression technology is expected as a international standard which defines the compression technique of 3D image and video has been progressed in form of 3DAV which is a part of MPEG. As we can identify in case of 3DAV, the coding technique has high possibility to be formed into the hybrid type which is a merged, refined, or mixid with the various previous technique. Therefore, we wish to present the relationship between various image/video coding techniques and digital hologram In this paper, we propose an efficient coding method of digital hologram using standard compression tools for video and image. At first, we convert fringe patterns into video data using a principle of CGH(Computer Generated Hologram), and then encode it. In this research, we propose a compression algorithm is made up of various method such as pre-processing for transform, local segmentation with global information of object image, frequency transform for coding, scanning to make fringe to video stream, classification of coefficients, and hybrid video coding. Finally the proposed hybrid compression algorithm is all of these methods. The tool for still image coding is JPEG2000, and the toots for video coding include various international compression algorithm such as MPEG-2, MPEG-4, and H.264 and various lossless compression algorithm. The proposed algorithm illustrated that it have better properties for reconstruction than the previous researches on far greater compression rate above from four times to eight times as much. Therefore we expect that the proposed technique for digital hologram coding is to be a good preceding research.