• International Journal of Advanced Culture Technology
• /
• v.9 no.4
• /
• pp.268-273
• /
• 2021

In this paper, we propose a sentiment analysis model that improves performance on small-scale data. A sentiment analysis model for small-scale data is proposed and verified through experiments. To this end, we propose Bagging-Bi-GRU, which combines Bi-GRU, which learns GRU, which is a variant of LSTM (Long Short-Term Memory) with excellent performance on sequential data, in both directions and the bagging technique, which is one of the ensembles learning methods. In order to verify the performance of the proposed model, it is applied to small-scale data and large-scale data. And by comparing and analyzing it with the existing machine learning algorithm, Bi-GRU, it shows that the performance of the proposed model is improved not only for small data but also for large data.

• Journal of Information Processing Systems
• /
• v.15 no.6
• /
• pp.1489-1502
• /
• 2019

This work introduces a novel unweighted combination method (UCSS) for business failure perdition (BFP). With considering features of BFP in the age of big data, UCSS integrates the quantitative and qualitative analysis by utilizing soft set theory (SS). We adopt the conventional expert system (ES) as the basic qualitative classifier, the logistic regression model (LR) and the support vector machine (SVM) as basic quantitative classifiers. Unlike other traditional combination methods, we employ soft set theory to integrate the results of each basic classifier without weighting. In this way, UCSS inherits the advantages of ES, LR, SVM, and SS. To verify the performance of UCSS, it is applied to real datasets. We adopt ES, LR, SVM, combination models utilizing the equal weight approach (CMEW), neural network algorithm (CMNN), rough set and D-S evidence theory (CMRD), and the receiver operating characteristic curve (ROC) and SS (CFBSS) as benchmarks. The superior performance of UCSS has been verified by the empirical experiments.

• Journal of the Korean Data and Information Science Society
• /
• v.13 no.2
• /
• pp.185-193
• /
• 2002

The purpose of this paper is to suggest a method for detecting the linear regression change-points or variance change-points in regression model by the combination of Schwarz information criteria. The advantage of the suggested method is to detect change-points more detailed when one compares the suggest method with Chen (1998)'s method.

• Bioinformatics and Biosystems
• /
• v.1 no.2
• /
• pp.128-133
• /
• 2006

This paper compares domain combination based protein-protein interaction prediction method with domain based protein-protein interaction method. The prediction accuracy and reliability of the methods are compared using the same prediction technique and interaction data. According to the comparison, domain combination based prediction method has showed superior prediction accuracy to domain based prediction method for protein pairs with fully overlapped domains with protein pairs in learning sets. When we consider that domain combination based method has the effects of assigning a weight to each domain interaction, it implies that we can improve the prediction accuracies of currently available domain or domain combination based protein interaction prediction methods further by developing more advanced weight assignment techniques. Several significant facts revealed from the comparative studies are also described in this paper.

• Journal of Korea Multimedia Society
• /
• v.19 no.2
• /
• pp.386-395
• /
• 2016

In trajectory-based navigation systems, a huge amount of trajectory data is needed for efficient route explorations. However, it would be very hard to collect trajectories from all the possible start and destination combinations. To provide a practical solution to this problem, we suggest a method combining collected GPS trajectories data into additional generated trajectories with new start and destination combinations without road information. We present a trajectory combination algorithm and its implementation with Scala programming language on Spark platform for big data processing. The experimental results proved that the proposed method can effectively populate the collected trajectories into valid trajectory paths more than three hundred times.

• Proceedings of the KOR-KST Conference
• /
• 1995.02a
• /
• pp.84-98
• /
• 1995

Network-based model were developed to predict short term future traffic volume based on current traffic, historical average, and upstream traffic. It is presumed that upstream traffic volume can be used to predict the downstream traffic in a specific time period. Three models were developed for traffic flow prediction; a combination of historical average and upstream traffic, a combination of current traffic and upstream traffic, and a combination of all three variables. The three models were evaluated using regression analysis. The third model is found to provide the best prediction for the analyzed data. In order to balance the variables appropriately according to the present traffic condition, a heuristic adaptive weighting system is devised based on the relationships between the beginning period of prediction and the previous periods. The developed models were applied to 15-minute freeway data obtained by regular induction loop detectors. The prediction models were shown to be capable of producing reliable and accurate forecasts under congested traffic condition. The prediction systems perform better in the 15-minute range than in the ranges of 30-to 45-minute. It is also found that the combined models usually produce more consistent forecasts than the historical average.

• Journal of Family Resource Management and Policy Review
• /
• v.5 no.1
• /
• pp.15-31
• /
• 2001

The purpose of this study is to identify participation-types of the family in the purchasing decision-making on items of commodities (TV, refrigerator, furniture, passenger car, housing, vacation, saving, husbands’clothing and children’s clothing). The other thing is to find who is major factors in each of the decision-making stages(the recognition of problem, the searching of information, the evaluation of alternatives and purchasing) among the subjects of decision-making (husband, wife, couple combination and family combination). To complete the purpose, the data were collected through questionnaire which 549 wives living in Seoul answered. From the data, frequency, percentage and correspondence analysis were executed by SPSS. The results of this study were briefly summarized as follows; Wives played an important role on all of decision-making stages for refrigerator, furniture, children’s clothing and husbands’clothing. Husbands played an important role on vacation and passenger car. And in case of saving, TV and housing, couple combination played an important role. It means that wives have a major influence on the purchasing process.

• Journal of Korea Multimedia Society
• /
• v.8 no.6
• /
• pp.836-843
• /
• 2005

The Wireless Internet services widely spread out with the developments of CDMA(Code Division Multiple Access) networks and wireless units. In contrast to the telecommunication network, WPAN (Wireless Personal Area Network) enables to transmit data and voice in personal area. Although WPAN technologies are commercially utilized, the combined services with COMA network are not so poplar up to now. Various services can be provided using the combination between COMA and WPAN. This paper presents the practical and united model between COMA and WPAN. Specially, the main focus of this research lies on the design of the Middle-ware system of a handset which could be managing both COMA and WPAN. This system used Bluethooth by WPAN. For the devices with the proposed WPAN Middle-ware, service areas of the COMA network can be expanded to WPAN, various services can be realized by the transmission of data and voice, and consequently, the user computing environment could be improved.

• Journal of Information Technology Services
• /
• v.22 no.3
• /
• pp.29-47
• /
• 2023

With the recent advancement of information and communication technologies such as artificial intelligence, big data, cloud computing, and 5G, data is being produced and digitized in unprecedented amounts. As a result, data has emerged as a critical resource for the future economy, and overseas countries have been revising laws for data protection and utilization. In Korea, the 'Data 3 Act' was revised in 2020 to introduce institutional measures that classify personal information, pseudonymized information, and anonymous information for research, statistics, and preservation of public records. Among them, it is expected to increase the added value of data by combining pseudonymized personal information, and to this end, "the Expert Data Combination Agency" and "the Expert Data Agency" (hereinafter referred to as the Expert Data Processing Agency) system were introduced. In comparison to these domestic systems, we would like to analyze similar overseas systems, and it was recently confirmed that the Vermont government in the United States enacted the first "Data Broker Act" in the United States as a measure to protect personal information held by data brokers. In this study, we aim to compare and analyze the roles and functions of the "Expert Data Processing Agency" and "Data Broker," and to identify differences in designated standards, security measures, etc., in order to present ways to contribute to the activation of the data economy and enhance information protection.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.10 no.2
• /
• pp.37-50
• /
• 1986

The purpose of this study is to compare and to analyse the difference of middle school girls clothing behaviors and their preference for clothing styles according to their intelligence, socio-economic status, and physical constitution. For the study, data were collected from 378 middle school girls resided in Taegu: one half from high intelligence group over I.Q. 113, others from low intelligence group under I.Q. 87 using the questionaire method. For the measurement of the relationship clothing behavior, socio-economic status. Rohrer index, preference style of clothing were examined. The analysis of the data was managed by computer; frequency, percentage, mean, standard deviation, t-test, and ANOVA. The results of the study are as follows; 1. The significant difference in clothing behavior according to intelligence was verified in four: modesty, comfort, management, and psychological dependence. 2. The significant difference in clothing behavior according to socio-economic status was verified in all of eight clothing behavior variables. 3. There was no significant difference according to physical constitution in all clothing behavior variables. 4. There was no significant difference in the preference styles of clothing according to intelligence, but high intelligence group took more interest in detailed factors; design, style, total combination, color, print, and comfort. 5. The upper and middle class preferred slacks and lower class preferred skirts. The upper class took interest in design-style, total combination, and comfort, the middle class in total combination. and comfort, and the lower class in total combination. 6. Thin group preferred skirt, the average group preferred slacks, fatty group preferred slacks and shirts-blouse. Thin group took interest in design-style, and color-print, the average group in total combination, and fatty group in total combination, color-print, and design-style.