• Journal of Korea Society of Digital Industry and Information Management
• /
• v.15 no.3
• /
• pp.197-206
• /
• 2019

Telemarketing has been used in finance due to the reduction of offline channels. In order to select telemarketing target customers, various machine learning techniques have emerged to maximize the effect of minimum cost. However, there are problems that the class imbalance, which the number of marketing success customers is smaller than the number of failed customers, and the recall rate is lower than accuracy. In this paper, we propose a method that solve the imbalanced class problem and increase the recall rate to improve the efficiency. The hybrid sampling method is applied to balance the data in the class, and the stacked deep network is applied to improve the recall and precision as well as the accuracy. The proposed method is applied to actual bank telemarketing data. As a result of the comparison experiment, the accuracy, the recall, and the precision is improved higher than that of the conventional methods.

• International conference on construction engineering and project management
• /
• 2011.02a
• /
• pp.7-11
• /
• 2011

Quantity take-off (QT) is one of the most important tasks for determining the total cost of a construction project, and it requires accuracy and reliability for the result. Accuracy and reliability in BIM-based QT are also required to assure the completeness of the result. However, there has been no basis to measure the completeness of the BIM-based QT result. As BIM is increasingly being adopted in the construction industry with a wide variety of purposes, it becomes more important to have a method to improve the completeness of BIM-based QT and criteria to measure it. This research focuses on the hypothesis that the completeness of BIM-based QT relies on the accuracy and reliability of BIM and the BIM-based QT process. As a basic research to determine the completeness of BIM-based QT, this research analyzes and derives factors that affect the completeness of BIM-based QT.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.38 no.4
• /
• pp.35-52
• /
• 2013

Markowitz's portfolio selection model is used to construct an optimal portfolio which has minimum variance, while satisfying a minimum required expected return. The model uses estimators based on analysis of historical data to estimate the returns, standard deviations, and correlation coefficients of individual stocks being considered for investment. However, due to the inaccuracies involved in estimations, the true optimality of a portfolio constructed using the model is questionable. To investigate the effect of estimation inaccuracy on actual portfolio performance, we study the changes in a portfolio's realized return and standard deviation as the accuracy of the estimations for each stock's return, standard deviation, and correlation coefficient is increased. Furthermore, we empirically analyze the portfolio's performance by comparing it with the performance of active mutual funds that are being traded in the Korean stock market and the KOSPI benchmark index, in terms of portfolio returns, standard deviations of returns, and Sharpe ratios. Our results suggest that, among the three input parameters, the accuracy of the estimated returns of individual stocks has the largest effect on performance, while the accuracy of the estimates of the standard deviation of each stock's returns and the correlation coefficient between different stocks have smaller effects. In addition, it is shown that even a small increase in the accuracy of the estimated return of individual stocks improves the portfolio's performance substantially, suggesting that Markowitz's model can be more effectively applied in real-life investments with just an incremental effort to increase estimation accuracy.

• Quality Improvement in Health Care
• /
• v.3 no.1
• /
• pp.94-103
• /
• 1996

Background : Blood pressure is an important indicator in diagnosis and assessing treatment of a patient. Clinical staffs use blood pressure on the assumption that measured value is accurate and reliable. However, whether measured blood pressure is accurate has been rarely investigated in Korea. Objectives : The aims of this study are to evaluate clinical staffs' knowledge and technique as well as accuracy of sphygmomanometer. Also the program to improve the measurement is developed. Methods : Seventy-three registered nurses were asked nine multiple choice questions including Korotkoff sound, cuff size, and deflation rate. Simultaneously characteristics of nurses were examined, age, working place, duration of employment and academic degree. A testing videotape(Standardizing Measurement Video-Tutored Course) was used for evaluating the accuracy of measurement. Testees were to read and record the 12 cases of blood pressure measurement, watching a falling mercury column and hearing Korotkoff sounds. After 10 minutes' education, they were again tested with the same cases. Additionally, 83 mercury sphygmomanometers were checked to find defects such as inaccurate calibration and zero setting, leaky bladder, etc. Results: For the knowledge testing correct response rate was 41.1%. They were the lowest in selecting the proper cuff size and Korotkoff sound. In examining accuracy of blood pressure with videotape, nurses had 67.7% correct response rate. The correct response rate was significantly improved by a session of education. About 23% of sphygmomanometers was without discernable defects. Conclusion : The knowledge and skill of clinical staffs along with the accuracy of equipment have to be improved. A properly designed education program would contribute to the accuracy improvement of blood pressure measurement. Also, more concerns should be given to the precision and maintenance of equipment.

• International conference on construction engineering and project management
• /
• 2009.05a
• /
• pp.1068-1074
• /
• 2009

The progress is widely used as a critical index for successful construction project management. In spite of the importance of progress measurement, the excessive management effort to collect and maintain detailed data has been highlighted as a major barrier to measurement of highly accurate progress. In order to reduce the required workload and to enhance accuracy, several researches have been conducted. These researches can be categorized into two groups. First group focuses on automated data collection utilizing advanced technologies only for limited construction tasks. The second group is a research area where the standard progress measurement methodologies encompassing entire construction tasks are investigated. Topics include the adjusting the level of details, standardizing work processes, and applying flexible WBS. However, the techniques for automated data collection are not fully investigated yet in the second group. Combining these two research areas can provide a solution for more effective progress management in terms of enhancing accuracy and optimizing workload. However, there has been no comprehensive research addressing these two research groups in an integrated manner. In this context, the purpose of this paper is to propose a methodology that identifies the most suitable measurement method and data acquisition technology (e.g., GPS, RFID, etc.) for entire construction tasks of a project. The proposed methodology in this paper will be able to facilitate the selection process of data acquisition technologies for entire construction tasks of a project and to support the overall enhancement of automated progress management.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.6
• /
• pp.1800-1817
• /
• 2022

With the development of the economy and the improvement of living standards, the hot issues in the subject area have become the main research direction, and the mining of the hot issues in the subject currently has problems such as a large amount of data and a complex algorithm structure. Therefore, in response to this problem, this study proposes a method for extracting hot keywords in scientific journals based on the improved BERT model.It can also provide reference for researchers,and the research method improves the overall similarity measure of the ensemble,introducing compound keyword word density, combining word segmentation, word sense set distance, and density clustering to construct an improved BERT framework, establish a composite keyword heat analysis model based on I-BERT framework.Taking the 14420 articles published in 21 kinds of social science management periodicals collected by CNKI(China National Knowledge Infrastructure) in 2017-2019 as the experimental data, the superiority of the proposed method is verified by the data of word spacing, class spacing, extraction accuracy and recall of hot keywords. In the experimental process of this research, it can be found that the method proposed in this paper has a higher accuracy than other methods in extracting hot keywords, which can ensure the timeliness and accuracy of scientific journals in capturing hot topics in the discipline, and finally pass Use information technology to master popular key words.

• Journal of Communications and Networks
• /
• v.15 no.1
• /
• pp.54-60
• /
• 2013

In this paper, a novel method called location-based delivery (LBD), which combines the short message service (SMS) and global position system (GPS), is proposed, and further, a realistic system for tracking a target's movement is developed. LBD reduces the number of short message transmissions while maintaining the location tracking accuracy within the acceptable range. The proposed approach, LBD, consists of three primary features: Short message format, location prediction, and dynamic threshold. The defined short message format is proprietary. Location prediction is performed by using the current location, moving speed, and bearing of the target to predict its next location. When the distance between the predicted location and the actual location exceeds a certain threshold, the target transmits a short message to the tracker to update its current location. The threshold is dynamically adjusted to maintain the location tracking accuracy and the number of short messages on the basis of the moving speed of the target. The experimental results show that LBD, indeed, outperforms other methods because it satisfactorily maintains the location tracking accuracy with relatively fewer messages.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.5
• /
• pp.1489-1515
• /
• 2022

The number of interconnected real-world devices such as sensors, actuators, and physical devices has increased with the advancement of technology. Due to this advancement, users face difficulties searching for the location of these devices, and the central issue is the findability of Things. In the WoT environment, keyword-based and geospatial searching approaches are used to locate these devices anywhere and on the web interface. A few static methods of indexing and ranking are discussed in the literature, but they are not suitable for finding devices dynamically. The authors have proposed a mechanism for dynamic and efficient searching of the devices in this paper. Indexing and ranking approaches can improve dynamic searching in different ways. The present paper has focused on indexing for improving dynamic searching and has indexed the Things Description in Solr. This paper presents the Things Description according to the model of W3C JSON-LD along with the open-access APIs. Search efficiency can be analyzed with query response timings, and the accuracy of response timings is critical for search results. Therefore, in this paper, the authors have evaluated their approach by analyzing the search query response timings and the accuracy of their search results. This study utilized different indexing approaches such as key-words-based, spatial, and hybrid. Results indicate that response time and accuracy are better with the hybrid approach than with keyword-based and spatial indexing approaches.

• International conference on construction engineering and project management
• /
• 2013.01a
• /
• pp.475-478
• /
• 2013

To carry out a one-off construction project successfully, effective and accurate early cost estimation is crucial, especially during the conceptual stage where very limited minimum information of construction project is given. As the level of accuracy of the early cost estimation has huge impacts on precise budgeting and cost management of a project, in other words, reducing the risk of a project, cost must be managed with special awareness. In an effort to improve the estimate accuracy of cost during the conceptual stage, this research introduces a Parameter Impact (PI) which can quantify weights of parameters and rank them; and PI development derived from the principle of impulse in physics is explicated. For a case study, 76 public apartment building cases in Korea are analyzed. To examine the validity of the proposed PI, a validation in terms of CBR applicability test and estimate accuracy comparisons using 10-nearest neighbor cases are carried out. The validation results support that the suggested PI can be applied in quantifying the weights of the parameters and CBR method for early cost estimation.

• Advances in nano research
• /
• v.16 no.6
• /
• pp.531-548
• /
• 2024

The study focuses on using remote sensing to gather data about the Earth's surface, particularly in urban environments, using satellites and aircraft-mounted sensors. It aims to develop a classification framework for road targets using multi-spectral imagery. By integrating Convolutional Neural Networks (CNNs) with XGBoost, the study seeks to enhance the accuracy and efficiency of road target identification, aiding urban infrastructure management and transportation planning. A novel aspect of the research is the incorporation of quantum sensors, which improve the resolution and sensitivity of the data. The model achieved high predictive accuracy with an MSE of 0.025, R-squared of 0.85, RMSE of 0.158, and MAE of 0.12. The CNN model showed excellent performance in road detection with 92% accuracy, 88% precision, 90% recall, and an f1-score of 89%. These results demonstrate the model's robustness and applicability in real-world urban planning scenarios, further enhanced by data augmentation and early stopping techniques.