• Journal of the Korean Operations Research and Management Science Society
• /
• v.41 no.2
• /
• pp.81-100
• /
• 2016

Much effort has been exerted to analyze online texts and understand how empirical results can help improve sales performance. In this research, we aim to extend this stream of research by decomposing online texts based on text sources, namely, companies and consumers. To be specific, we investigate how online texts driven by companies differ from those generated by consumers, and the extent to which both types of online texts have different effects on online sales. We obtained sales data from one of the biggest game publishers and merged them with online texts provided by companies using news articles and those created by consumers in user communities. The empirical analyses yield the following findings. Word visualization and topic analyses show that firms and consumers generate different contexts. Specifically, companies spread word to promote their own events whereas consumers produce online words to share winning strategies. Moreover, online sales are influenced by consumer-generated community topics whereas firm-driven topics in news articles have little to no effect. These findings suggest that companies should focus more on online texts generated by consumers rather than spreading their own words. Moreover, online sales strategies should take advantage of specific topics that have been proven to increase online sales. In particular, these findings give startup companies and small business owners in variety of industries the advantage when they use the online channel for distribution and as a marketing platform.

• Computers and Concrete
• /
• v.32 no.2
• /
• pp.217-232
• /
• 2023

Numerous studies have been performed on the behavior of pile foundations in cold regions. This study first attempted to employ artificial neural networks (ANN) to predict pile-bearing capacity focusing on pile data recorded primarily on cold regions. As the ANN technique has disadvantages such as finding global minima or slower convergence rates, this study in the second phase deals with the development of an ANN-based predictive model improved with an Elephant herding optimizer (EHO), Dragonfly Algorithm (DA), Genetic Algorithm (GA), and Evolution Strategy (ES) methods for predicting the piles' bearing capacity. The network inputs included the pile geometrical features, pile area (m²), pile length (m), internal friction angle along the pile body and pile tip (Ø°), and effective vertical stress. The MLP model pile's output was the ultimate bearing capacity. A sensitivity analysis was performed to determine the optimum parameters to select the best predictive model. A trial-and-error technique was also used to find the optimum network architecture and the number of hidden nodes. According to the results, there is a good consistency between the pile-bearing DA-MLP-predicted capacities and the measured bearing capacities. Based on the R2 and determination coefficient as 0.90364 and 0.8643 for testing and training datasets, respectively, it is suggested that the DA-MLP model can be effectively implemented with higher reliability, efficiency, and practicability to predict the bearing capacity of piles.

• Journal of Wetlands Research
• /
• v.26 no.2
• /
• pp.147-159
• /
• 2024

Lack of streamflow observations makes model calibration difficult and limits model performance improvement. Satellite-based remote sensing products offer a new alternative as they can be actively utilized to obtain hydrological data. Recently, several studies have shown that artificial intelligence-based solutions are more appropriate than traditional conceptual and physical models. In this study, a data-driven approach combining various recurrent neural networks and decision tree-based algorithms is proposed, and the utilization of satellite remote sensing information for AI training is investigated. The satellite imagery used in this study is from MODIS and SMAP. The proposed approach is validated using publicly available data from 25 watersheds. Inspired by the traditional regionalization approach, a strategy is adopted to learn one data-driven model by integrating data from all basins, and the potential of the proposed approach is evaluated by using a leave-one-out cross-validation regionalization setting to predict streamflow from different basins with one model. The GRU + Light GBM model was found to be a suitable model combination for target basins and showed good streamflow prediction performance in ungauged basins (The average model efficiency coefficient for predicting daily streamflow in 25 ungauged basins is 0.7187) except for the period when streamflow is very small. The influence of satellite remote sensing information was found to be up to 10%, with the additional application of satellite information having a greater impact on streamflow prediction during low or dry seasons than during wet or normal seasons.

• Journal of Information Science Theory and Practice
• /
• v.12 no.3
• /
• pp.49-62
• /
• 2024

Strategy monitoring is essential for business management and for administrators, including managers and executives, to build a data-driven organization. Having a tool that is able to visualize strategic data is significant for business intelligence. Unfortunately, there are gaps between business users and information technology departments or business intelligence experts that need to be filled to meet user requirements. For example, business users want to be self-reliant when using business intelligence systems, but they are too inexperienced to deal with the technical difficulties of the business intelligence systems. This research aims to create an automatic matching framework between the key performance indicators (KPI) formula and the data in database systems, based on ontology concepts, in the case study of Prince of Songkla University. The mapping data schema with ontology (MapDSOnto) framework is created through knowledge adaptation from the literature review and is evaluated using sample data from the case study. String similarity methods are compared to find the best fit for this framework. The research results reveal that the "fuzz.token_set_ratio" method is suitable for this study, with a 91.50 similarity score. The two main algorithms, database schema mapping and domain schema mapping, present the process of the MapDS-Onto framework using the "fuzz.token_set_ratio" method and database structure ontology to match the correct data of each factor in the KPI formula. The MapDS-Onto framework contributes to increasing self-reliance by reducing the amount of database knowledge that business users need to use semantic business intelligence.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.15 no.4
• /
• pp.68-80
• /
• 2016

This study proposes the traffic prediction and optimal traffic control system based on cell transmission model and genetic algorithm in cloud environment. The proposed prediction and control system consists of four parts. 1) Data preprocessing module detects and imputes the corrupted data and missing data points. 2) Data-driven traffic prediction module predicts the future traffic state using Multi-level K-Nearest Neighbor (MK-NN) Algorithm with stored historical data in SQL database. 3) Online traffic simulation module simulates the future traffic state in various situations including accident, road work, and extreme weather condition with predicted traffic data by MK-NN. 4) Optimal road control module produces the control strategy for large road network with cell transmission model and genetic algorithm. The results show that proposed system can effectively reduce the Vehicle Hours Traveled upto 60%.

• Korean Journal of Construction Engineering and Management
• /
• v.20 no.2
• /
• pp.95-104
• /
• 2019

EPC plant projects are large and complex, requiring systematic working methodologies, accumulated data, and thorough planning through communications between the entities. In this study, the method of extracting the process planning information using asset data of the plant project and using it to present the initial process plan is presented through the concept of IAP(Interactive Planning). In order to carry out the effective IAP at the early stage of the project, this study extracted the schedule element information from the asset data, created the process plan for each work package, and applied it to the sample project case. Through the proposed IAP methodology, it is possible to promote the utilization of asset data, to identify schedule risks, and to develop countermeasures, which can form the basis for establishing the process management strategy to successfully complete the project.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.11 no.4
• /
• pp.182-192
• /
• 2008

In oder to develop a decision supporting system for oil spill response, the prototype of pollution response support system which has integrated oil spill prediction system and pollution risk prediction system has developed for Incheon-Daesan area. Spill prediction system calculates oil spill aspects based on real-time wind data and real-time water flow and the residual volume of spilt oil and spread pattern are calculated considering the characteristic of spilt oil. In this study, real-time data is created from results of real-time meteorological forecasting model(National Institute of Environmental Research) using ftp, real-time tidal currents datasets are built using CHARRY(Current by Harmonic Response to the Reference Yardstick) model and real-time wind-driven currents are calculated applying the correlation function between wind and wind-driven currents. In order to model the feature which is spilt oil spreading according to real-time water flow is weathered, the decrease ratio by oil kinds was used. These real-time data and real-time prediction information have been integrated with ESI(Environmental Sensitivity Index) and response resources and then these are provided using GIS as a whole system to make the response strategy.

• Resources Recycling
• /
• v.31 no.4
• /
• pp.3-11
• /
• 2022

Owing to the increasing demand for electric vehicles (EVs), appropriate management of their waste batteries is required urgently for scrapped vehicles or for addressing battery aging. With respect to technological developments, data-driven diagnosis of waste EV batteries and management technologies have drawn increasing attention. Moreover, robot-based automatic dismantling technologies, which are seemingly interesting, require industrial verifications and linkages with future battery-related database systems. Among these, it is critical to develop and disseminate various advanced battery diagnosis and assessment techniques to improve the efficiency and safety/environment of the recirculation of waste batteries. Incorporation of lithium-related chemical substances in the public pollutant release and transfer register (PRTR) database as well as in-depth risk assessment of gas emissions in waste EV battery combustion and their relevant fire safety are some of the necessary steps. Further research and development thus are needed for optimizing the lifecycle management of waste batteries from various aspects related to data-based diagnosis/classification/disassembly processes as well as reuse/recycling and final disposal. The idea here is that the data should contribute to clean design and manufacturing to reduce the environmental burden and facilitate reuse/recycling in future production of EV batteries. Such optimization should also consider the future technological and market trends.

• Journal of Korean Society for Quality Management
• /
• v.38 no.4
• /
• pp.521-530
• /
• 2010

Global regulatory pressures relating climate change and environmental responsibility are asking companies to find out the best way for sustaining their continuous business growths. It could be known that inadequate management for environmental issues are bad for business, negatively affecting brand image, causing unnecessary losses and costs for environmental preservation. For this reason, environmentally conscious green business growth has been recognized as an essential requirement for a company to stay in business. Many companies are looking for green business opportunities of improving their environmental and financial results, and struggling with how green fits into their business. In this paper, the Green Six Sigma, an environmentally conscious Six Sigma methodology, is presented as a way to find solutions for green growths. The Six Sigma is known as a disciplined, data-driven approach and methodology for achieving world-class performance in any process from manufacturing to transactional. In chronological order, the Six Sigma has been evolved from Motorola's quality-oriented methodology to GE's cost-oriented lean approach, and is being evolved and developed as an environment-oriented green growth approach. There is no doubt that the Green Six Sigma, as an engine of green growth, is a power tool for achieving competitive business performance and reducing the impact on the environment.

• ETRI Journal
• /
• v.34 no.3
• /
• pp.308-318
• /
• 2012

As mobile markets in most developed countries are rapidly coming close to saturation, it is increasingly challenging to cover the cost of providing the network, as revenues are not growing. This has driven mobile operators, thus far mostly involved in facility-based competition, to turn their attention to network sharing. There exist various types of mobile network sharing (MNS), from passive to active sharing. In this paper, we propose a model, based on the supply-demand model, for evaluating the economic effects of using six types of MNS. Our study measures the economic effects of employing these six types of MNS, using actual WiBro-related data. Considering lower service price and expenditure reduction, the total economic effect from a year's worth of MNS use is estimated to be between 513 million and 689 million USD, which is equal to three to four percent of the annual revenue of Korean mobile operators. The results of this study will be used to support the establishment of a MNS policy in Korea. In addition, the results can be used as a basic model for developing various network sharing models.