• Proceedings of the Korea Information Processing Society Conference
• /
• 2017.04a
• /
• pp.916-919
• /
• 2017

최근 딥러닝의 발달로 인해 Sentiment analysis분야에서도 다양한 기법들이 적용되고 있다. 이미지, 음성인식 분야에서 높은 성능을 보여주었던 Convolutional Neural Networks (CNN)은 최근 자연어처리 분야에서도 활발하게 연구가 진행되고 있으며 Sentiment analysis에도 효과적인 것으로 알려져 있다. 기존의 머신러닝에서는 lexicon을 이용한 기법들이 활발하게 연구되었지만 word embedding이 등장하면서 이러한 시도가 점차 줄어들게 되었다. 그러나 lexicon은 여전히 sentiment analysis에서 유용한 정보를 제공한다. 본 연구에서는 SemEval 2017 Task4에서 제공한 Twitter dataset과 다양한 lexicon corpus를 사용하여 lexicon을 CNN과 결합하였을 때 모델의 성능이 얼마큼 향상되는지에 대하여 연구하였다. 또한 word embedding과 lexicon이 미치는 영향에 대하여 분석하였다. 모델을 평가하는 metric은 positive, negative, neutral 3가지 class에 대한 macroaveraged F1 score를 사용하였다.

• Journal of Korea Multimedia Society
• /
• v.10 no.6
• /
• pp.760-769
• /
• 2007

In this paper, we consider a group mobility model to formulate a clustering mechanism called Dual Coalescent Energy-Efficient Algorithm (DCEE) which is scalable, distributed and energy-efficient for wireless mesh network. The differences of the network nodes will be distinguished to exploit heterogeneity of the network. Furthermore, a topology control, that is, adjusting the transmission range to further reduce power consumption will be integrated with the cluster formation to improve network lifetime and connectivity. Along with network lifetime and power consumption, clusterhead changes will be measured as a performance metric to evaluate the. effectiveness and robustness of the algorithm.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.32 no.4A
• /
• pp.379-386
• /
• 2007

In this paper, we propose an efficient diagnosis algorithm for multiple stuck-at faults. Because of using vectorwise intersections as an important metric of diagnosis, the proposed diagnosis algorithm can diagnose multiple defects in single stuck-at fault simulator. In spite of multiple fault diagnosis, the number of candidate faults is drastically reduced. For identifying faults, the variable weight, positive calculations and negative calculations are used for the matching algorithm. To verify our algorithm, experiments were performed for ISCAS85 and full-scan version of ISCAS89 benchmark circuits.

• Journal of Information Processing Systems
• /
• v.14 no.4
• /
• pp.904-915
• /
• 2018

We demonstrate how social media content can be used to predict the unemployment rate, a real-world indicator. We present a novel method for predicting the unemployment rate using social media analysis based on natural language processing and statistical modeling. The system collects social media contents including news articles, blogs, and tweets written in Korean, and then extracts data for modeling using part-of-speech tagging and sentiment analysis techniques. The autoregressive integrated moving average with exogenous variables (ARIMAX) and autoregressive with exogenous variables (ARX) models for unemployment rate prediction are fit using the analyzed data. The proposed method quantifies the social moods expressed in social media contents, whereas the existing methods simply present social tendencies. Our model derived a 27.9% improvement in error reduction compared to a Google Index-based model in the mean absolute percentage error metric.

• Journal of Environmental Impact Assessment
• /
• v.7 no.2
• /
• pp.71-82
• /
• 1998

This study focuses on one of the most typical energy-intensive industries, the steel industry. The two-fold purpose of the study is to develop a model to assess measures to alleviate sulfur dioxide($SO_2$) emissions from the steel industry and to propose a concrete $SO_2$ emission reduction measure from the steel industry. This study partially employed and modified AIM(Asia-Pacific Integrated Model) developed by Japan National Environmental Research Institute to develop AIM/KOREA SULFUR model for simulation. In the study, a base scenario, which is BAU(Business As Usual) scenario, and mitigation scenarios(a use of low-sulfur contain fuel, fuel conversion to cleaner energy, an induction of desulfurization systems, and energy saving) were employed. The results of the simulation are summarized below: The sulphur dioxide emission from the steel industry in 1992 was estimated to be 252,000 metric tons; however, according to BAU scenario, sulphur dioxide emission is expected to be increased to 586,000 metric tons, which is 2.3 times greater than that in 1992 by year 2020. To alleviate such increasement, simulation results under various 7scenarios proved that some degrees of reduction may be possible by an induction of desulfurization systems although there may be numerous ways to interpretate the simulation results; however, the bottom line is that it appears to be difficult to achieve the Korean Ministry of Environment's policy goal-a mitigation of sulphur dioxide concentration to 0.01ppm.

• Korean Journal of Ecology and Environment
• /
• v.56 no.1
• /
• pp.70-82
• /
• 2023

Recently, Korean government has introduced Multi Metric Indices (MMI) using various biocommunity information for aquatic ecosystem monitoring and ecosystem health assessment at the national level. MMI is a key tool in national ecosystem health assessment programs. The MMI consists of indices that respond to different target environmental factors, including environmental disturbance (e.g. nutrients, hydrological and hydraulic situation of site etc.). We used zooplankton community information collected from Korean lakes to estimate the availability of candidate zooplankton MMI indices that can be used to assess lake ecosystem health. First, we modified the candidate indices proposed by the U.S. EPA to suit Korean conditions. The modified indices were subjected to individual index suitability analysis, correlation analysis with environmental variables, and redundancy analysis among indices, and 19 indices were finally selected. Taxonomic diversity was suggested to be an important indicator for all three taxonomic groups (cladoceran, copepod, rotifer), on the other hand, the indices using biomass for large cladocerans and copepods, while the indices using abundance were suggested for small cladocerans and rotifers.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.11
• /
• pp.1463-1470
• /
• 2021

With the development of ICT technology, a monitoring system to check the status of an object to be managed in real time in various industrial fields is widely used. Existing monitoring systems implemented individual systems according to monitoring targets, but recently, monitoring systems have been implemented using open sources such as Prometheus and Grafana. When using Prometheus and Grafana, many parts become more convenient compared to the existing monitoring system development method, but there are still problems. In this paper, to solve this problem, we propose an integrated monitoring system that supports Prometheus and Grafana. The proposed system is a detailed module that collects, stores, visualizes, and manages data to be monitored, and each module is implemented so that roles can be divided and existing problems can be solved. The proposed system can conveniently manage and monitor monitoring targets stored in heterogeneous databases, and create dashboards through simple operation.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.11 no.4
• /
• pp.1171-1177
• /
• 2010

This paper presents a robust design procedure for minimizing warpage in plastic injection-molded products, where the Pick-the-Winner rule based on Taguchi's Orthogonal Array experiments and the Design Space Reduction Method are integrated for optimization. Two-step optimization approach is applied to reduce warpage in the part design stage and additionally to minimize the warpage in the process conditions design stage. Taguchi's S/N ratio is introduced as a design metric to evaluate robustness against process variations. The effectiveness of proposed optimization process is shown with an example of warpage minimization problem.

• JSTS:Journal of Semiconductor Technology and Science
• /
• v.17 no.1
• /
• pp.65-78
• /
• 2017

Single Event Transient has a critical impact on highly integrated logic circuits which are currently common in various commercial and consumer electronic devices. Reliability against the soft and intermittent faults will become a key metric to evaluate such complex system on chip designs. Our previous work analyzing soft errors was focused on parallelizing and optimizing error propagation procedures for individual transient faults on logic and sequential cells. In this paper, we present a new propagation technique where a fault binary decision diagram (BDD) continues to merge every new fault generated from the subsequent logic gate traversal. BDD-based transient fault analysis has been known to provide the most accurate results that consider both electrical and logical properties for the given design. However, it suffers from a limitation in storing and handling BDDs that can be increased in size and operations by the exponential order. On the other hand, the proposed method requires only a visit to each logic gate traversal and unnecessary BDDs can be removed or reduced. This results in an approximately 20-200 fold speed increase while the existing parallelized procedure is only 3-4 times faster than the baseline algorithm.

• Journal of Construction Engineering and Project Management
• /
• v.2 no.1
• /
• pp.27-34
• /
• 2012

Recent requirements in component management of building systems have focused on the requirement for improving methods and metric tools to support component condition assessment and appropriate decisions for infrastructure owned facilities. Although engineers and researchers have focused on developing methodologies for component assessment in recent years but there is not enough attention dedicate to facilities and components that have been constructed. This paper is a literature study of scientific papers within the topic of component condition index system (CCIS) in the period 1976 to 2009. Infrastructure component condition index had existed for some 40 years. The purpose of this paper is to provide an overview of CCIS to identify the suitable method for component condition assessment during its service life. This paper finds that the focus of CCIS, surveyed in several aspects during the 40 years that have been investigated, from technology to measurement and from assessment function to component maintenance as an integrated part of the infrastructure component management. This study offers help to researchers in understanding the selection of an appropriate method for component condition assessment in building and non-building systems.