• Journal of KIISE
• /
• v.43 no.1
• /
• pp.80-86
• /
• 2016

A speech-act is a behavior intended by users in an utterance. Speech-act classification is important in a dialogue system. The machine learning and rule-based methods have mainly been used for speech-act classification. In this paper, we propose a speech-act classification method based on the combination of support vector machine (SVM) and transformation-based learning (TBL). The user's utterance is first classified by SVM that is preferentially applied to categories with a low utterance rate in training data. Next, when an utterance has negative scores throughout the whole of the categories, the utterance is applied to the correction phase by rules. The results from our method were higher performance over the baseline system long with error-reduction.

• The KIPS Transactions:PartB
• /
• v.9B no.5
• /
• pp.579-586
• /
• 2002

The edge descriptor proposed by MPEG-7 standard is a representative approach for the contents-based image retrieval using the edge information. In the edge descriptor, the edge information is the edge histogram derived from a gray-level value image. This paper proposes a new method which extracts color edge information from color images and a new approach for the contents-based image retrieval based on the color edge histogram. The poposed method and technique are applied to image retrieval of the e-catalogue. For the evaluation, the results of image retrieval using the proposed approach are compared with those of image retrieval using the edge descriptor by MPEG-7 and the statistics shows the efficiency of the proposed method. The proposed color edge model is made by combining the R,G,B channel components vectorially and by characterizing the vector norm of the edge map. The color edge histogram using the direction of the color edge model is subsequently used for the contents-based image retrieval.

• Journal of the Korea Convergence Society
• /
• v.10 no.4
• /
• pp.17-23
• /
• 2019

In this paper, quality control (QC) is applied to each meteorological element of weather data collected from seven IoT sensors such as temperature. In addition, we propose a method for estimating the data regarded as error by means of machine learning. The collected meteorological data was linearly interpolated based on the basic QC results, and then machine learning-based QC was performed. Support vector regression, decision table, and multilayer perceptron were used as machine learning techniques. We confirmed that the mean absolute error (MAE) of the machine learning models through the basic QC is 21% lower than that of models without basic QC. In addition, when the support vector regression model was compared with other machine learning methods, it was found that the MAE is 24% lower than that of the multilayer neural network and 58% lower than that of the decision table on average.

• Journal of the Korean Geographical Society
• /
• v.47 no.5
• /
• pp.677-699
• /
• 2012

Climate changes may cause substantial changes in spatial patterns and distribution of vector-borne diseases (VBD's), which will result in a significant threat to humans and emerge as an important public health problem that the international society needs to solve. As global warming becomes widespread and the Korean peninsula characterizes subtropical climate, the potentials of climate-driven disease outbreaks and spread rapidly increase with changes in land use, population distributions, and ecological environments. Vector-borne diseases are typically infected by insects such as mosquitoes and ticks, and infected hosts and vectors increased dramatically as the habitat ranges of the VBD agents have been expanded for the past 20 years. Medical geography integrates and processes a wide range of public health data and indicators at both local and regional levels, and ultimately helps researchers identify spatiotemporal mechanism of the diseases determining interactions and relationships between spatial and non-spatial data. Spatial epidemiology is a new and emerging area of medical geography integrating geospatial sciences, environmental sciences, and epidemiology to further uncover human health-environment relationships. An introduction of GIS-based disease monitoring system to the public health surveillance system is among the important future research agenda that medical geography can significantly contribute to. Particularly, real-time monitoring methods, early-warning systems, and spatial forecasting of VBD factors will be key research fields to understand the dynamics of VBD's.

• Communications for Statistical Applications and Methods
• /
• v.16 no.3
• /
• pp.449-461
• /
• 2009

Higher sigma quality level is generally perceived by customers as improved performance by assigning a correspondingly higher satisfaction score. The process capability indices and the sigma level $Z_{st}$ ave been widely used in six sigma industries to assess process performance. Most evaluations on process capability indices focus on statistical estimation under normal process which may result in unreliable assessments of process performance. In this paper, we consider statistical estimation for bivariate VPCI(Vector-valued Process Capability Index) $C_{pkl}=(C_{pklx},\;C_{pklx})$ under Marshall and Olkin (1967)'s bivariate exponential process. First, we derive some limiting distribution for statistical inference of bivariate VPCI $C_{pkl}$. And we propose two asymptotic normal confidence regions for bivariate VPCI $C_{pkl}$. The proposed method may be very useful under bivariate exponential process. A numerical result based on our proposed method shows to be more reliable.

• Journal of the Korean Society for information Management
• /
• v.7 no.1
• /
• pp.79-95
• /
• 1990

A retrieval technique was defined as a technique for comparing the document representations. So this study classified retrieval technique in terms of the charactristics of the retrieved set of documents and the representations that are used. The distinction is whether the set of retrieved documents contains only documents whose representations are an exact match with the query, or a partial match with query. For a partial match, the set of retrieved document will include also those that are an exact match with the query. Boolean-logic as one of the exact match retrieval techniques is in current in most of the large operational information retrieval systems despite of its problems and limitatlons. Partial match as an alternative technique has also various problems. Existing information retrieval systems are successful in aSSisting the user whose needs are well- defined (e.g. Boolean-logic), to retrieve relevant documents but it should be successful in providing retrieval assistance to the browser whose information requirements is ill-defined.

• Journal of KIISE:Software and Applications
• /
• v.32 no.12
• /
• pp.1260-1270
• /
• 2005

Support Vector Machines (SVM) is well known for a representative learner as one of the kernel methods. SVM which is based on the statistical learning theory shows good generalization performance and has been applied to various pattern recognition problems. However, SVM is basically to deal with a two-class classification problem, so we cannot solve directly a multi-class problem with a binary SVM. One-Per-Class (OPC) and All-Pairs have been applied to solve the face recognition problem, which is one of the multi-class problems, with SVM. The two methods above are ones of the output coding methods, a general approach for solving multi-class problem with multiple binary classifiers, which decomposes a complex multi-class problem into a set of binary problems and then reconstructs the outputs of binary classifiers for each binary problem. In this paper, we introduce the output coding methods as an approach for extending binary SVM to multi-class SVM and propose new output coding schemes based on the Error-Correcting Output Codes (ECOC) which is a dominant theoretical foundation of the output coding methods. From the experiment on the face recognition, we give empirical results on the properties of output coding methods including our proposed ones.

• (The) Research of the performance art and culture
• /
• no.43
• /
• pp.293-322
• /
• 2021

The present study examined the scientific awareness appearing in Korean tokusatsu series by focusing on Vectorman: Warriors of the Earth. As a work representing Korean tokusatsu series, Vectorman: Warriors of the Earth achieved the greatest success among tokusatsu series. This work was released thanks to the continued popularity of Japanese tokusatsu since the mid-1980s and the trend of robot animations. Due to the chronic problems regarding Korean children's programs-the oversupply of imported programs and repeated reruns-the need for domestically produced children's programs has continued to come to the fore. However, as the popularity of Korean animation waned beginning in the mid-1990s, inevitably the burden fr producing animation increased. As a result, Vectorman: Warriors of the Earth was produced as a tokusatsu rather than an animation, and because this was a time when an environment for using special effects technology was being fostered in broadcasting stations, computer visual effects were actively used for the series. The response to the new domestically produced tokusatsu series Vectorman: Warriors of the Earth was explosive. The Vectorman series explained the abilities of cosmic beings by using specific scientific terms such as DNA synthesis, brain cell transformation, and special psychological control device instead of ambiguous words like the scientific technology of space. Although the series is unable to describe in detail about the process and cause, the way it defines technology using concrete terms rather than science fiction shows how scientific imagination is manifesting in specific forms in Korean society. Furthermore, the equal relationship between Vectorman and the aliens shows how the science of space, explained with the scientific terms of earth, is an expression of confidence regarding the advancement of Korean scientific technology which represents earth. However, the female characters fail to gain entry into the domain of science and are portrayed as unscientific beings, revealing limitations in terms of scientific awareness.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.10b
• /
• pp.88-90
• /
• 2004

대표적인 현재 및 미래 위치 색인인 TPR-Tree는 이동 객체의 위치 좌표와 속도 벡터 정보를 이용하여 시간에 대해 선형적으로 이동 객체의 현재 및 미래 위치를 예측한다. 그러나 이동 객체의 이동 방향 및 속도가 특정한 임계값을 벗어날 경우에는 서버로 새로운 위치 보고를 수행하기 때문에, 차량과 같이 이동 방향과 속도가 빈번하게 변하는 환경에 적용할 경우 서버로 잦은 보고를 필요로 하게 되어 통신비용을 크게 증가시키는 문제가 있다. 통신비용을 일정하게 유지하기 위한 방법으로 이동 객체의 보고를 일정한 시간 간격으로 수행하게 하는 방법이 있다. 그러나 일정한 시간 간격으로 이동 객체의 위치 보고가 수행되는 환경에서는 보고간격 사이에 속도와 방향이 변하게 되면 시간에 대해 선형적인 위치 예측 시에 오차가 발생할 수 있다. 본 논문에서는 일정한 시간 간격으로 이동 객체의 위치 보고가 수행되는 환경에서 보고 간격 사이에 이동객체의 이동 속도와 방향의 변화에 대한 불확실성을 반영하기 위하여 도로 네트워크를 이용한 이동 객체의 불확실 위치데이터의 질의 처리 기법을 제시한다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.04c
• /
• pp.552-554
• /
• 2003

전자우편은 기존 우편 기능을 대체하는 대표적인 정보 전달 수단으로 자리 잡고 있다. 전자매일 사용자의 증가에 따라 망은 기업들은 전자 메일을 통해 광고를 하게 되었다. 이에 따라 전자매일 사용자들은 인터넷 상에 개인 전자메일 주소가 노출됨으로 많은 스팸메일을 수신하게 되는데, 이것은 전자메일 사용자에게 많은 부담이 되고있다. 본 논문은 전자우편 문서내의 단어들을 대상으로 통계적 방법의 SVM을 이용하여 스팸메일을 필터링 하였으며, 학습 단계에서 단어 자질공간의 축소를 위해 DF값 변화에 따른 학습을 통하여 분류의 성능을 비교하였다. SVM의 성능 평가를 위해 확률적 방법의 나이브 베이지안과 벡터 모텔을 이용한 분류기와 성능을 비교함으로써 SVM 방법이 우수한 성능을 보임을 검증하였다.