• The Korean Journal of Applied Statistics
• /
• v.29 no.6
• /
• pp.1129-1145
• /
• 2016

Functional neuro-connectivity is one of the main issues in brain science in the sense that it is closely related to neurodynamics in the brain. In the paper, we choose fMRI as a main form of response data to brain activity due to its high resolution. We review methods for analyzing functional neuro-connectivity assuming that measurements are made on physiological responses to neuron activation. This means that we deal with a state-space and measurement model, where the state-space model is assumed to represent neurodynamics. Analysis methods and their interpretation should vary subject to what was measured. We included analysis results of real fMRI data by applying a high-dimensional autoregressive model, which indicated that different neurodynamics were required for solving different types of geometric problems.

• The KIPS Transactions:PartD
• /
• v.18D no.6
• /
• pp.493-500
• /
• 2011

The trends of technical terms express the changes of particular subjects in a specific research field over time. However, the amount of academic literature and patent data is too large to be analyzed by human resources. In this paper, we propose a method that can detect and analyze the trends of terms by modeling the life cycle of the terms. The proposed method is composed of the following steps. First, the technical terms are extracted from academic literature data, and the TDVs(Term Dominance Values) of terms are computed on a periodic basis. Based on the TDVs, the life cycles of terms are modeled, and technical terms with similar temporal patterns of the life cycles are classified into the same trends class. The experiments shown in this paper is performed by exploiting the NDSL academic literature data maintained by KISTI.

• Journal of the Korea Society of Computer and Information
• /
• v.18 no.3
• /
• pp.119-126
• /
• 2013

In this paper, we propose and describe the system to visualize the story of contents such as movies and novels. Character-net is applied as story model in order to visualize story. However, it is the form to be accumulated for total movie story, though it can depict the relationship between characters. We have developed the system that analyzes and shows the variation of Character-net and characters' tendency in order to represent story variation depending on movie progression. This system is composed by two windows that can play and analyze sequential Character-nets by time, and can analyze time variant graph of characters' degree centrality. First window has a function that supports to find important story points like the scenes that main characters appear or meet firstly. Second window supports a function that track each character's tendency or a variation of his tendency through analyzing in-degree graph and out-degree. This paper describes the proposed system and discusses additional requirements.

• Journal of Cadastre & Land InformatiX
• /
• v.45 no.1
• /
• pp.123-135
• /
• 2015

In order to satisfy the high expectations of consumers for a variety of consumer's desired subject area, information could be provided in the form of a map according to the analysis information. With the name change in 2015, LX would intend to play a role in building the information infrastructure that can be supported government policy as an intermediary between the government and private sector. Therefore, in this study, we would like to propose a plan that provide personalized information to the consumer. Through compositing a variety of time-series data(inner or outer of LX) based on public information, and analyzing spatially and temporally the rapidly changing land status. For these purpose, prior research and domestic or abroad thematic map service about thematic map making were reviewed. And the reason why the LX makes information map was presented. Also, themes of 3 field were selected, and depending on the data processing or analysis level and theme were subdivided, and then production and expression method were proposed.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.21 no.1
• /
• pp.12-18
• /
• 2011

In this paper, we proposed Interval Type-2 polynomial Radial Basis Function Neural Networks. In the receptive filed of hidden layer, Interval Type-2 fuzzy set is used. The characteristic of Interval Type-2 fuzzy set has Footprint Of Uncertainly(FOU), which denotes a certain level of robustness in the presence of un-known information when compared with the type-1 fuzzy set. In order to improve the performance of proposed model, we used the linear polynomial function as connection weight of network. The parameters such as center values of receptive field, constant deviation, and connection weight between hidden layer and output layer are optimized by Conjugate Gradient Method(CGM) and Space Search Evolutionary Algorithm(SSEA). The proposed model is applied to gas furnace dataset and its result are compared with those reported in the previous studies.

• Journal of the Korea Society of Computer and Information
• /
• v.8 no.3
• /
• pp.113-123
• /
• 2003

Many researches on the application of neural networks for making financial index prediction have proven their advantages over statistical and other methods. In this paper, a neural network model is proposed for the Buying, Holding or Selling timing prediction in stocks by the price index of stocks by inputting the closing price and volume of dealing in stocks and the technical indexes(MACD, Psychological Line). This model has an equalized multi-layer arithmetic function as well as the time series prediction function of backpropagation neural network algorithm. In the case that the numbers of learning data are unbalanced among the three categories (Buying, Holding or Selling), the neural network with conventional method has the problem that it tries to improve only the prediction accuracy of the most dominant category. Therefore, this paper, after describing the structure, working and learning algorithm of the neural network, shows the equalized multi-layer arithmetic method controlling the numbers of learning data by using information about the importance of each category for improving prediction accuracy of other category. Experimental results show that the financial index prediction using the equalized multi-layer arithmetic neural network has much higher correctness rate than the other conventional models.

• Journal of KIISE:Computing Practices and Letters
• /
• v.15 no.8
• /
• pp.534-546
• /
• 2009

Query by humming is an effective and intuitive querying mechanism when a user wants to find a song without knowing lyrics. The query by humming system takes a user-hummed melody as input, compares it with melodies in a music database, and returns top-k similar melodies to the input. In this paper, we propose a novel algebra for query by humming, and design and implement a real query by humming system called HummingBase by exploiting the algebra. By analyzing existing similarity search techniques, we derive 10 core operators for the algebra. By using the well-defined algebra, we can easily implement such a system in a extensible and modular way. With two case studies, we show that the proposed algebra can easily represent the query processing processes of existing query-by-humming systems.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2011.05a
• /
• pp.53-53
• /
• 2011

가뭄은 강수 부족 및 온도 상승에 따른 물 수지의 불균형으로, 그 특성상 점진적이고 홍수에 비해 피해규모가 광범위 하여 효율적인 대처방안을 마련하기가 어려운 특성을 가지고 있다. 현재 국내의 경우 가뭄관리를 위해 비구조적 대책 방안인 가뭄지수를 활용하여 해당 지역의 부족한 용수의 정도를 시 공간적으로 측정하고 크기와 강도에 대한 정량적 또는 정성적인 평가를 수행하고 있다. 그러나 대부분 강수 및 기온자료를 토대로 한 평가가 주를 이루고 있으며, 그나마 제공되는 지수들의 경우 가뭄을 나타내는 기준이 상이하여 사용자에게 많은 혼란을 가중시키고 있는 실정이다. 따라서 효율적인 가뭄관리를 위해서는 장주기 기상정보를 토대로 국가 또는 권역별 가뭄감시가 이루어져야하며, 기상 분만 아니라 지표와의 물 수지 해석이 반영된 수문정보(유량, 토양수분 등) 기반의 가뭄 정보가 생산되어야 할 것이다. 본 연구에서는 전지구 수문해석이 가능한 지표수문해석모형을 활용하여 남한에 대한 수문성분 기반의 가뭄평가를 수행하고자 한다. 우선 남한 전역에 대한 기상 및 지형 정보를 구축하고 지표수문해석모형에 적용하여 격자별 수문성분을 생산하였다. 수문성분은 가뭄평가에 필요한 정보로 전환되어야 하며, 본 연구에서는 빈도해석기법을 적용하여 가뭄에 대한 발생 빈도 및 규모를 정량화 하였다. 즉, 모형에서 산정된 수문정보로 부터 빈도해석을 수행하여 적정 확률분포형을 결정한 후, 해당기간에 대한 확률값을 산정하여 과거 대비 가뭄에 대한 여부를 판단하였다. 산정된 지수에 대한 평가를 위해 국내 과거 가뭄기록사례를 조사 및 기존 가뭄지수인 SPI 및 PDSI를 활용하였다. 평가 방법은 시계열 및 지역별 분석과 유역별 물수지 분석으로 구분되며, 주로 가뭄기간동안의 가뭄심도와 가뭄 발생 및 해갈에 따른 재현여부를 평가하였다. 평가 결과 가뭄발생 및 해갈시기 그리고 피해지역에 대한 표현에 있어 기록된 사항을 적절히 반영하는 것으로 나타났으며, 기존 가뭄지수 보다 가뭄 재현에 있어 비교적 신뢰성이 높은 것으로 확인되었다. 따라서 지표수문해석모형 기반의 가뭄평가의 경우 적용성이 우수한 것으로 판단되며, 이상의 연구결과는 향후 국내 및 동아시아 가뭄감시 전망에 있어 기초자료로 활용될 것이다.

• Journal of Korea Water Resources Association
• /
• v.54 no.12
• /
• pp.1275-1284
• /
• 2021

This paper presents a machine learning technique applied to prediction of time-dependent local scour around bridge piers in both non-cohesive and cohesive beds. The support vector machines (SVM), which is known to be free from overfitting, is used. The time-dependent scour depths are expressed by 7 and 9 variables for the non-cohesive and cohesive beds, respectively. The SVM models are trained and validated with time series data from different sources of experiments. Resulting Mean Absolute Percentage Error (MAPE) indicates that the models are trained and validated properly. Comparisons are made with the results from Choi and Choi's formula and Scour Rate in Cohesive Soils (SRICOS) method by Briaud et al., as well as measured data. This study reveals that the SVM is capable of predicting time-dependent local scour in both non-cohesive and cohesive beds under the condition that sufficient data of good quality are provided.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.11
• /
• pp.1586-1591
• /
• 2022

Kirchhoff-Love plate (KLP) equation is a well established theory for a description of a deformation of a thin plate under certain outer source. Meanwhile, analysis of a vibrating plate in a frequency domain is important in terms of obtaining the main frequency/eigenfunctions and predicting the vibration of plate. Among various modal analysis methods, dynamic mode decomposition (DMD) is one of the efficient data-driven methods. In this work, we carry out DMD based modal analysis for KLP where thin plate is under effects of sine-type outer force. We first construct discrete time series of KLP solutions based on a finite difference method (FDM). Over 720,000 number of FDM-generated solutions, we select only 500 number of solutions for the DMD implementation. We report the resulting DMD-modes for KLP. Also, we show how DMD can be used to predict KLP solutions in an efficient way.