• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.1
• /
• pp.87-92
• /
• 2021

Bias, variance, error and learning are important factors for performance in modeling a big data based recommendation system. The recommendation model in this system must reduce complexity while maintaining the explanatory diagram. In addition, the sparsity of the dataset and the prediction of the system are more likely to be inversely proportional to each other. Therefore, a product recommendation model has been proposed through learning the similarity between products by using a factorization method of the sparsity of the dataset. In this paper, the generalization ability of the model is improved by applying the max-norm regularization as an optimization method for the loss function of this model. The solution is to apply a stochastic projection gradient descent method that projects a gradient. The sparser data became, it was confirmed that the propsed regularization method was relatively effective compared to the existing method through lots of experiment.

• The Journal of the Acoustical Society of Korea
• /
• v.26 no.6
• /
• pp.259-268
• /
• 2007

This paper contributes to an improvement of the statistical bandwidth extension(BWE) system based on Hidden Markov Model(HMM). First, the existing HMM training method for BWE, which is suggested originally by Jax, is analyzed in comparison with the general Baum-Welch training method. Next, based on this analysis, a new HMM-based BWE method is suggested which adopts the Baum-Welch re-estimation algorithm instead of the Jax's to train HMM model. Conclusionally speaking, the Baum-Welch re-estimation algorithm is a generalized form of the Jax's training method. It is flexible and adaptive in modeling the statistical characteristic of training data. Therefore, it generates a better model to the training data, which results in an enhanced BWE system. According to experimental results, the new method performs much better than the Jax's BWE systemin all cases. Under the given test conditions, the RMS log spectral distortion(LSD) scores were improved ranged from 0.31dB to 0.8dB, and 0.52dB in average.

• Journal of the Korea Concrete Institute
• /
• v.26 no.2
• /
• pp.159-169
• /
• 2014

This study generalizes the lateral load-displacement relationship of reinforced concrete shear walls from the section analysis for moment-curvature response to straightforwardly evaluate the flexural capacity and ductility of such members. Moment and curvature at different selected points including the first flexural crack, yielding of tensile reinforcing bar, maximum strength, 80% of the maximum strength at descending branch, and fracture of tensile reinforcing bar are calculated based on the strain compatibility and equilibrium of internal forces. The strain at extreme compressive fiber to determine the curvature at the descending branch is formulated as a function of reduction factor of maximum stress of concrete and volumetric index of lateral reinforcement using the stress-strain model of confined concrete proposed by Razvi and Saatcioglu. The moment prediction models are simply formulated as a function of tensile reinforcement index, vertical reinforcement index, and axial load index from an extensive parametric study. Lateral displacement is calculated by using the moment area method of idealized curvature distribution along the wall height. The generalized lateral load-displacement relationship is in good agreement with test result, even at the descending branch after ultimate strength of shear walls.

• Journal of Korea Multimedia Society
• /
• v.13 no.2
• /
• pp.216-224
• /
• 2010

A RAM-based Neural Network(RNN) has the advantages of processing speed and hardware implementation. In spite of these advantages, it has a saturation problem, weakness of repeated learning and extract of a generalized pattern. To resolve these problems of RNN, the 3DNS model using cumulative multi discriminator was proposed. But that model does not solve the saturation problem yet. In this paper, we proposed a adaptive weight cumulative neural net(AWCNN) using the adaptive weight neuron (AWN) for solving the saturation problem. The proposed nets improved a recognition rate and the saturation problem of 3DNS. We experimented with the MNIST database of NIST without preprocessing. As a result of experimentations, the AWCNN was 1.5% higher than 3DNS in a recognition rate when all input patterns were used. The recognition rate using generalized patterns was similar to that using all input patterns.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.31 no.2A
• /
• pp.119-127
• /
• 2006

In this two-part paper, we consider dynamic resource allocation in orthogonal frequency division multiple access(OFDMA). To reduce the reverse link overhead for channel quality information(CQI) feedback, a set of sub-carriers are tied up to a sub-channel to be used as the unit of CQI feedback, user-multiplexing and the corresponding power/rate allocation. Specifically, we focus on two sub-channel structures, either aggregated or distributed, where the SNR distribution over a sub-channel is modeled as Ricean in general, and the channel quality of a sub-channel is summarized as the mean and variance of channel gain envelop divided by noise standard deviation. Then, we develop a generalized two step channel/resource allocation algorithm, which uses the two statistical measurements, and analyze the spectral efficiency of the OFDMA system in terms of average frequency utilization. An extension to proportional fair algorithm will also be addressed. As confirmed by numerical results, the aggregated structure is preferred especially when intending aggressive link adaptation.

• Journal of the Economic Geographical Society of Korea
• /
• v.11 no.1
• /
• pp.78-93
• /
• 2008

Location is the key theme in geography and various location theories are well established for diverse economic activities. As normative theory, the logical presentation of the location theories tends to make students appraise the class difficult and lose interests. This study presents a teaching materials and inductive method for the location of venture firms using newspaper articles as an example of the knowledge and information industry. Presented are various newspaper articles about venture firms in which the locational notions are selected to find out general location factors. And then their locational preference of large cities, diseconomies of agglomeration, the possibilities of deconcentration from large cities and new establishments in peripheral regions are presented in order. The inductive way of teaching location from specific examples to general tendencies using NIE could bring students' interests and participation which is in line with the student-oriented inquiry-based teaching emphasized recently in educational methods. It could be an effective teaching method for the location theories in economic geography and other difficult subjects.

• Journal of the Korea Society of Computer and Information
• /
• v.19 no.1
• /
• pp.149-156
• /
• 2014

NGE(Nested Generalized Exemplar Method) can increase the performance of the noisy data at the same time, can reduce the size of the model. It is the optimal distance-based classification method using a matching rule. NGE cross or overlap hyperrectangles generated in the learning has been noted to inhibit the factors. In this paper, We propose the DHGen(Dominant Hyperrectangle Generation) algorithm which avoids the overlapping and the crossing between hyperrectangles, uses interval weights for mixed hyperrectangles to be splited based on the mutual information. The DHGen improves the classification performance and reduces the number of hyperrectangles by processing the training set in an incremental manner. The proposed DHGen has been successfully shown to exhibit comparable classification performance to k-NN and better result than EACH system which implements the NGE theory using benchmark data sets from UCI Machine Learning Repository.

• School Mathematics
• /
• v.13 no.4
• /
• pp.581-598
• /
• 2011

Given the growing agreement that algebra should be taught in the early stage of the curriculum, considerable studies have been conducted with regard to early algebra in the elementary school. However, there has been lack of research on how to organize mathematic lessons to develop of algebraic reasoning ability of the elementary school students. This research attempted to gain specific and practical information on effective algebraic teaching and learning in the elementary school. An exploratory qualitative case study was conducted to the fourth graders. This paper focused on the associative law of the multiplication. This paper showed what kinds of activities a teacher may organize following three steps: (a) focus on the properties of numbers and operations in specific situations, (b) discovery of the properties of numbers and operations with many examples, and (c) generalization of the properties of numbers and operations in arbitrary situations. Given the steps, this paper included an analysis on how the students developed their algebraic reasoning. This study provides implications on the important factors that lead to the development of algebraic reasoning ability for elementary students.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.11
• /
• pp.3566-3575
• /
• 2000

In this paper, we propose an simple and efficient method to estunate the camera operation by using compressed information, which is extracted diracily from MPEG2 stream without complete decoding. In the method, the motion vector is converted into approximate optical flow by using the feature of predicted frame, because the motion vector in MPEG2 video stream is not regular sequene. And they are used to estimate the camera operation, which consist of pan, and zoom by Hough transform technique. The method provided better results than the least square method for video stream of basketball and socer games. The proposed method can have a reduced computational complexity because the information is directiv abtained in compressed domain. Additionally it can be a useful technology in content-based searching and analysis of video information. Also, the estimatd cameral operationis applicable in searching or tracking objects in MPEG2 video stream without decoding.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.10 no.7
• /
• pp.1473-1479
• /
• 2009

This study have been carried out as a part of R&D project to develop a low cost high efficiency combined grain drying & storage system. The design of sweeping auger must meet the various conditions : capacity and dimensions of silo, discharging capacity of auger, operation conditions of auger such as revolution and rotation and density of grains and even the first-in & first-out of grains in the system. Through the experimental observations with the apparatus which enables the direct observation of discharging behavior of grains, the limitations of the performance of existing straight type auger was observed. Generalized mathematical model for the profile of new variable section sweeping auger was obtained, which ensures the uniform descending of grains and also meets various operating conditions. The experimental results with a prototype sweeping auger showed that the mathematical model for the variable section sweeping auger was quitely correct.