In this paper, we proposed an auto-encoder model of observation-wise linear transformation to reduce the dimensionality of data. While nonlinear models can reduce the dimensionality more effectively than linear models, such as the principal component analysis, the non-linear methods can hardly provide a simple linear relationship between the original and the dimensionally reduced data. The proposed model overcomes this difficulty while maintaining the effectiveness of the dimensionality reduction. We assessed the proposed model and compared with PCA and a typical auto-encoder model in terms of the loss function and the degree of reconstruction of the original data. By applying the proposed method to a public data of MNIST and Fashion-MNIST, we showed the effectiveness in the dimensionality reduction and relationship between the original data to the reduced data.

In this study, we proposed a model for forecasting power energy demand by investigating how outside temperature at a given time affected power consumption and. To this end, we analyzed the time series of power consumption in terms of the power spectrum and found the periodicities of one day and one week. With these periodicities, we investigated two time series of temperature and power consumption, and found, for a given hour, an approximate linear relation between temperature and power consumption. We adopted an exponential smoothing model to examine the effect of the linearity in forecasting the power demand. In particular, we adjusted the exponential smoothing model by using the variation of power consumption due to temperature change. In this way, the proposed model became a mixture of a time series model and a regression model. We demonstrated that the adjusted model outperformed the exponential smoothing model alone in terms of the mean relative percentage error and the root mean square error in the range of 3%~8% and 4kWh~27kWh, respectively. The results of this study can be used to the energy management system in terms of the effective control of the cross usage of the electric energy together with the outside temperature.

The base quality score recalibration (BQSR) is an important step in the variant calling from high-throughput sequence data. Motivated by the fact that BQSR necessarily requires a database of known variants such as the dbSNP, we present an extensive analysis on BQSR results for human and rice genome. We showed that the recalibration results depended on the size of the database. The more variants are there in the database, the larger averaged value of the recalibrated quality scores is obtained. This implies that the recalibrated quality score is lower than it should be when the number of variants in the database is not large enough. Based on the finding that the size of the database should play a crucial role in BQSR, we proposed a method to create a database when the size of a database is not large enough for BQSR results to be reliable. We demonstrated that, in the case of human, the database constructed by the proposed method generated almost the same results as the human dbSNP. In the case of rice, however, we showed that the proposed database is more reasonable than the rice dbSNP by illustrating how the proposed method is effective.

Joint order and transportation system which can be used for the franchise business is considered. It is assumed that several agents exist on the same vehicle route and each agent has an EOQ inventory policy and the same order cycle. It is also assumed that transportation cost is proportional to the distance of farthest agent from the supplier. Various methods for rational and fair cost allocation such as line rule, AMEF, the Shapley value, nucleolus, and proportional method were researched and applied to the inventory transportation problem. The core is the basic condition for rational and fair cost allocation. Therefore, we examined if the solutions of various allocation methods are existed in the core through an example.

In this study, we utilize the cross and partial correlation analyses in order to investigate the dependence of power energy consumption on the temperature. To this end, we use a time series data that consists of three attributes : an hourly measured electric power consumption, temperature, and humidity. We, in particular, divide the yearly data into monthly base, and estimate the cross correlation coefficients between all possible pairs of attributes for each monthly based data. We found that temperature and power consumption are negatively correlated in the winter; positively correlated in the summer. A similar trend was found between humid and power consumption. This implies that when temperature or humidity is relatively high or low, the power consumption increases due to the cooling and heating system at work. In contrast, the correlation between temperature and humid behaves differently from those between temperature and power consumption. These results can be used to effectively manage the power system.

In this paper, we propose an Elman recurrent neural network to predict and analyze a time series of gas energy consumption in an air handling unit. To this end, we consider the volatility of the time series and demonstrate that there exists a correlation in the time series of the volatilities, which suggests that the gas consumption time series contain a non-negligible amount of the non-linear correlation. Based on this finding, we adopt the Elman recurrent neural network as the model for the prediction of the gas consumption. As the simplest form of the recurrent network, the Elman network is designed to learn sequential or time-varying pattern and could predict learned series of values. The Elman network has a layer of “context units” in addition to a standard feedforward network. By adjusting two parameters in the model and performing the cross validation, we demonstrated that the proposed model predicts the gas consumption with the relative errors and the average errors in the range of 2%~5% and 3kWh~8kWh, respectively. The results of this study can be used to the energy management system in terms of the effective control of the cross usage of the electric and the gas energies.

Cost allocation studies on the rational allocation method for the common cost of the joint products or services that provide different benefits to each economic entity under the constraints of the efficiency and fairness. Cooperative game theory is often used for cost allocation and studies on a fair and efficient allocation of the utility if some feasible utility for a whole or subset of the players in a game is given. This study shows a variety of cooperative game theory approaches and discusses the pros and cons of each approach.

If the coalition of players orders the product together, then they can reduce the inventory costs such as ordering cost and holding cost. Inventory game can be defined as the allocation of the inventory costs to the players in a fair and rational manner. The characteristics recommended for the solutions provided by the method for the inventory game are completeness, rationality, and marginality. The solutions that satisfy these characteristics are in the core, where the proportional method may depend on the allocator. This study has found out that the solutions of the proportional method with some allocators for economic order quantity model exist in the core.

In this paper, we propose an Elman recurrent neural network to predict and analyze a time series of power energy consumption. To this end, we consider the volatility of the time series and apply the sample variance and the detrended fluctuation analyses to the volatilities. We demonstrate that there exists a correlation in the time series of the volatilities, which suggests that the power consumption time series contain a non-negligible amount of the non-linear correlation. Based on this finding, we adopt the Elman recurrent neural network as the model for the prediction of the power consumption. As the simplest form of the recurrent network, the Elman network is designed to learn sequential or time-varying pattern and could predict learned series of values. The Elman network has a layer of “context units” in addition to a standard feedforward network. By adjusting two parameters in the model and performing the cross validation, we demonstrated that the proposed model predicts the power consumption with the relative errors and the average errors in the range of 2%~5% and 3kWh~8kWh, respectively. To further confirm the experimental results, we performed two types of the cross validations designed for the time series data. We also support the validity of the model by analyzing the multi-step forecasting. We found that the prediction errors tend to be saturated although they increase as the prediction time step increases. The results of this study can be used to the energy management system in terms of the effective control of the cross usage of the electric and the gas energies.

In this study, we proposed a method for correcting non-compliance of the web standard in webpages based on the characteristics of sampled webpages. We collected about 70,000 webpages by using a web crawler and analyzed them by applying various statistical methods. We found that the top three most frequent types of errors and warnings consist of about 60% and 80% of the total number of errors and warnings, respectively. Based on these results, we focus on correcting four most frequent errors and warnings, and showed that the strategy corrected about 95% of these errors and warnings. In addition, we found that different types of errors and warnings are correlated in such a way that correcting one type of error or warning influences on the correction of another types.