• Journal of Korean Society of Transportation
• /
• v.12 no.1
• /
• pp.25-42
• /
• 1994

The traffic forecast is one of the most important analysis objects in the urban transportation planning process. The results of traffic forecast are the most widely used informations and give a critical influence on the major decision makings in the transportation planning process. Thus, they should be as much accurate and credible data, and evaluated to determine whether they are enough reliable to directly use in the planning process. However, the evaluation process is usually overlooked or abbreviated with a few exceptions according to the size and character of the project. Even though a planner or engineer tries to evaluate the assignment results, he/she is usually faced with certain difficulties since there are no established criteria and methods for the accuracy evaluation. Accordingly, the main purpose of this research placed on establishing the criteria and methods for the accuracy evaluation of the assignment results. The secondary purpose was to evaluate which assignment technique produces the most accurate assignment results by applying the established evaluation criteria and methods to an actual network. The research found that the proposed evaluation methods well operated in testing the accuracy of assignment results with few limits on application. Also, the incremental assignment was found to provide the best assignment results of existing assignment techniques (Stochastic, Iterative, Incremental, Equilibrium assignment) for the Seoul city network applied.

• BMB Reports
• /
• v.44 no.2
• /
• pp.118-122
• /
• 2011

Most widely used secondary structure assignment methods such as DSSP identify structural elements based on N-H and C=O hydrogen bonding patterns from X-ray or NMR-determined coordinates. Secondary structure assignment algorithms using limited $C{\alpha}$ information have been under development as well, but their accuracy is only ~80% compared to DSSP. We have hereby developed SABA (Secondary Structure Assignment Program Based on only Alpha Carbons) with ~90% accuracy. SABA defines a novel geometrical parameter, termed a pseudo center, which is the midpoint of two continuous $C{\alpha}s$. SABA is capable of identifying $\alpha$-helices, $3_{10}$-helices, and $\beta$-strands with high accuracy by using cut-off criteria on distances and dihedral angles between two or more pseudo centers. In addition to assigning secondary structures to $C{\alpha}$-only structures, algorithms using limited $C{\alpha}$ information with high accuracy have the potential to enhance the speed of calculations for high capacity structure comparison.

• Journal of Institute of Control, Robotics and Systems
• /
• v.6 no.10
• /
• pp.845-853
• /
• 2000

In this paper, we introduce a relationship between the sensitivity and the robustness of a system, and we propose a robust eigenstructure assignment scheme using a novel performance index which can consider the performance and the robustness of the system simultaneously. We also propose an assignment accuracy measure and a robustness measure which are used for the performance examination of the proposed robust eigenstructure assignment scheme. The usefulness of the proposed algorithm and the measures are verified by applying to controller design of a simple numerical example and the EMRAAT missile.

• Asian-Australasian Journal of Animal Sciences
• /
• v.18 no.11
• /
• pp.1529-1534
• /
• 2005

Individual-breed assignments were implemented in six swine populations using twenty six microsatellites recommended by the Food and Agriculture Organization and the International Society for Animal Genetics (FAO-ISAG). Most microsatellites exhibited high polymorphisms as shown by the number of alleles and the polymorphism information content. The assignment accuracy per locus obtained by using the Bayesian method ranged from 33.33% (CGA) to 68.47% (S0068), and the accumulated assignment accuracy of the top ten loci combination added up to 96.40%. The assignment power of microsatellites based on the Bayesian method had positive correlations with the number of alleles and the gene differential coefficient ($G_{st}$) per locus, while it has no relationship to genetic heterozygosity, polymorphism information content per locus and the exclusion probabilities under case II and case III. The percentage of corrected assignment was highest for the Bayesian method, followed by the gene frequency and distancebased methods. The assignment efficiency of microsatellites rose with increase in the number of loci used, and it can reach 98% when using a ten-locus combination. This indicated that such a set of ten microsatellites is sufficient for breed verification purposes.

• Proceedings of the KIEE Conference
• /
• 1990.07a
• /
• pp.539-541
• /
• 1990

A new neural network model, based on the Hopfield's crossbar associative network, is presented and shown to be an effective tool for the NP-Complete problems. This model is applied to a class of layer assignment problems for VLSI routing. The results indicate that this modified Hopfield model improves stability and accuracy.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.40 no.2
• /
• pp.234-237
• /
• 1991

A new neural network model, based on the Hopfield crossbar associative network, is presented and shown to be an effective tool for the NP-Complete problems. This model is applied to a class of layer assignment problems for VLSI routing. The results indicate that this modified Hopfield model, improves stability and accuracy.

• The Journal of Asian Finance, Economics and Business
• /
• v.9 no.10
• /
• pp.159-168
• /
• 2022

The time series of foreign trade turnover is complex and variable and contains linear and nonlinear information. This paper proposes preprocessing the dataset by the EMD algorithm and combining the linear prediction advantage of the SARIMA model with the nonlinear prediction advantage of the EMD-LSTM model to construct the SARIMA-EMD-LSTM hybrid model by the weight assignment method. The forecast performance of the single models is compared with that of the hybrid models by using MAPE and RMSE metrics. Furthermore, it is confirmed that the weight assignment approach can benefit from the hybrid models. The results show that the SARIMA model can capture the fluctuation pattern of the time series, but it cannot effectively predict the sudden drop in foreign trade turnover caused by special reasons and has the lowest accuracy in long-term forecasting. The EMD-LSTM model successfully resolves the hysteresis phenomenon and has the highest forecast accuracy of all models, with a MAPE of 7.4304%. Therefore, it can be effectively used to forecast the Sino-Russia foreign trade turnover time series post-epidemic. Hybrid models cannot take advantage of SARIMA linear and LSTM nonlinear forecasting, so weight assignment is not the best method to construct hybrid models.

• Asian-Australasian Journal of Animal Sciences
• /
• v.27 no.5
• /
• pp.616-627
• /
• 2014

The Lanyu is a miniature pig breed indigenous to Lanyu Island, Taiwan. It is distantly related to Asian and European pig breeds. It has been inbred to generate two breeds and crossed with Landrace and Duroc to produce two hybrids for laboratory use. Selecting sets of informative genetic markers to track the genetic qualities of laboratory animals and stud stock is an important function of genetic databases. For more than two decades, Lanyu derived breeds of common ancestry and crossbreeds have been used to examine the effectiveness of genetic marker selection and optimal approaches for individual assignment. In this paper, these pigs and the following breeds: Berkshire, Duroc, Landrace and Yorkshire, Meishan and Taoyuan, TLRI Black Pig No. 1, and Kaohsiung Animal Propagation Station Black pig are studied to build a genetic reference database. Nineteen microsatellite markers (loci) provide information on genetic variation and differentiation among studied breeds. High differentiation index ($F_{ST}$) and Cavalli-Sforza chord distances give genetic differentiation among breeds, including Lanyu's inbred populations. Inbreeding values ($F_{IS}$) show that Lanyu and its derived inbred breeds have significant loss of heterozygosity. Individual assignment testing of 352 animals was done with different numbers of microsatellite markers in this study. The testing assigned 99% of the animals successfully into their correct reference populations based on 9 to 14 markers ranking D-scores, allelic number, expected heterozygosity ($H_E$) or $F_{ST}$, respectively. All miss-assigned individuals came from close lineage Lanyu breeds. To improve individual assignment among close lineage breeds, microsatellite markers selected from Lanyu populations with high polymorphic, heterozygosity, $F_{ST}$ and D-scores were used. Only 6 to 8 markers ranking $H_E$, $F_{ST}$ or allelic number were required to obtain 99% assignment accuracy. This result suggests empirical examination of assignment-error rates is required if discernible levels of co-ancestry exist. In the reference group, optimum assignment accuracy was achievable achieved through a combination of different markers by ranking the heterozygosity, $F_{ST}$ and allelic number of close lineage populations.

• Journal of Korea Multimedia Society
• /
• v.23 no.4
• /
• pp.595-602
• /
• 2020

We propose a generative probabilistic model with Dirichlet prior distribution for topic modeling and text similarity analysis. It assigns a topic and calculates text correlation between documents within a corpus. It also provides posterior probabilities that are assigned to each topic of a document based on the prior distribution in the corpus. We then present a Gibbs sampling algorithm for inference about the posterior distribution and compute text correlation among 50 abstracts from the papers published by IEEE. We also conduct a supervised learning to set a benchmark that justifies the performance of the LDA (Latent Dirichlet Allocation). The experiments show that the accuracy for topic assignment to a certain document is 76% for LDA. The results for supervised learning show the accuracy of 61%, the precision of 93% and the f1-score of 96%. A discussion for experimental results indicates a thorough justification based on probabilities, distributions, evaluation metrics and correlation coefficients with respect to topic assignment.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.41 no.3
• /
• pp.23-36
• /
• 2016

We consider a model that minimizes the total cost incurred by assigning available weapons to existing targets in order to reduce enemy threats, which is called the weapon target assignment problem (WTAP). This study addresses the stochastic versions of WTAP, in which data, such as the probability of destroying a target, are given randomly (i.e., data are identified with certain probability distributions). For each type of random data or parameter, we provide a stochastic optimization model on the basis of the expected value or scenario enumeration. In particular, when the probabilities of destroying targets depending on weapons are stochastic, we present a stochastic programming formulation with a simple recourse. We show that the stochastic model can be transformed into a deterministic equivalent mixed integer programming model under a certain discrete probability distribution of randomness. We solve the stochastic model to obtain an optimal solution via the mixed integer programming model and compare this solution with that of the deterministic model.