• Journal of KIISE:Computer Systems and Theory
• /
• v.33 no.1_2
• /
• pp.15-34
• /
• 2006

This paper introduces an empirical modeling technique. This technique uses a set of sample results which are collected from a few small scale simulations. Empirical models are developed by applying a couple of statistical estimation techniques to these samples. We built two types of models for cache miss rates in Symmetric Multiprocessor systems. One is for the changes of input data set size while the specification of target system is fixed. The other is for the changes of the number of processors in target system while the input data set size is fixed. To develop accurate models, we built individual model for every kind of cache misses for each shared data structure in a program. The final model is then obtained by integrating them. Besides, combined use of Least Mean Squares and Robust Estimations enhances the quality of models by minimizing the distortion due to outliers. Empirical modeling technique produces extremely accurate models without analysis on sample data. In addition, since only snail scale simulations are necessary, once a set of samples can be collected, empirical method can be adopted in any research areas. In 17 cases among 24 trials, empirical models present extremely low prediction errors below $1\%$. In the remaining cases, the accuracy is excellent, as well. The models sustain high quality even when the behavioral characteristics of programs are irregular and the number of samples are barely enough.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.20 no.6
• /
• pp.157-165
• /
• 2020

Aiming at the development of neural network and self-driving data set, it is also an idea to improve the performance of network model to detect moving objects by dividing the data set. In Darknet network framework, the YOLOv4 (You Only Look Once v4) network model was used to train and test Udacity data set. According to 7 proportions of the Udacity data set, it was divided into three subsets including training set, validation set and test set. K-means++ algorithm was used to conduct dimensional clustering of object boxes in 7 groups. By adjusting the super parameters of YOLOv4 network for training, Optimal model parameters for 7 groups were obtained respectively. These model parameters were used to detect and compare 7 test sets respectively. The experimental results showed that YOLOv4 can effectively detect the large, medium and small moving objects represented by Truck, Car and Pedestrian in the Udacity data set. When the ratio of training set, validation set and test set is 7:1.5:1.5, the optimal model parameters of the YOLOv4 have highest detection performance. The values show mAP50 reaching 80.89%, mAP75 reaching 47.08%, and the detection speed reaching 10.56 FPS.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.39 no.1
• /
• pp.19-26
• /
• 2015

This paper proposes a design method for improving the frequency-domain performance reliability of dynamic systems with uncertain and degrading components. Discrete frequencies are used in this method as surrogates for the frequency band of interest, and the conformance of the frequency responses to the specification at these frequencies is utilized to model the frequency-domain performance reliability. A meta-model for the frequency responses, an extreme-value event, and the set-theory are integrated to improve the computational efficiency of the reliability estimation. In addition, a sample-based approach is presented to evaluate and optimize the estimated performance reliability. A case study of a vibration absorber system showed that the proposed design method has engineering applications.

• Journal of Digital Convergence
• /
• v.16 no.10
• /
• pp.205-212
• /
• 2018

With the development of high performance CPU / GPU, artificial intelligence algorithms such as deep neural networks, and a large amount of data, machine learning has been extended to various applications. In particular, a large amount of data collected from the Internet of Things, social network services, web pages, and public data is accelerating the use of machine learning. Learning data sets for machine learning exist in various formats according to application fields and data types, and thus it is difficult to effectively process data and apply them to machine learning. Therefore, this paper studied a method for building a learning data set for machine learning in accordance with standardized procedures. This paper first analyzes the requirement of learning data set according to problem types and data types. Based on the analysis, this paper presents the reference model to build learning data set for machine learning applications. This paper presents the target standardization organization and a standard development strategy for building learning data set.

• Journal of the military operations research society of Korea
• /
• v.8 no.1
• /
• pp.53-70
• /
• 1982

This thesis develops a more realistic and applicable new set covering model that is adjusted and supplied by the existing set covering models, and induces an algorithm for solving the new set covering model, and applies the new model and the algorithm to an actual set covering problems. The new set covering model introduces a probabilistic covering aistance ($0{\eqslantless}p{\eqslantless}1$)or time($0{\eqslantless}p{\eqslantless}1$) instead of a deterministic covering distance(0 or 1) or time (0 or 1) of the existing set covering model. The existing set covering model has not considered the merit of the overcover of customers. But this new set covering model leads a concept of this overcover to a concept of the parallel system reliability. The algorithm has been programmed on the UNIVAC 9030 for solving large-scale covering problems. An application of the new set covering model is presented in order to determine the locations of the air surveillance radars as a set covering problem for a case-study.

• Journal of KIISE:Software and Applications
• /
• v.34 no.12
• /
• pp.1045-1055
• /
• 2007

Latent topic models are statistical models which automatically captures salient patterns or correlation among features underlying a data collection in a probabilistic way. They are gaining an increased popularity as an effective tool in the application of automatic semantic feature extraction from text corpus, multimedia data analysis including image data, and bioinformatics. Among the important issues for the effectiveness in the application of latent topic models to the massive data set is the efficient learning of the model. The paper proposes an accelerated learning technique for PLSA model, one of the popular latent topic models, by an incremental EM algorithm instead of conventional EM algorithm. The incremental EM algorithm can be characterized by the employment of a series of partial E-steps that are performed on the corresponding subsets of the entire data collection, unlike in the conventional EM algorithm where one batch E-step is done for the whole data set. By the replacement of a single batch E-M step with a series of partial E-steps and M-steps, the inference result for the previous data subset can be directly reflected to the next inference process, which can enhance the learning speed for the entire data set. The algorithm is advantageous also in that it is guaranteed to converge to a local maximum solution and can be easily implemented just with slight modification of the existing algorithm based on the conventional EM. We present the basic application of the incremental EM algorithm to the learning of PLSA and empirically evaluate the acceleration performance with several possible data partitioning methods for the practical application. The experimental results on a real-world news data set show that the proposed approach can accomplish a meaningful enhancement of the convergence rate in the learning of latent topic model. Additionally, we present an interesting result which supports a possible synergistic effect of the combination of incremental EM algorithm with parallel computing.

• Journal of Internet Computing and Services
• /
• v.1 no.2
• /
• pp.11-18
• /
• 2000

In this paper we present a new extensible model based upon the concept of subtransactions in real-time transaction systems. The nested transaction model originally proposed by J. Moss is extended for real-time uniprocessor transaction systems by adding explicit timing constraints. Based upon the model, an integrated concurrency control and scheduling algorithm is developed, that not only guarantees timing constraints of a set of real-time transactions but also maintains consistency of the database. The algorithm is based on the priority ceiling protocol of Sha et al. We prove that the Real-Time Nested Priority Ceiling Protocol prevents unbounded blocking and deadlock, and maintains the serializability of a set of real-time transactions. We use the upper bound on the duration that a transaction can be blocked to show that it is possible to analyze the schedulability of a transaction set using rate-monotonic priority assignment. This work is viewed as a step toward multiprocessor and distributed real-time nested transaction systems. Also, it is possible to be extended to include the real-time multimedia transactions in the emerging web-based database application areas.

• Journal of Cadastre & Land InformatiX
• /
• v.52 no.1
• /
• pp.81-93
• /
• 2022

Recently, there have been many studies on object recognition around the vehicle and recognition of traffic signs and traffic lights in autonomous driving. In particular, such the recognition of traffic lights is one of the core technologies in autonomous driving. Therefore, many studies for such the recognition of traffic lights have been performed, the studies based on various deep learning models have increased significantly in recent. In addition, as a high-quality AI training data set for voice, vision, and autonomous driving is released on AIHub, it makes it possible to develop a recognition model for traffic lights suitable for the domestic environment using the data set. In this study, we developed a recognition model for traffic lights that can be used in Korea using the AIHub's training data set. In particular, in order to improve the recognition performance, we used various models of YOLOv4 and YOLOv5, and performed our recognition experiments by defining various classes for the training data. In conclusion, we could see that YOLOv5 shows better performance in the recognition than YOLOv4 and could confirm the reason from the architecture comparison of the two models.

• Journal of Korea Multimedia Society
• /
• v.2 no.2
• /
• pp.195-199
• /
• 1999

While object-oriented models are effective in achieving sharing and code reusability, they unfortunately lack a mechanism for giving scope to objects. We propose an object-oriented model in which each object can be given a scope, i.e., an object becomes available only when it is needed. Thus, the set of currently available objects is dynamically changing and only the needed set of objects is maintained in this model. We illustrate the usefulness of this model through some examples.

• Proceedings of the KIEE Conference
• /
• 2006.07d
• /
• pp.2089-2090
• /
• 2006

본 논문은 기호 코딩 및 정보입자를 이용한 유전자 알고리즘의 전방향 퍼지 집합 기반 뉴럴네트워크 (Information Granules and Symbolic Encoding-based Fuzzy Set Polynomial Neural Networks ; IG and SE based FSPNN)의 모델 설계를 제안한다. 기존 퍼지 집합기반 다항식 뉴럴네트워크(FSPNN)의 구조 최적화를 위해 이진코딩을 사용하였다. 그러나 이진코딩에서 스트링의 길이가 길면 길수록 인접한 두 수 사이에 발생하는 급격한 비트 차이라는 해밍절벽이 발생하였다. 이에 제안된 모델에서는 해밍절벽의 문제를 해결하기 위해 기호코딩을 사용하였다. 제안된 모델은 각 입력에 대해 MFs의 개수 만큼 규칙을 생성하는 Fuzzy 집합기반 다항식 뉴럴네트워크(FSPNN)를 그대로 사용한다. 그리고 IG based gFSPNN의 평가을 위해 실험적 예제를 통하여 제안된 모델의 성능 및 근사화 능력의 우수함을 보인다.