• IE interfaces
• /
• v.25 no.3
• /
• pp.290-299
• /
• 2012

Keystroke dynamics refers to a way of typing a string of characters. Since one has his/her own typing behavior, one's keystroke dynamics can be used as a distinctive biometric feature for user authentication. In this paper, two authentication algorithms based on keystroke dynamics of long and free texts are proposed. The first is the K-S score, which is based on the Kolmogorov-Smirnov test, and the second is the 'R-A' measure, which combines 'R' and 'A' measures proposed by Gunetti and Picardi (2005). In order to verify the authentication performance of the proposed algorithms, we collected more than 3,000 key latencies from 34 subjects in Korean and 35 subjects in English. Compared with three benchmark algorithms, we found that the K-S score was outstanding when the reference and test key latencies were not sufficient, while the 'R-A' measure was the best when enough reference and test key latencies were provided.

• Industrial Engineering and Management Systems
• /
• v.12 no.3
• /
• pp.198-206
• /
• 2013

The team orienteering problem (TOP) or the multiple tour maximum collection problem can be considered as a generic model that can be applied to a number of challenging applications in logistics, tourism, and other fields. This problem is generally defined as the problem of determining P paths, in which the traveling time of each path is limited by $T_{max}$ that maximizes the total collected score. In the TOP, a set of N vertices i is given, each with a score $S_i$. The starting point (vertex 1) and the end point (vertex N) of all paths are fixed. The time $t_{ij}$ needed to travel from vertex i to j is known for all vertices. Some exact and heuristics approaches had been proposed in the past for solving the TOP. This paper proposes a new solution methodology for solving the TOP using the particle swarm optimization, especially by proposing a solution representation and its decoding method. The performance of the proposed algorithm is then evaluated using several benchmark datasets for the TOP. The computational results show that the proposed algorithm using specific settings is capable of finding good solution for the corresponding TOP instance.

• Management Science and Financial Engineering
• /
• v.17 no.1
• /
• pp.79-94
• /
• 2011

Team orienteering, an extension of single-competitor orienteering, is the problem of determining multiple paths from a starting node to a finishing node for a given allowed time or distance limit fixed for each of the paths with the objective of maximizing the total collected score. Each path is through a subset of nodes, each of which has an associated score. The team orienteering problem has many applications such as home fuel delivery, college football players recruiting, service technicians scheduling, military operations, etc. Unlike existing optimal and heuristic algorithms often leading to heavy computation, this paper suggests two types of priority rule based heuristics-serial and parallel ones-that are especially suitable for practically large-sized problems. In the proposed heuristics, all nodes are listed in an order using a priority rule and then the paths are constructed according to this order. To show the performances of the heuristics, computational experiments were done on the small-to-medium sized benchmark instances and randomly generated large sized test instances, and the results show that some of the heuristics give reasonable quality solutions within very short computation time.

• Journal of Computing Science and Engineering
• /
• v.11 no.4
• /
• pp.130-141
• /
• 2017

Feature selection is one of the most challenging problems of pattern recognition and data mining. In this paper, a feature selection algorithm based on an improved version of binary differential evolution is proposed. The method simultaneously optimizes two feature selection criteria, namely, set approximation accuracy of rough set theory and relational algebra based derived score, in order to select the most relevant feature subset from an entire feature set. Superiority of the proposed method over other state-of-the-art methods is confirmed by experimental results, which is conducted over seven publicly available benchmark datasets of different characteristics such as a low number of objects with a high number of features, and a high number of objects with a low number of features.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.35 no.2
• /
• pp.220-227
• /
• 2012

The case company has driven the six sigma innovation programme company wide for the last seven years without any stop in spite of the CEO change. There was neither any benchmark nor the sufficient number of internal experts during the initial stage of driving six sigma. However, the company raised a lot of innovation experts such as MBB, BB and GB, thereby successfully changed the way of employees' doing work and reaped an enormous amount of either visible or invisible performances as a result of implementing the six sigma innovation programme. This paper deals with both main activities undertaken at each stage of the innovation life cycle of six sigma and their performances in a ICT (Information Communication Technology) industry. The performances are described from the four aspects of a balanced score card (BSC) and finally some strategic implications suggested.

• 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.

• Proceedings of the Korea Association of Information Systems Conference
• /
• 2005.12a
• /
• pp.326-333
• /
• 2005

The purpose of this paper is to provide a decision support method to a rational buyer, who wants to pay the least price for the product with the highest quality and service. We suggest a minimum efficiency game model and DEA game model to valuate many vendors whose qualifies of outputs are measured by percentage. Our results gave a decision maker (buyer) the upper bound and lower bound of the true efficiency score of a decision making unit (vendor) with respect to the benchmark (target) set by the buyer. As a result, a buyer can choose the best vendor in terms of his/her preference.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2019.01a
• /
• pp.41-42
• /
• 2019

본 논문에서는 다양한 기계학습 모델을 이용한 신용 데이터 분석 기법에 대해 서술한다. 기계학습 모델은 크게 Canonical models, Committee machines, 그리고 Deep learning models로 분류된다. 이러한 다양한 기계학습 모델 중 일부 학습 모델을 기반으로 Benchmark dataset인 Credit Approval 데이터를 분석하고 성능을 평가한다. 성능 평가에는 k-fold evaluation method를 사용하며, k-fold evaluation 결과에 대한 평균 성능을 측정하기 위해 Accuracy, Precision, Recall, 그리고 F1-score가 사용되었다.

• East Asian mathematical journal
• /
• v.37 no.3
• /
• pp.333-354
• /
• 2021

A support vector machine (SVM) is a state-of-the-art machine learning model rooted in structural risk minimization. SVM is underestimated with regards to its application to real world problems because of the difficulties associated with its use. We aim at showing that the performance of SVM highly depends on which kernel function to use. To achieve these, after providing a summary of support vector machines and kernel function, we constructed experiments with various benchmark datasets to compare the performance of various kernel functions. For evaluating the performance of SVM, the F1-score and its Standard Deviation with 10-cross validation was used. Furthermore, we used taylor diagrams to reveal the difference between kernels. Finally, we provided Python codes for all our experiments to enable re-implementation of the experiments.

• Journal of Digital Convergence
• /
• v.15 no.10
• /
• pp.83-92
• /
• 2017

The purpose of this study is to establish the Balanced Score Card for 'I Company' which is a small and medium sized manufacturing company in Busan City. It is suitable for SMEs and suitable for the management environment. The study was intended to contain the detailed needs of managers and employees when developing the performance measurement system. It also allowed other SMEs to benchmark through this study. We also proposed a solution to the problems after BSC construction. In addition, BSC has been developed for the purpose of shifting business strategy from RDS to SSS in accordance with changes in the demand market environment. Strategy Maps were divided into the whole company level and each team level. You can look at strategic goals, core success factors, and key performance indicators at each glance. Finally, we developed a smart performance evaluation system that can easily calculate the score, strategic goal, key success factor, weight of key performance indicators, target score, performance, and achievement rate by creating a smart chart. Have significance.