• ETRI Journal
• /
• v.30 no.3
• /
• pp.421-431
• /
• 2008

In this paper, we offer a new technique to discover frequent spatiotemporal patterns from a moving object database. Though the search space for spatiotemporal knowledge is extremely challenging, imposing spatial and timing constraints on moving sequences makes the computation feasible. The proposed technique includes two algorithms, AllMOP and MaxMOP, to find all frequent patterns and maximal patterns, respectively. In addition, to support the service provider in sending information to a user in a push-driven manner, we propose a rule-based location prediction technique to predict the future location of the user. The idea is to employ the algorithm AllMOP to discover the frequent movement patterns in the user's historical movements, from which frequent movement rules are generated. These rules are then used to estimate the future location of the user. The performance is assessed with respect to precision and recall. The proposed techniques could be quite efficiently applied in a location-based service (LBS) system in which diverse types of data are integrated to support a variety of LBSs.

• KIPS Transactions on Computer and Communication Systems
• /
• v.2 no.6
• /
• pp.279-290
• /
• 2013

In digital forensics log files have been stored as a form of large data for the purpose of tracing users' past behaviors. It is difficult for investigators to manually analysis the large log data without clues. In this paper, we propose a text mining technique for extracting intrusion logs from a large log set to recommend reliable evidences to investigators. In the training stage, the proposed method extracts intrusion association words from a training log set by using Apriori algorithm after preprocessing and the probability of intrusion for association words are computed by combining support and confidence. Robinson's method of computing confidences for filtering spam mails is applied to extracting intrusion logs in the proposed method. As the results, the association word knowledge base is constructed by including the weights of the probability of intrusion for association words to improve the accuracy. In the test stage, the probability of intrusion logs and the probability of normal logs in a test log set are computed by Fisher's inverse chi-square classification algorithm based on the association word knowledge base respectively and intrusion logs are extracted from combining the results. Then, the intrusion logs are recommended to investigators. The proposed method uses a training method of clearly analyzing the meaning of data from an unstructured large log data. As the results, it complements the problem of reduction in accuracy caused by data ambiguity. In addition, the proposed method recommends intrusion logs by using Fisher's inverse chi-square classification algorithm. So, it reduces the rate of false positive(FP) and decreases in laborious effort to extract evidences manually.

• Journal of KIISE:Databases
• /
• v.28 no.3
• /
• pp.279-300
• /
• 2001

Data mining plays an important role in knowledge discovery process and usually various existing algorithms are selected for the specific purpose of the mining. Currently, data mining techniques are actively to the statistics, business, electronic commerce, biology, and medical area and currently numerous algorithms are being researched and developed for these applications. However, in a long run, only a few algorithms, which are well-suited to specific applications with excellent performance in large database, will survive. So it is reasonable to focus our effort on those selected algorithms in the future. This paper classifies about 30 existing algorithms into 7 categories - association rule, clustering, neural network, decision tree, genetic algorithm, memory-based reasoning, and bayesian network. First of all, this work analyzes systematic hierarchy and characteristics of algorithms and we present 14 criteria for classifying the algorithms and the results based on this criteria. Finally, we propose the best algorithms among some comparable algorithms with different features and performances. The result of this paper can be used as a guideline for data mining researches as well as field applications of data mining.

• Journal of the Society of Naval Architects of Korea
• /
• v.42 no.5 s.143
• /
• pp.534-541
• /
• 2005

Until now, Korean shipyards have accumulated a great amount of data. But they do not have appropriate tools to utilize the data in practical works. Engineering data contains experts' experience and know-how in its own. It is very useful to extract knowledge or information from the accumulated existing data by using data mining technique This paper treats an evolutionary computation based on genetic programming (GP), which can be one of the components to realize data mining. The paper deals with linear models of GP for the regression or approximation problem when given learning samples are not sufficient. The linear model, which is a function of unknown parameters, is built through extracting all possible base functions from the standard GP tree by utilizing the symbolic processing algorithm. In addition to a standard linear model consisting of mathematic functions, one variant form of a linear model, which can be built using low order Taylor series and can be converted into the standard form of a polynomial, is considered in this paper. The suggested model can be utilized as a designing tool to predict design parameters with small accumulated data.

• Knowledge Management Research
• /
• v.20 no.1
• /
• pp.45-75
• /
• 2019

Recently, the growth of beauty industry such as plastic surgery and beauty is continued every year in Korea. With the increased interest in appearance based on the improvement of life standard and the development of media, people's perception of cosmetic plastic surgery is changing. Now, as the service for consumer satisfaction based on their desire, the perception of plastic surgery medical service is changed to the high value-added industry with the high growth potential. Thus, this study aims to suggest the strategies for providing the medical service that could satisfy customers, by drawing the factors cognized as important when customers aim to get the cosmetic plastic surgery, and then additionally analyzing the relationships of those factors. On top of performing the topic modeling based on customers' comments data of social commerce related to cosmetic plastic surgery, this study also conducted the network analysis for visualizing the relations of each keywords. The drawn main factors were divided by applying the sub-categories of the SERVQUAL theory, and the additional characteristics of plastic surgery were shown by referring the relevant previous researches. Moreover, the interview with the cosmetic plastic surgery specialists (plastic surgeons) and customers who actually received the plastic surgery, helped the understanding of the interpretation of each factor and the actual relevant phenomenons. The significance of this study is to draw and discuss the main factors that should be observed by Korean cosmetic plastic surgery medical institutes, by mining and analyzing the opinions of customers interested in the cosmetic plastic surgery and procedure with the use of topic modeling. In other words, the quality of medical service of cosmetic plastic surgery could be improved by presenting the key factors that could be considered by the cosmetic plastic surgery medical service suppliers and also the actual strategies.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2005.10a
• /
• pp.409-413
• /
• 2005

The logical analysis of data(LAD) is an effective Boolean-logic based data mining tool. A critical step in analyzing data by LAD is the pattern generation stage where useful knowledge and hidden structural information in data is discovered in the form of patterns. A conventional method for pattern generation in LAD is based on term enumeration that renders the generation of higher degree patterns practically impossible. In this paper, we present a new optimization-based pattern generation methodology and propose two mathematical programming medels, a mixed 0-1 integer and linear programming(MILP) formulation and a well-studied set covering problem(SCP) formulation for the generation of optimal and heuristic patterns, respectively. With benchmark datasets, we demonstrate the effectiveness of our models by automatically generating with much ease patterns of high complexity that cannot be generated with the conventional approach.

• The KIPS Transactions:PartD
• /
• v.16D no.2
• /
• pp.185-190
• /
• 2009

The event sequence of the same type from a set of events having time property can be summarized in one event. But if the event sequence having an interval, It is reasonable to be summarized more than one in independent sub event sequence of each other. In this paper, we suggest a method of temporal data mining that summarizes the interval events based on Allen's interval algebra and finds out interval event association rule from interval events. It provides better knowledge than others by using concept of an independent sub sequence and finding interval event association rules.

• Journal of Intelligence and Information Systems
• /
• v.5 no.2
• /
• pp.15-27
• /
• 1999

산업계 전반에 걸친 오랜 정보시스템 운용의 결과로 대용량의 데이터들이 축적되고 있다. 이러한 데이터로부터 유용한 지식을 추출하기 위해 여러 가지 데이터마이닝 기법들이 연구되어 왔다. 특히 데이터웨어하우스의 등장은 이러한 데이터마이닝에 있어 필요한 데이터 제공 환경을 주고 있다. 그러나 전문가의 적절한 판단과 해석을 거치지 않은 데이터마이닝의 결과는 당연한 사실이거나, 사실과 다른 가짜이거나 또한 관련성 없는(Trivial, Spurious and Irrelevant) 내용만 무수히 쏟아낼 수 있다. 그러므로 데이터마이닝의 결과가 비록 통계적 유의성을 가진다 하더라고 그 정당성과 유용성에 대한 검증과정과 방법론의 정립이 필요하다. 데이터마이닝의 가장 어려운 점은 귀납적 오류를 없애기 위해 사람이 직접 그 결과를 해석하고 판단하며 아울러 새로운 탐색 방향을 제시해야 한다는 것이다. 본 논문의 목적인 이러한 데이터마이닝에서 추출된 결과를 검증하고 아울러 새로운 지식 탐색 방향을 제시하는 방법론을 정립하는데 있다. 본 논문에서는 데이터마이닝 기법 중 연관규칙탐사(Associations)로 얻어진 결과를 설명가능성 여부의 판단을 통해 검증하는 기법을 제안하였고, 이를 위해 도메인 지식(Domain Knowledge)과 연관규칙탐사를 통해 얻어진 결과를 표현하기 위한 지식표현방법으로 관계형 술어논리(RPL : Relational Predicate Logic)를 개발하였다. 연관규칙탐사로 얻어진 결과를 설명하기 위한 방법으로는 연관규칙탐사로 얻어진 연관규칙에 대한 RPL로 표현된 도메인 지식으로서 설명됨을 보이게 한다. 또한 이러한 설명(Explanation)을 토대로 검증된 지식을 일반화하여 새로운 가설을 연역적으로 생성하고 이를 연관규칙탐사를 통해 검증한 후 새로운 지식을 얻는 설명기반 데이터마이닝 구조(Explanation-based Data Mining Architecture)를 제시하였다.

• The Journal of the Korea Contents Association
• /
• v.12 no.12
• /
• pp.35-45
• /
• 2012

Researchers and scientists spend huge amount of time in analyzing the previous studies and their results. In order to timely take the advantageous position, they usually analyze various resources such as paper, patents, and Web documents on recent research issues to preoccupy newly emerging technologies. However, it is difficult to select invest-worthy research fields out of huge corpus by using the traditional information search based on keywords and bibliographic information. In this paper, we propose a method for efficient creation, storage, and utilization of semantically relevant information among technologies, products and research agents extracted from 'big data' by using text mining. In order to implement the proposed method, we designed an ontology that creates technological knowledge for semantic web environment based on the relationships extracted by text mining techniques. The ontology was utilized for InSciTe Adaptive, a R&D trends analysis and forecast service which supports the search for the relevant technological knowledge.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.1
• /
• pp.77-92
• /
• 2020

When automatic speech recognition (ASR) is provided as a cloud service, it is easy to collect voice and application domain data from users. Harnessing these data will facilitate the provision of more personalized services. In this paper, we demonstrate our transfer learning-based knowledge service that built with the user-generated data collected through our novel system that deliveries personalized ASR service. First, we discuss the motivation, challenges, and prospects of building up such a knowledge-based service-oriented system. Second, we present a Quadruple Transfer Learning (QTL) method that can learn a classification model from a source domain and transfer it to a target domain. Third, we provide an overview architecture of our novel system that collects voice data from mobile users, labels the data via crowdsourcing, utilises these collected user-generated data to train different machine learning models, and delivers the personalised real-time cloud services. Finally, we use the E-Book data collected from our system to train classification models and apply them in the smart TV domain, and the experimental results show that our QTL method is effective in two classification tasks, which confirms that the knowledge transfer provides a value-added service for the upper-layer mobile applications in different domains.