• 한국멀티미디어학회논문지
• /
• 제11권12호
• /
• pp.1739-1748
• /
• 2008

When we conduct a data mining procedure on sample data sources, several rules are generated. But some rules are redundant or logically disjoint and therefore they can be removed. We suggest a new rule minimization algorithm inspired from logic synthesis to improve comprehensibility and eliminate redundant rules. The method can merge several relevant rules into one based on data mining and logic synthesis without high loss of accuracy. In case of two or more rules are candidates to be merged, we merge the rules with the attribute having the lowest information gain. To show the proposed method could be a reasonable solution, we applied the proposed approach to a problem domain constructing user preferred ontology in anti-spam systems.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2019년도 춘계학술대회
• /
• pp.325-326
• /
• 2019

본 논문은 텍스트 마이닝 기법으로 명사의 빈도수를 조사하여 워드클라우드를 나타내는 기존의 방법을 개선하여 지능적 워드클라우드를 구현하는 방법을 제안한다. 텍스트 마이닝 시에 명사 단어를 추출하는 사전에 누락된 신조어 등의 단어를 효과적으로 추가하고, 동사 등 다른 품사위주의 워드클라우드를 시각적으로 보여주는 방법을 제안한다. 실험에서 기존 명사의 빈도수 추출에는 KoNLP 패키지를 사용하였고, 지원되지 않는 신조어 80개를 추가하였고 빈도수를 수동으로 조사하여 추가하였다.

• 정보처리학회논문지D
• /
• 제18D권2호
• /
• pp.81-88
• /
• 2011

시퀀스(sequence) 데이터가 주어졌을 때 그 중에서 빈번(frequent)한 순차 패턴을 찾는 순차 패턴 마이닝(sequential pattern mining)은 여러 어플리케이션(application)에 사용되는 중요한 데이터마이닝 문제이다. 순차 패턴 마이닝은 웹 접속 패턴, 고객 구매 패턴, 특정 질병의 DNA 시퀀스를 찾는 등 광범위한 분야에서 사용된다. 본 논문에서는 맵리듀스(MapReduce) 프레임웍 상에서 맵리듀스 함수 호출을 최적화하는 순차 패턴 마이닝 알고리즘을 개발하였다. 이 알고리즘은 여러 대의 기계에 데이터들을 분산시켜 병렬적으로 빈번한 순차 패턴을 찾는다. 실험적으로 다양한 데이터를 이용하여 파라미터 값을 변화시켜가며 제안된 알고리즘의 성능을 종합적으로 확인하였다. 그리고 실험 결과를 통해 제안된 알고리즘은 기계 수에 대해 선형적인 속도 개선을 보인다는 것을 확인하였다.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2006년도 Proceedings of ISRS 2006 PORSEC Volume II
• /
• pp.627-630
• /
• 2006

Sequential Pattern Mining is the mining approach which addresses the problem of discovering the existent maximal frequent sequences in a given databases. In the daily and scientific life, sequential data are available and used everywhere based on their representative forms as text, weather data, satellite data streams, business transactions, telecommunications records, experimental runs, DNA sequences, histories of medical records, etc. Discovering sequential patterns can assist user or scientist on predicting coming activities, interpreting recurring phenomena or extracting similarities. For the sake of that purpose, the core of sequential pattern mining is finding the frequent sequence which is contained frequently in all data sequences. Beside the discovery of frequent itemsets, sequential pattern mining requires the arrangement of those itemsets in sequences and the discovery of which of those are frequent. So before mining sequences, the main task is checking if one sequence is a subsequence of another sequence in the database. In this paper, we implement the subsequence matching method as the preprocessing step for sequential pattern mining. Matched sequences in our implementation are the normalized sequences as the form of number chain. The result which is given by this method is the review of matching information between input mapped sequences.

• 인터넷정보학회논문지
• /
• 제20권6호
• /
• pp.21-28
• /
• 2019

Process mining is state-of-the-art technology in the workflow field. Recently, process mining becomes more important because of the fact that it shows the status of the actual behavior of the workflow model. However, as the process mining get focused and developed, the material of the process mining - workflow event log - also grows fast. Thus, the process mining algorithms cannot operate with some data because it is too large. To solve this problem, there should be a lightweight process mining algorithm, or the event log must be divided and processed partly. In this paper, we suggest a set of operations that control and edit XES based event logs for process mining. They are designed based on relational algebra, which is used in database management systems. We designed three operations for tailoring XES event logs. Select operation is an operation that gets specific attributes and excludes others. Thus, the output file has the same structure and contents of the original file, but each element has only the attributes user selected. Union operation makes two input XES files into one XES file. Two input files must be from the same process. As a result, the contents of the two files are integrated into one file. The final operation is a slice. It divides anXES file into several files by the number of traces. We will show the design methods and details below.

• Journal of Power Electronics
• /
• 제15권3호
• /
• pp.841-848
• /
• 2015

As a vital part of renewable energy and electrical traction, power converters are supposed to have high reliability and good performance. However, power semiconductors produce considerable heat when the power converter works, which results in high junction temperatures that lower the reliability and performance of the power semiconductors. Many studies show that ambient humidity has a significant effect on power devices, but the influence of high humidity on junction temperatures has yet to be studied. Therefore, this paper presents a thermal model for power converters in moist air to obtain the junction temperature increase, which is utilized for the power converter used in a Switched Reluctance Motor System. Simulation results show that the law of converter temperature distribution is independent of the relative humidity in the case of fixed ambient temperature, whereas the temperature in the power converter decreases as the ambient relative humidity increases. These simulation results are validated with the experimental results.

• Journal of Power Electronics
• /
• 제14권2호
• /
• pp.315-323
• /
• 2014

An optimized space vector pulse width modulation (SVPWM) method with common mode voltage elimination and neutral point potential balancing is proposed for an open-end winding induction motor. The motor is fed from both of the ends with two neutral point clamped (NPC) three-level inverters. In order to eliminate the common mode voltage of the motor ends and balance the neutral point potential of the DC link, only zero common mode voltage vectors are used and a balancing control factor is gained from calculation in the strategy. In order to improve the harmonic characteristics of the output voltages and currents, the balancing control factor is regulated properly and the theoretical analysis is provided. Simulation and experimental results show that by adopting the proposed method, the common mode voltage can be completely eliminated, the neutral point potential can be accurately balanced and the harmonic performance for the output voltages and currents can be effectively improved.

• Journal of Power Electronics
• /
• 제12권4호
• /
• pp.615-622
• /
• 2012

The switching frequency of the power electronic devices used in large synchronous motor drives is usually kept low (less than 1 kHz) to reduce the switching losses and to improve the converter power capability. However, this results in a couple of problems, e.g. an increase in the harmonic components of the stator current, and an undesired cross-coupling between the magnetization current component ($i_m$) and the torque component ($i_t$). In this paper, a novel complex matrix model of electrically excited synchronous motors (EESM) was established with a new control scheme for coping with the low switching frequency issues. First, a hybrid observer was proposed to identify the instantaneous fundamental component of the stator current, which results in an obvious reduction of both the total harmonic distortion (THD) and the low order harmonics. Then, a novel complex current controller was designed to realize the decoupling between $i_m$ and $i_t$. Simulation and experimental results verify the effectiveness of this novel control system for EESM drives.

• 정보보호학회논문지
• /
• 제23권1호
• /
• pp.33-43
• /
• 2013

정보의 양이 많아짐에 따라 많은 양의 정보를 효과적으로 관리, 운용할 수 있는 데이터 마이닝 기법의 연구가 활발해졌다. 다양한 데이터 마이닝 기법들이 연구되었는데 그 중에는 프라이버시를 보호할 수 있는 프라이버시 보호 데이터 마이닝(Privacy Preserving Data Mining) 연구도 진행됐다. 프라이버시 보호 데이터 마이닝은 크게 연관규칙, 군집화, 분류 등의 알고리즘이 존재한다. 그 중 연관규칙 알고리즘은 데이터간의 연관규칙을 찾아내는 알고리즘으로 주로 마케팅에 주로 사용된다. 본 논문에서는 Shamir의 비밀 분배 기법을 이용하여 다자간 프라이버시 보호 데이터 마이닝 환경에서 단일 비트가 아닌 멀티 비트 정보를 공유할 수 있는 내적연산 기법을 제안한다.

• Journal of Power Electronics
• /
• 제13권5호
• /
• pp.896-908
• /
• 2013

Under high grid impedance conditions, it is difficult to guarantee the stability of grid-connected inverters with an LCL filter designed based on ideal grid conditions. In this paper, the theoretical basis for output impedance calculation is introduced. Based on the small-signal model, the d-d channel closed-loop output impedance models adopting the converter-side current control method and the grid-side current control method are derived, respectively. Specifically, this paper shows how to simplify the stability analysis which is usually complemented based on the generalized Nyquist stability criterion (GNC). The stability of each current-controlled grid-connected system is analyzed via the proposed simplified method. Moreover, the influence of the LCL parameters on the stability margin of grid-connected inverter controlled with converter-side current is studied. It is shown that the stability of grid-connected systems is fully determined by the d-d channel output admittance of the grid-connected inverter and the inductive component of the grid impedance. Experimental results validate the proposed theoretical stability analysis.