• Journal of the Korea Society of Computer and Information
• /
• v.15 no.9
• /
• pp.35-46
• /
• 2010

Sequential pattern mining is one of the essential data mining tasks, and it is widely used to analyze data generated in various application fields such as web-based applications, E-commerce, bioinformatics, and USN environments. Recently data generated in the application fields has been taking the form of continuous data streams rather than finite stored data sets. Considering the changes in the form of data, many researches have been actively performed to efficiently find sequential patterns over data streams. However, conventional researches focus on reducing processing time and memory usage in mining sequential patterns over a target data stream, so that a research on mining more interesting and useful sequential patterns that efficiently reflect the characteristics of the data stream has been attracting no attention. This paper proposes a mining method of sequential patterns over data streams with a gap constraint, which can help to find more interesting sequential patterns over the data streams. First, meanings of the gap for a sequential pattern and gap-constrained sequential patterns are defined, and subsequently a mining method for finding gap-constrained sequential patterns over a data stream is proposed.

• Journal of Intelligence and Information Systems
• /
• v.15 no.3
• /
• pp.67-80
• /
• 2009

By a number of sensors and sensor networks, we can collect environmental information from a certain sensor space. To discover more useful information and knowledge, we want to employ data mining methodologies to sensor data stream from such sensor spaces. In this paper, we present a novel data preprocessing scheme to improve the performances of the data mining algorithms. Especially, ontologies are applied to represent meanings of the sensor data. For evaluating the proposed method, we have collected sensor streams for about 30 days, and simulated them to compare with other approaches.

• The KIPS Transactions:PartD
• /
• v.15D no.1
• /
• pp.23-30
• /
• 2008

The basic research of context aware in ubiquitous environment is an internet technique and XML. The XML data of continuous stream type are popular in network application through the internet. And also there are researches related to query processing for streaming XML data. As a basic research to efficiently query, we propose not only a labeled ordered tree model representing the XML but also a mining method to extract frequent structures from streaming XML data. That is, XML data to continuously be input are modeled by a stream tree which is called by XFP_tree and we exactly extract the frequent structures from the XFP_tree of current window to mine recent data. The proposed method can be applied to the basis of the query processing and index method for XML stream data.

• The KIPS Transactions:PartD
• /
• v.15D no.6
• /
• pp.741-754
• /
• 2008

Due to the characteristics of a data stream, it is very important to confine the memory usage of a data mining process regardless of the amount of information generated in the data stream. For this purpose, this paper proposes the Prime pattern tree(PPT) for finding frequent itemsets over data streams with using the confined memory space. Unlike a prefix tree, a node of a PPT can maintain the information necessary to estimate the current supports of several itemsets together. The length of items in a prime pattern can be reduced the total number of nodes and controlled by split_delta $S_{\delta}$. The size and the accuracy of the PPT is determined by $S_{\delta}$. The accuracy is better as the value of $S_{\delta}$ is smaller since the value of $S_{\delta}$ is large, many itemsets are estimated their frequencies. So it is important to consider trade-off between the size of a PPT and the accuracy of the mining result. Based on this characteristic, the size and the accuracy of the PPT can be flexibly controlled by merging or splitting nodes in a mining process. For finding all frequent itemsets over the data stream, this paper proposes a PPT to replace the role of a prefix tree in the estDec method which was proposed as a previous work. It is efficient to optimize the memory usage for finding frequent itemsets over a data stream in confined memory space. Finally, the performance of the proposed method is analyzed by a series of experiments to identify its various characteristics.

• The KIPS Transactions:PartD
• /
• v.19D no.2
• /
• pp.139-146
• /
• 2012

Data Mining is widely used to discover knowledge in many fields. Although there are many methods to discover association rule, most of them are based on frequency-based approaches. Therefore it is not appropriate for stream environment. Because the stream environment has a property that event data are generated continuously. it is expensive to store all data. In this paper, we propose a new method to discover association rules based on stream environment. Our new method is using a variable window for extracting data items. Variable windows have variable size according to the gap of same target event. Our method extracts data using COBJ(Count object) calculation method. FPMDSTN(Frequent pattern Mining over Data Stream using Terminal Node) discovers association rules from the extracted data items. Through experiment, our method is more efficient to apply stream environment than conventional methods.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2009.04a
• /
• pp.360-363
• /
• 2009

실시간으로 심장 질환의 진단을 위하여 심전도 신호의 스트림 처리 및 데이터 마이닝 프로토타입을 구현하였다. 신체부착형 센서로부터 전송되는 심전도를 전처리하여 심장질환의 진단 지표를 추출하였고 실시간 진단을 위한 출현 패턴 마이닝 알고리즘을 구현 및 적용하였다. 이를 기반으로 심혈관계 질환에 대하여 실시간 자동 진단 및 예측이 가능한 생체 신호 스트림 데이터 처리 분석 프로토타입을 구현하였다.

• 한국IT서비스학회:학술대회논문집
• /
• 2010.05a
• /
• pp.291-295
• /
• 2010

최근 급격한 하드웨어 기술과 데이터베이스 시스템의 발전은 우리 주변에서 발생하는 다양한 분야의 데이터를 자동으로 수집하는 것을 가능하게 하였다. 흔히 데이터 스트림(data stream)이라고 언급되는 끊임없이 생산되는 대용량의 데이터를 효율적으로 처리하여 유용한 정보를 얻어내는 기술은 이미 많은 응용 분야에서 광범위하게 연구되고 있다. 인터넷은 이러한 데이터 스트림을 양산해 내는 주요 원천 중의 하나이다. 인터넷 비즈니스의 활성화와 더불어 웹로그 데이터 스트림은 마케팅, 전략 수립, 고객관리 등 여러 부분에 광범위하게 활용되기 시작했으며, 보다 정확하고 효율적인 분석에 대한 요구사항도 점점 늘어나고 있다. 데이터 웨어하우스(Data Warehouse)는 수집된 데이터를 주제 기반으로 통합하여 시계열 형태로 적재하는 저장소서 유용한 분석이나 의사결정에 많이 사용되어 왔다. 데이터웨어하우스는 데이터를 요약하고 통합 및 정제하는 기능을 제공하여 대용량의 데이터 처리에 적합하고 데이터의 품질을 향상시키기 때문에 데이터 마이닝 분야에서 전처리 과정으로도 많이 이용되어 왔다. 본 논문에서는 웹로그 데이터 스트림에 대한 데이터 웨어하우스를 구축하여 보다 고품질의 유용한 정보를 효율적으로 얻어내는 시스템을 제안한다.

• The KIPS Transactions:PartD
• /
• v.18D no.4
• /
• pp.215-224
• /
• 2011

U-Health collects patient information with various kinds of sensor. Stream data can be summarized as an interval event which has aninterval between start-time-point and end-time-point. Most of temporal mining techniques consider only the event occurrence-time-point and ignore stream data change of state. In this paper, we propose the temporal mining technique considering the event change of state in U-Health. Our method overcomes the restrictions of the environment by sending a significant event in U-Health from sensors to a server. We define four event states of stream data and perform the temporal data mining considered the event change of state. Finally, we can remove an ambiguity of discovered rules by describing cause-and-effect relations among events in temporal relation sequences.

• Journal of Digital Convergence
• /
• v.20 no.2
• /
• pp.217-223
• /
• 2022

Subspace clustering for online data streams requires a large amount of memory resources as all subsets of data dimensions must be examined. In order to track the continuous change of clusters for a data stream in a finite memory space, in this paper, we propose a grid-based subspace clustering algorithm that effectively uses memory resources. Given an n-dimensional data stream, the distribution information of data items in data space is monitored by a grid-cell list. When the frequency of data items in the grid-cell list of the first level is high and it becomes a unit grid-cell, the grid-cell list of the next level is created as a child node in order to find clusters of all possible subspaces from the grid-cell. In this way, a maximum n-level grid-cell subspace tree is constructed, and a k-dimensional subspace cluster can be found at the k^th level of the subspace grid-cell tree. Through experiments, it was confirmed that the proposed method uses computing resources more efficiently by expanding only the dense space while maintaining the same accuracy as the existing method.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.11b
• /
• pp.145-147
• /
• 2005

최근 들어 저장장치의 발전과 네트워크의 발달로 인하여 대용량의 데이터가 매우 빠르게 증가되고 있다. 또한, 대용량의 데이터에 내재되어 있는 정보를 빠른 시간 내에 처리하여 새로운 지식을 창출하려는 요구가 증가하고 있다. 연속적이고 빠르게 증가하는 데이터를 지칭하는 데이터 스트림에서 데이터 마이닝 기법을 이용하여 시간이 흐름에 따라 변하고, 무한적으로 증가하는 데이터 스트림에서의 빈발항목을 찾는 연구가 활발하게 진행되고 있다. 하지만 기존의 연구들은 시간의 흐름에 따른 빈발항목 탐색방법을 적절히 제시하지 못하고 있으며 단지 집계를 이용하여 빈발항목을 탐색하고 있다. 본 논문에서는 데이터 스트림에서 시간적 측면을 고려하여 상대적인 빈발항목을 탐색하기 위한 새로운 알고리즘을 제안한다. 논문에서 제안하는 알고리즘의 성능은 다양한 실험을 통해서 검증된다.