• Journal of KIISE:Databases
• /
• v.31 no.2
• /
• pp.91-98
• /
• 2004

DHP mining association rules algorithm maintains previously independent direct hash table to reduce the sire of hash tree containing the frequency number of each candidate large itemset. It performs pruning by using the direct hash table when the hash tree is constructed. The mort large the size of direct hash table increases, the higher the effort of pruning becomes. Especially, the effect of pruning in phase 2 which generate 2-large itemsets is so high that it dominates the overall performance of DHP algorithm. So, following the speedy trends of producing VLM(Very Large Memory) systems, extreme increment of direct hash table size is being tried and one of those trials is perfect hash table in phase 2. In case of using perfect hash table in phase 2, we found that some rearrangement of DHP algorithm got about 20% performance improvement compared to simply ｜H$_2$｜ reconfigured DHP algorithm. In this paper, we examine the feasibility of perfect hash table in phase 2 and propose PHP algorithm, a rearranged DHP algorithm, which uses the characteristics of perfect hash table sufficiently, then make an analysis on the results in experimental environment.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2008.05a
• /
• pp.193-196
• /
• 2008

네트워크 인프라가 급속히 발전하면서 네트워크 상에서 발생되는 트래픽을 관리하기 위해 마이닝 기법을 적용하려는 여러 연구가 활발히 진행되고 있다. 그러나 기존의 방법들은 DBMS를 이용하여 개개의 플로우를 저장 후 분석하는 방식을 채택함으로써 엄청난 부하와 실시간 마이닝을 어렵게 하는 문제점이 있다. 본 논문에서는 제한된 크기의 메모리를 사용하여 실시간으로 발생하는 네트워크 플로우 데이터 중 빈발한 플로우를 추출하는 방법을 제안한다. 오직 빈발하게 발생하는 플오우만을 메모리에서 모니터링 트리를 사용하여 관리함으로써 메모리를 효율적으로 사용한다. 제안 된 방법은 기존의 방법들과 비교할 때 적은 시스템 부하를 주면서 초고대역폭의 트래픽을 실시간으로 모니터링 할 수 있다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2007.11a
• /
• pp.471-474
• /
• 2007

본 논문은 연관규칙탐사 기술에서 사용되는 빈발항목집합과 동일한 개념으로 문서분류의 문서에서 빈발단어집합을 정의하고, 이를 사용하여 문서분류 방법으로 잘 알려진 kNN에 적용하였다. 이를 위하여 하나의 문서는 여러 개의 문단으로 나뉘어졌으며, 각 문단에 나타나는 단어들의 집합을 트랜잭션화하여 빈발단어집합을 찾을 수 있도록 하였다. 제안한 방법은 AI::Categorizer 프레임워크에서 구현되었으며 로이터-21578 데이터를 사용하여 학습문서의 크기에 따라 그 정확도가 측정되었다. 정확도의 측정된 결과로 부터 제안된 방법이 기존의 방법에 비하여 정확도를 개선한다는 사실을 알 수 있었다.

• Journal of Internet Computing and Services
• /
• v.7 no.3
• /
• pp.169-178
• /
• 2006

In this paper, we define the large word sets which are noble variations the large item sets in mining association rules, and improve the accuracy for NaiveBayes based on the defined large word sets. In order to use them, a document is divided into the several paragraphs, and then each paragraph can be transformed as the transaction by extracting words in it. The proposed method was implemented by using Al:Categorizer framework and its accuracies were measured by the experiments for reuter-21578 data set. The results of the experiments show that the proposed method improves the accuracy of the conventional NaiveBayes.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2003.11c
• /
• pp.1359-1362
• /
• 2003

최근 인터넷과 웹 기술의 발전 그리고 이를 기반으로 하는 다양한 멀티미디어 컨텐츠가 홍수를 이루고 있지만 멀티미디어 데이터에서 체계적으로 연관 규칙을 마이닝 하는 연구는 초기 단계이다. 본 논문에서는 이미지 프로세싱 분야 및 내용 기반 이미지 검색에 대한 기존 연구를 바탕으로 이미지 데이터 저장소에 저장된 재생성 항목과 희소하게 발생하지만 상대적으로 특정 항목과 높은 비율로 동시에 나타나는 희소 항목을 포함한 내용기반의 이미지 연관 규칙을 찾아내기 위한 탐사 기법을 제안한다 실험 결과 제안된 알고리즘은 기존의 재생성 항목만을 고려한 알고리즘보다 희소 항목을 포함하여 연관 규칙을 탐사하므로 같은 종류의 이미지가 모여 있는 저장소에서 이미지 오브젝트간의 연관 관계를 발견하는 이미지 데이터 마이닝에 효과적이다.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• v.10 no.1
• /
• pp.107-119
• /
• 2006

Recently with developments of an internet and web techniques, the amount of data that are stored in database is increasing rapidly. So the range of adaption in database has been expanded and a research of Data Mining techniques finding useful skills from the huge database has been progressed. Many original algorithms have been developed by cutting down the item set and the size of database isn't required in the entire course of creating frequent item sets. Although those skills could save time in some course, it requires too much time for adapting those techniques in other courses. In this paper, an algorithm is proposed. In an Transaction Database that the length of it's transactions are short or the number of items are relatively small, this algorithm scans a database once by using a Hashing Technique and at the same time, stores all parts of the set, can be appeared at each transaction, in an Hash-table. So without an influence of n minimum percentage of support, it can discover a set of frequent items in more shorter time than the time what is used by an original algorithm.

• Journal of the Korea Society of Computer and Information
• /
• v.16 no.2
• /
• pp.33-46
• /
• 2011

Since an e-mail has been an important mean of communication and information sharing, there have been much effort to classify e-mails efficiently by their contents. An e-mail has various forms in length and style, and words used in an e-mail are usually irregular. In addition, the criteria of an e-mail classification are subjective. As a result, it is quite difficult for the conventional text classification technique to be adapted to an e-mail classification efficiently. An e-mail classification technique in a commercial e-mail program uses a simple text filtering technique in an e-mail client. In the previous studies on automatic classification of an e-mail, the Naive Bayesian technique based on the probability has been used to improve the classification accuracy, and most of them are on an e-mail in English. This paper proposes the personalized recommendation technique of an email in Korean using a data mining technique of frequent patterns. The proposed technique consists of two phases such as the pre-processing of e-mails in an e-mail folder and the generating a profile for the e-mail folder. The generated profile is used for an e-mail to be classified into the most appropriate e-mail folder by the subjective criteria. The e-mail classification system is also implemented, which adapts the proposed technique.

• The KIPS Transactions:PartD
• /
• v.13D no.6 s.109
• /
• pp.765-774
• /
• 2006

An association rule discovery, a technique to analyze the stored data in databases to discover potential information, has been a popular topic in stream data system. Most of the previous researches are concerned to single stream data. However, this approach may ignore in mining to multidimensional stream data. In this paper, we study the techniques discovering the association rules to multidimensional stream data. And we propose a AR-MS method reflecting the characteristics of stream data since make the summarization information by one data scan and discovering the association rules for significant rare data that appear infrequently in the database but are highly associated with specific event. Also, AR-MS method can discover the maximal frequent item of multidimensional stream data by using the summarization information. Through analysis and experiments, we show that AR-MS method is superior to other previous methods.

• Journal of KIISE:Databases
• /
• v.35 no.3
• /
• pp.199-207
• /
• 2008

The FP-tree(Frequency Pattern Tree) mining association rules algorithm was proposed to improve mining performance by reducing DB scan overhead dramatically, and it is recognized that the performance of it is better than that of any other algorithms based on different approaches. But the FP-tree algorithm needs a few more memory because it has to store all transactions including frequent itemsets of the DB. This paper implements a FP-tree algorithm on a general purpose UNK system and compares it with the DHP(Direct Hashing and Pruning) algorithm which uses hash tree and direct hash table from the point of memory usage and execution time. The results show surprisingly that the FP-tree algorithm is poor than the DHP algorithm in some cases even if the system memory is sufficient for the FP-tree. The characteristics of the test data are as follows. The site of DB is look, the number of total items is $1K{\sim}7K$, avenrage length of transactions is $5{\sim}10$, avergage size of maximal frequent itemsets is $2{\sim}12$(these are typical attributes of data for large-scale convenience stores).

• Proceedings of the Korean Information Science Society Conference
• /
• 2002.10d
• /
• pp.337-339
• /
• 2002

연관규칙 탐사기법은 거래(사건) 속에 포함된 품목(항목)간의 연관관계를 발견하고자 할 때 사용하는 기법이며, 독특한 형태의 자료구조를 사용하는 다양한 연관규칙 알고리즘들이 제안되었다. 다양한 특성을 갖는 대용량의 데이터에 대해 효율적으로 연관규칙 탐사를 수행하기 위해서는 저장공간과 실행시간을 모두 고려해야 한다. 본 논문에서는 후보항목집합 발생과정 없이 압축빈발항목집합과 동적링크집합을 이용하여 저장공간 축소와 실행시간 개선을 동시에 고려한 연관규칙 알고리즘을 제안하며, 그 우수성을 증명하기 위해 연관규칙 탐사의 대표적인 자료 구조인 FP-struct, H-Struct와의 저장공간 비교 및 이들 저장구조를 사용하는 FP-growth, H-mine 알고리즘과의 실행시간을 비교한다.