Dissertation > Excellent graduate degree dissertation topics show
Research on Frequent Subtree Mining and Pruning Strategies
Author: DengDongJie
Tutor: MaZhiXin
School: Lanzhou University
Course: Computer Software and Theory
Keywords: Frequent subtree mining Frequent sub- tree mode Candidate patterns. Support calculation Pruning strategy
CLC: TP311.13
Type: Master's thesis
Year: 2010
Downloads: 21
Quote: 1
Read: Download Dissertation
Abstract
|
Frequent subtree mining is an important data mining problems , and its main task of mining frequent subtrees from large databases . Due to its wide range of application areas , has been involved in the XML database , molecular databases , relational databases , graphics, databases and other applications and research areas , frequent subtree mining has caused strong concern for database researchers and practitioners . Most existing frequent subtree mining algorithms based on Apriori nature , the main candidate mode generation and support of two major steps. (K 1) -order candidate mode are connected by frequent k -order mode generated in candidate mode generation phase . The major bottleneck of these algorithms is that generates a lot of candidate modes , apparently do support for such a large number of candidate patterns calculation takes consideration is very expensive . In fact , a large number of sub-tree candidate pattern is not frequent or in the database would not have arisen . This paper analyzes the existing frequent subtree mining pruning strategy used in the algorithm , on this basis, we propose a new the the pruning strategy F2SC ( Frequent 2 - Subtree Checking ) . The strategy frequent order subtree stored in a hash table when generating the candidate patterns of order k sub-tree (k ≥ 3) , can be saved by detecting goes F2 hash tables infrequent candidate mode prune . F2SC can be used for all frequent subtree mining algorithm based on Apriori property . In order to verify the validity of the measurement pruning strategy F2SC , we will apply it in classic subtree algorithm TreeMiner of proposed improved algorithm TMp . Finally, by the comparison of TreeMiner and TMP Evaluation the effectiveness of F2SC . Experimental results show good improvement , TreeMiner the the pruning strategy F2SC on proposed effective to reduce the candidate pattern generation, and reduce the cost of support .
|
Related Dissertations
- Research of Detection Outlier Based on Outlier Degree,TP311.13
- The Studying of Sequence Pattern Mining Based on DF2Ls,TP311.13
- Study on Algorithms for Fast Outlier Detection,TP311.13
- Integrated management approach based on XML document clustering,TP311.13
- The Design of Foundation Pit Supporting and Its Application in Zhumadian District,TU753
- Study on Frequent Subtree Mining and Its Application in XML Mining,TP311.13
- A Research on a New Algorithm for Mining Frequent Subtrees,TP311.13
- Single fulcrum pile-anchor retaining structure prestressed lock Factors Affecting,TU473.1
- Research on the Heuristic-search-based AI Planning,TP311.52
- Study on Frequent Pattern Mining Algorithms and Pruning Strategies,TP311.13
- Research on F&B Index Structure Supporting XML Query,TP311.13
- Query Processing and Optimization in Massive Multi-Database Integration,TP311.13
- Research on Compression, Operation and Query Processing Methods of Massive Datasets,TP311.13
- Research and Implementation of Mining Implicit User Interest,TP311.13
- Implementation of Data Compression, Operation and Query Processing System Based on BAP,TP311.13
- Research and Improvement on K-Means Clustering Algorithm,TP311.13
- Study of Data Reduction Technique Based on Manifold Learning,TP311.13
- Public Security 110 Command Decisions Business Systems,TP311.13
- Clustering Method Research Based on Divided and Conquered Method,TP311.13
- Incomplete information on the completeness of the system and its knowledge acquisition,TP311.13
- Evolutionary Clustering Algorithm and Its Application,TP311.13
CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer software > Program design,software engineering > Programming > Database theory and systems
© 2012 www.DissertationTopic.Net Mobile
|