X3-Miner: Mining patterns from XML database
dc.contributor.author | Chang, Elizabeth | |
dc.contributor.author | Tan, H. | |
dc.contributor.author | Dillon, Tharam S. | |
dc.contributor.author | Feng, L. | |
dc.contributor.author | Hadzic, Fedja | |
dc.date.accessioned | 2017-01-30T15:26:17Z | |
dc.date.available | 2017-01-30T15:26:17Z | |
dc.date.created | 2008-11-12T23:21:42Z | |
dc.date.issued | 2005 | |
dc.identifier.citation | Chang, Elizabeth and Tan, Henry and Dillon, Tharam and Feng, Ling and Hadzic, Fedja. 2005. : X3-Miner: Mining patterns from XML database, in Zanasi, A. and Brebbia, C.A. and Ebecken, N.F.F. (ed), 6th International Conference on Data Mining, Text Mining and their Business Applications, May 25 2005, pp. 287-298. Skiathos, Greece: WIT Press. | |
dc.identifier.uri | http://hdl.handle.net/20.500.11937/46300 | |
dc.description.abstract |
An XML enabled framework for representation of association rules in databases was first presented in [Feng03]. In Frequent Structure Mining (FSM), there are techniques proposed to mine frequent patterns from complex trees and graphs databases. One of the popular approaches is to use graph matching. Graph matching algorithms use data structures such as the adjacency matrix [Inokuchi00] or adjacency list [FSG01]. Another approach represents semi-structured tree-like structures using a string representation, which is more space efficient and relatively easy for manipulation [Zaki02]. However, in the XML Era, mining association rules is faced with more challenges due to the inherent flexibilities of XML in both structure and semantics. The primary challenges include 1) a more complicated hierarchical data structure with tags and attributes; 2) an ordered data context; and 3) a much bigger data size. To tackle these challenges, in this paper, we propose an approach, X3-Miner, that efficiently extracts patterns from a large XML data set, and overcomes the challenges by:(1) Exploring the use of a model validating approach in deducing the number of candidates generated. The basic idea is that by taking into account of the semantics embedded in the tree-like structure in an XML database while generating candidates directly from the XML tree, we can obtain only valid (i.e., possibly existing) candidates out of the XML database;(2) Minimising I/O overhead by first trimming the infrequent 1-itemset in the XML database. The XML database is intersected with the frequent 1-itemset resulting in a smaller XML tree that contains only the frequent 1-itemset. The algorithm also progressively trims infrequent k-itemsets that contain infrequent (k-1)-itemsets.(3) Extending the notion of string representation of a tree structure proposed in [Zaki02] to xstring for describing an XML document in a flat format without loss of both structure and semantics. Such an extension enables an easier traversal of the tree-structured XML data during our model-validating candidate generation.Our experiments with both synthetic and real-life data sets demonstrate the effectiveness of the proposed model-validating approach in mining XML data. | |
dc.publisher | WIT Press | |
dc.relation.uri | http://www.witpress.com | |
dc.relation.uri | http://library.witpress.com/pages/listpapers.asp?q_bid=327&q_subject=Computing%20_%20Information%20Management | |
dc.subject | X3-Miner | |
dc.subject | Data Mining | |
dc.subject | Semantic Relationships | |
dc.subject | Association Mining | |
dc.subject | information systems | |
dc.subject | Algorithm | |
dc.subject | database mining | |
dc.subject | XML | |
dc.title | X3-Miner: Mining patterns from XML database | |
dc.type | Conference Paper | |
dcterms.source.startPage | 287 | |
dcterms.source.endPage | 298 | |
dcterms.source.title | Data Mining VI: Data mining, text mining and their business applications | |
dcterms.source.series | Data Mining VI: Data mining, text mining and their business applications | |
dcterms.source.conference | 6th International Conference on Data Mining, Text Mining and their Business Applications | |
dcterms.source.conference-start-date | May 25 2005 | |
dcterms.source.conferencelocation | Skiathos, Greece | |
dcterms.source.place | Southampton, Boston | |
curtin.note |
Originally published by WIT Press, Southampton, UK. | |
curtin.department | Centre for Extended Enterprises and Business Intelligence | |
curtin.identifier | EPR-561 | |
curtin.accessStatus | Open access | |
curtin.faculty | Curtin Business School | |
curtin.faculty | School of Information Systems |