Dissertation > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer applications > Information processing (information processing) > Text Processing

The literature of resources for research and application clustering system

Author CuiHongYang
Tutor LiuQingTang
School Central China Normal University
Course Educational Technology
Keywords Text Clustering Singular Value Decomposition Ontology Educational Technology Knowledge Fusion
CLC TP391.1
Type Master's thesis
Year 2011
Downloads 77
Quotes 0
Download Dissertation

With the explosive growth of text messaging , text clustering text information processing technology has become an important means of research , and knowledge discovery, information retrieval , bioinformatics and other fields has been widely used. Text clustering is the use of unsupervised machine learning methods automatically recognize text Category facilitate the user to select useful for knowledge category , and are conducive to knowledge, and similar texts related knowledge for further integration of knowledge provides the premise . Thesis example of educational technology field of ontology library was constructed as a text clustering system data sources to achieve a discipline -oriented literature clustering systems, and Lingo clustering algorithm has been optimized to get a better clustering results . The main work includes : ( 1 ) analysis expounded the theory of text clustering . Main narrative of the text clustering technology research status, introduced the main clustering algorithms and clustering system is currently more mature . ( 2 ) describes the subject domain ontology library construction method . In this paper the concept of domain ontology library contains tables and relational tables to collect educational technology core materials and recent scientific journals in the field of technical terms form the set of concepts , and to indicate the relationship between the concept ( including synonymy , hyponymy , part and whole relationship ) . ( 3 ) paper designed for clustering academic literature resources system , the system is largely based on text preprocessing module , text clustering algorithm module , clustering results visualization module three parts were designed and implemented , and finally through experiments with conventional poly class algorithm are compared . ( 4 ) introduced the document resource clustering results in an information retrieval and knowledge fusion . This feature of the Department : ( 1 ) introduces the field of educational technology construction method of ontology . ( 2 ) on the Lingo clustering algorithm is optimized in the analysis of algorithms based on the concept of relationship ontology repository merge synonyms for word frequency - document matrix to reduce the dimensions of the label used to extract keywords field to punish the label more specification. ( 3 ) for the same category of high similarity of documents, automatically find the same or similar knowledge element , the knowledge -based topic map element integration , so as to achieve the purpose of integration of knowledge between documents .

Related Dissertations
More Dissertations