Dissertation > Excellent graduate degree dissertation topics show

Research on Mining Community from Emails

Author: ZhangXiaoDong
Tutor: WangZuo
School: National University of Defense Science and Technology
Course: Computer Science and Technology
Keywords: Community mining Target discovery The level of importance Relationship
CLC: TP311.52
Type: Master's thesis
Year: 2007
Downloads: 105
Quote: 4
Read: Download Dissertation

Abstract


Network data, personal electronic information (web, mail and published documents, etc.) can reflect a person involved in the field as well as the degree of concern in this area. E-mail as a form of network data, has become a powerful platform for mutual cooperation and the exchange of knowledge. Throughout the fields of business, social and technical exchanges, and highly relevant to the field of social network research. E-mail is often used for dynamic exchange behavior within the organization, and it is this dynamic fuzzy hierarchical relationships within the organization and structure of relationship. So when there is a great difficulty we are facing collected a large number of e-mail from the e-mail list from the complex connections precisely and quickly found that community and extraction of community collaboration and leadership relations. Targeted e-mail, the e-mail network community mining technology, mining methods and how to judge the level or degree of importance of the members of the collection, including collection of mail users. First of all, we have an important node according to the initial collection step given the expansion of the collection, the other node to identify more closely with the user focus node associated. The traditional community division method is almost impossible for the vast amounts of data processed in the reasonable range when asked to complete, for specific applications from the two aspects of the efficiency and accuracy of community division method were improved. Associated collections community divided in accordance with the improved method constitutes pending evaluation of the community collection. Then, in order to give the collection node ID on important level, the community will eventually have a higher level of value important node is added to the collection of important users, after analysis inspired by Google's PageRank algorithm, developed based communication frequency node level, (ImportanceRank, the following abbreviated IR) algorithm algorithm and theme-based IR. Finally, we carried out the experiments and achieved good results. The results of this study have been used in the prototype system, the same time provides us close to the initial set-associative importance node, and can be observed depending on the sampling data to valuable information. This indicates that described in this article the community mining methods can be more accurate to extract effective community, an important level algorithm provides us with more accurate node level, and provide better guidance to the judgment of the relationship of collaboration and leadership.

Related Dissertations

  1. Environmental ethics based on the concept of circular economy of tourism research,F590
  2. Development of EST-SSR Primers and Application in Analysis of Genetic Realtionships in Tree Peony,S685.11
  3. A Study on the Relationship between the Spirituality and Self-Consistency of College Students,B844.2
  4. The Study of the Relationship between the Party and the Masses from the View of Social Class Divisions,D252
  5. Research on the Construction of Self-class,G424.21
  6. Research on the Relationship between the University and Its Faculty from the Perspective of Contract Theory,G645.1
  7. The Research of Influence Factors of Class Management in Secondary Vocational School,G712.4
  8. The Establishment of Chemical Toxicological Database for Wool Industry,TS131
  9. Study on the Relationship between Chemical Interactions and Structure-Function Properties of Rabbit Myosin,TS251.1
  10. Studies on Genetic Diversity and Phylogenetic Relationship of Dendranthema and Its Related Genera,S682.11
  11. Analysis of Genetic Diversity in Pear (Pyrus) and Cherry (Cerasus) Germplasm Using DNA Molecular Markers,S661.2
  12. The Analysis of Parent-child Relationship Impacting on Youth Internet Addiction and Design of Play Therapy Strategy,D669.5
  13. Design and Implementation of Cross-selling System Based on Knowledge Reasoning,TP18
  14. A Study on Interpersonal Relationship of Primary Teacher Supporting Rural Education from Urban School,G451.1
  15. Study of Nurse-patient Relationship Based on Nurses’ Communicative Abilities,R47
  16. Parent-child relationship , the parent and child sexual awareness of the relationship between social,G78
  17. U.S. community college teacher-student relationship and Inspiration,G648.6
  18. Perspective of the socialist market economy , pharmaceutical sales staff pay satisfaction and job satisfaction Relations,F426.72
  19. Philosophical Thinking of Ecological Consumption,X2
  20. Environmental Ethics and Technology Development Relationship,B82-058
  21. The Research about the System of Prosecution Guiding Investigation,D925.2

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer software > Program design,software engineering > Software Engineering > Software Development
© 2012 www.DissertationTopic.Net  Mobile