Dissertation > Excellent graduate degree dissertation topics show

Research and Design of the Automatic Classification and Active Delivery System for Complaint Texts

Author: DuanXiaoYan
Tutor: YangWenChuan
School: Beijing University of Posts and Telecommunications
Course: Signal and Information Processing
Keywords: Complain Text Classification Model AutomaticClassification Search Engine Long-Polling
CLC: TP391.1
Type: Master's thesis
Year: 2013
Downloads: 17
Quote: 0
Read: Download Dissertation

Abstract


With the continuous development of communication services, a great number of new problems and contradictions have appeared in different way. With a growing number of customer complaints, how to make operators to find the problems, concerned by customers, analytical departments and other business, in the complaints has become an important issue at this stage. How to deal with it effectively has become an inevitable request. While recently there is a kind of incident which is hard to be found by the ordinary way due to the high speed of complaints increase. Those who are blind to this kind of issues will be confronted with the serious consequences.Based on analysis the Complaint data and research the users’ behaviors of screening target complaints, this paper is focus on the automatic classification and long-polling mechanism to solve these issues about above problems. An auxiliary platform of screening target data is provided to the personnel worked on complaint analysis based on the cooperation project as well.The main research work of this paper is briefly generalized as follows:1.According to research the complain texts and the analytical habit of users, we develop the conception and related attributes of automatic classification and long-polling model.2. This paper puts forward an extraction method of text features based on the TF-IDF algorithm. Association rules algorithm is used for mining the feature words of a category from text features of its positive texts.3.A method of constructing complaints space is proposed in this paper, which is based on VSM model and the feature words of every category, and a classification model is presented based on it, which includes classification algorithms and classification parameters as well.4.Relationship mapping is extracted by recording the users’ behaviors of labeling samples. In addition, an automatic delivery model is presented on that basis to deliver the results of automatic classification to the corresponding users.

Related Dissertations

  1. Evaluation on Mining Possibility of Extremely Thin Coal Seam in Heilongjiang Province,TD823.251
  2. Research of Ontology-Aided Method for Integrating Prior Knowledge into Bioinformatics Data Mining,TP311.13
  3. The PLS Variable Selection Method and Its Application,O651
  4. Nuclear power plants based on SVM classification model is not used spare parts,TM623
  5. Intrusion detection technology based on the mix of a variety of detector,TP393.08
  6. Model for Intrusion Detection Based Technology of Knowledge Discovery in Database,TP393.08
  7. Supplier classification model and index system,F274
  8. Based on Winnow algorithm and filtering of spam messages CAPTCHA,TN929.5
  9. Research on the Key Technique of Simple Action Recognition in Video Sequence,TP391.41
  10. Research on the Recognition Methods of Artificial Olfactory Signal,TP212
  11. CdTe nuclear radiation detector output signal noise processing method,TL81
  12. Application Research of ADABOOST Combination Classification Model in Credit Evaluation,F830.5
  13. Research on Mechanical Process Knowledge Management and Its Application,F426.4
  14. The Detection Model of Normal and Abnormal Instances Based-on Process-Mining,TP311.13
  15. A Study and Application on the Clinical Classification Model for Baby Pneumonia Based on Decision Tree Algorithm,TP311.13
  16. Research on Question Classifying of Chinese Question Answering System Based on Bayesian Classification,TP391.1
  17. Model Misfit in Classification and Its Solution,TP181
  18. Research on Classification Model and Standardized Building of Public Building Energy Consumption Quota in Hot and Summer and Warm Winter Area,TU111
  19. The Empirical Bayes Estimations of Parameters for Multi-classification Model,O212.8
  20. Study on Investment Decision-Making of Real Estate Based on Theory of Projection Pursuit,F293.3

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer applications > Information processing (information processing) > Text Processing
© 2012 www.DissertationTopic.Net  Mobile