Dissertation > Excellent graduate degree dissertation topics show

Research on Machine Learning-based Protein-Protein Interaction Extraction

Author: YuHuanHuan
Tutor: ZhouGuoDong
School: Suzhou University
Course: Applied Computer Technology
Keywords: Text Mining Protein relation extraction Machine Learning
CLC: Q51-3
Type: Master's thesis
Year: 2010
Downloads: 86
Quote: 1
Read: Download Dissertation


With the explosive growth in the field of biomedical literature , automatic acquisition of biomedical knowledge from the biomedical literature has become an important area of bioinformatics research . Since protein interactions relationship has a special significance for the life sciences , and thus protein interaction relation extraction has become hot issues bioinformatics . Due to the complexity and diversity of the biomedical literature , extract protein interactions is a rather difficult task . In this paper, machine learning methods for protein interaction relations extraction in-depth research , the study include: 1 . Research -based the eigenvectors protein relation extraction method , focus on exploring how a variety of surface features extracted from the free text and structural characteristics, and to analyze the contribution of these different characteristics of protein relationship extraction ; proposed relationship extraction method based on the the convolution tree nuclear function of the protein , expression analysis of the structured information on protein relationship extraction impact for to lay a good foundation for further research ; studied protein relation extraction method based on composite kernel function , linear the composite and polynomial composite way based the eigenvectors method and tree - based kernel function method organic combine to effectively capture the relationship instance plane characteristics and structural features . On AIMed Corpus protein relation extraction experiments show that the composite kernel function method based on the eigenvectors and convolution tree kernel function to obtain the highest value of F to 53.7 , reaching the current good level . Of this study and the results obtained on the future relationship of proteins has certain reference value .

Related Dissertations

  1. Based on Data Distribution Characteristics of Text Classification,TP391.1
  2. Prediction of Binding Affinity of Human Transporter Associated with Antigen Processing,R392.1
  3. The Study of Topic-Oriented IT News with Search Enging and Web Page Analysing,TP393.092
  4. Research and Implement of Chinese Word Segment Techniques Based on the Conditional Random Field,TP391.1
  5. Semi-supervised Learning and Active Learning of Sentiment Classification Coupled with Domain Knowledge,TP181
  6. Distortion effects on image quality evaluation and classification,TP391.41
  7. Social network algorithm and its application in recommendation,TP391.3
  8. Based on self-learning social relation extraction research,TP391.1
  9. Graph model based on statistical and machine learning algorithm and its application of certain,TP181
  10. P2P Traffic Identification Method,TP393.06
  11. Based on rough sets and SVM national defense Comprehensive Quality Assessment Methods,E075
  12. SVM Based on SIFT and scene classification,TP391.41
  13. Brain Connectivity Pattern Analysis Based on DTI,R445.2
  14. Text -oriented disciplines correlation analysis association rule mining technology research,TP311.13
  15. The Dynamic Distributed network intrusion patterns,TP393.08
  16. Research on Network Traffic Classification Based on Machine Learning Method,TP393.06
  17. Heterogeneous Information Based Financial Event Detection,TP391.1
  18. Commercial Information Mining from Competitors’ Websites,TP311.13
  19. Medical Image Analysis Based on Feature Extraction and Machine Learning,TP391.41
  20. Performance Prediction Methodology Based on Machine Learning,TP181

CLC: > Biological Sciences > Biochemistry > Protein
© 2012 www.DissertationTopic.Net  Mobile