#Paper Reading# Joint Matrix Factorization and Manifold-Ranking for Topic-Focused Multi-Document Sum
2017-07-01 18:48
531 查看
论文题目:Joint Matrix Factorization and Manifold-Ranking for Topic-Focused Multi-Document Summarization
论文地址:http://dl.acm.org/citation.cfm?id=2767765
论文发表于:SIGIR 2015(CCF A类) 短文
论文大体内容:
本文将矩阵分解与流形排序(Manifold-ranking)组合起来,得到JMFMR的新模型,用于多文档摘要(extractive式),通过实验发现取得了不错的效果。
1、流形排序(Manifold-ranking)的优化方程如下
![](https://img-blog.csdn.net/20170701184723896?watermark/2/text/aHR0cDovL2Jsb2cuY3Nkbi5uZXQvSm9objE1OTE1MQ==/font/5a6L5L2T/fontsize/400/fill/I0JBQkFCMA==/dissolve/70/gravity/SouthEast)
2、矩阵分解WTMF的优化方程如下
![](https://img-blog.csdn.net/20170701184732934?watermark/2/text/aHR0cDovL2Jsb2cuY3Nkbi5uZXQvSm9objE1OTE1MQ==/font/5a6L5L2T/fontsize/400/fill/I0JBQkFCMA==/dissolve/70/gravity/SouthEast)
3、将矩阵分解与流形排序(Manifold-ranking)组合起来,得到如下优化方程,其中两者的关联,体现在W’与W”矩阵
![](https://img-blog.csdn.net/20170701184742319?watermark/2/text/aHR0cDovL2Jsb2cuY3Nkbi5uZXQvSm9objE1OTE1MQ==/font/5a6L5L2T/fontsize/400/fill/I0JBQkFCMA==/dissolve/70/gravity/SouthEast)
4、将优化方程拆分为两个,不断迭代优化两个方程,从而得到每个句子的打分F向量;
![](https://img-blog.csdn.net/20170701184750474?watermark/2/text/aHR0cDovL2Jsb2cuY3Nkbi5uZXQvSm9objE1OTE1MQ==/font/5a6L5L2T/fontsize/400/fill/I0JBQkFCMA==/dissolve/70/gravity/SouthEast)
![](https://img-blog.csdn.net/20170701184757488?watermark/2/text/aHR0cDovL2Jsb2cuY3Nkbi5uZXQvSm9objE1OTE1MQ==/font/5a6L5L2T/fontsize/400/fill/I0JBQkFCMA==/dissolve/70/gravity/SouthEast)
5、使用论文[1]中的考虑diversity的生成摘要的方法,抽取摘要;
6、在DUC2006的实验结果
![](https://img-blog.csdn.net/20170701184806913?watermark/2/text/aHR0cDovL2Jsb2cuY3Nkbi5uZXQvSm9objE1OTE1MQ==/font/5a6L5L2T/fontsize/400/fill/I0JBQkFCMA==/dissolve/70/gravity/SouthEast)
![](https://img-blog.csdn.net/20170701184814351?watermark/2/text/aHR0cDovL2Jsb2cuY3Nkbi5uZXQvSm9objE1OTE1MQ==/font/5a6L5L2T/fontsize/400/fill/I0JBQkFCMA==/dissolve/70/gravity/SouthEast)
参考资料:
[1]、http://blog.csdn.net/John159151/article/details/74076521
以上均为个人见解,因本人水平有限,如发现有所错漏,敬请指出,谢谢!
论文地址:http://dl.acm.org/citation.cfm?id=2767765
论文发表于:SIGIR 2015(CCF A类) 短文
论文大体内容:
本文将矩阵分解与流形排序(Manifold-ranking)组合起来,得到JMFMR的新模型,用于多文档摘要(extractive式),通过实验发现取得了不错的效果。
1、流形排序(Manifold-ranking)的优化方程如下
2、矩阵分解WTMF的优化方程如下
3、将矩阵分解与流形排序(Manifold-ranking)组合起来,得到如下优化方程,其中两者的关联,体现在W’与W”矩阵
4、将优化方程拆分为两个,不断迭代优化两个方程,从而得到每个句子的打分F向量;
5、使用论文[1]中的考虑diversity的生成摘要的方法,抽取摘要;
6、在DUC2006的实验结果
参考资料:
[1]、http://blog.csdn.net/John159151/article/details/74076521
以上均为个人见解,因本人水平有限,如发现有所错漏,敬请指出,谢谢!
相关文章推荐
- #Paper Reading# Manifold-Ranking Based Topic-Focused Multi-Document Summarization
- #Paper Reading# Lifelong Machine Learning for Topic Modeling and Beyond
- #Paper Reading# SumView: A Web-based engine for summarizing product reviews and customer opinions
- #Paper Reading# Topic Modeling using Topics from Many Domains, Lifelong Learning and Big Data
- #Paper Reading# Robust Word-Network Topic Model for Short Texts
- #Paper Reading# Online Knowledge-Based Model for Big Data Topic Extraction
- #Paper Reading# Fast Online EM for Big Topic Modeling
- End-to-End Answer Chunk Extraction and Ranking for Reading Comprehension
- #Paper Reading# Graph Regularized Non-negative Matrix Factorization By Maximizing Correntropy
- [Paper note] Joint Learning of Single-image and Cross-image Representations for Person Re-id.
- mysql不预读数据库信息(use dbname 更快,不会卡)—Reading table information for completion of table and column names
- 解决:Reading table information for completion of table and column names
- SUSE Linux Toolbox: 1000+ Commands for openSUSE and SUSE, Linux Enterprise (Paperback) Dec.2007.eBook-BBL
- The first paper reading——Guttman, R-TREES A DYNAMIC INDEX STRUCTURE FOR SPATIAL SEARCHING
- [EMNLP2017]Global Normalization of Convolutional Neural Networks for Joint Entity and Relation(阅读笔记)
- 行人检索"Deep Ranking for Person Re-identification via Joint Representation Learning"
- #Paper Reading# Dual Learning for Machine Translation
- 论文阅读:Joint Learning of Single-image and Cross-image Representations for Person Re-identification
- paper reading——《Improving Person Re-identification by Attribute and Identity Learning》
- How to Read and Understand a Scientific Paper: A Step-by-Step Guide for Non-Scientists