http://blog.csdn.net/pipisorry/article/details/44040701

LDA (Bleiet al., 2003)

术语

Mustlink states that two words should belong to the same topic
Cannot-link states that two words should not belong to the same topic.

DF-LDA[Andrzejewski, David, Zhu, Xiaojin, and Craven, Mark. Incorporating domain knowledge into topic modeling via Dirichlet Forest priors. In ICML, pp. 2009.]

is perhaps the earliest KBTM, which can incorporate two forms of prior knowledge from the user:must-links and cannot-links.

DF-LDA [1]: A knowledge-based topic model that can use both must-links and cannot-links, but itassumes all the knowledge is correct.

DF-LDA [1] cannot distinguish multiple senses because its de nition of must-link is transitive.That is, if terms w1 and w2 form a must-link, and terms w2 and w3 form a must-link, it implies a must-link between w1 and w3, i.e., w1, w2, and w3 should be in the same topic. We call it thetransitivity problem.

http://blog.csdn.net/pipisorry

GK-LDA (Chen et al., 2013a)

A knowledge-based topic model that uses the ratio of word probabilities under each topic to reduce the effect of wrong knowledge. However, it can only use the must-link type of knowledge.

MC-LDA (Chen et al., EMNLP 2013)

is a recent knowledge-based model for aspect extraction.

A knowledge-based topic model that also use both the must-link and the cannot-link knowledge. It assumes that all knowledge is correct as well.

MC-LDA [10] assumes each must-link represents a distinct sense, and thus assigns each term only one relevant must-link and ignores the rest. This misses a lot of good must-links.

Note:although both DF-LDA and GK-LDA can take prior knowledge from the user, they cannot mine any prior knowledge.

Note:  Both of DF-LDA && MC-LDA   assume the knowledge to be correct and manually provided, both of them assume that the user-provided must-links and cannot-links are correct and there is no conflict among them.

AKL (Automated Knowledge LDA),(Chen et al., 2014: Aspect Extraction with Automated Prior Knowledge Learning):

A knowledge-based topic model that applies clustering to learn the knowledge and utilizes the knowledge in the form of knowledge clusters.

whose inference can exploit the automatically learned prior knowledge and handle the issues of incorrect knowledge to produce superior aspects.

LTM [7]:

A lifelong learning topic model that learns only the must-link type of knowledge automatically. It outperformed [8].

from:http://blog.csdn.net/pipisorry/article/details/44040701

knowledge_based topic model - 基于知识的主题模型概述相关推荐

  1. 【论文笔记8】Tree-Structured Neural Topic Model (分层神经主题模型 / TSNTM)

    Tree-Structured Neural Topic Model 树结构的神经主题模型 论文概述 题目:Tree-Structured Neural Topic Model 作者:Masaru I ...

  2. Incorporating Lexical Priors into Topic Models(即交互式主题模型的应用)论文阅读

    本文作者:合肥工业大学 管理学院 钱洋 email:1563178220@qq.com 内容可能有不到之处,欢迎交流. 未经本人允许禁止转载. 文章目录 论文来源 应用场景及模型 第一个模型 第二个模 ...

  3. 数据仓库之主题模型之IBM与TD金融模型

    数据仓库之主题模型之IBM与TD银行模型 IBM与TD IBM BDWM (Banking Date Warehouse Model) TD FS-LDM (Financal Services Log ...

  4. LDA主题模型评价指标汇总

    主题模型评价指标一览 前言 一.主题距离(Topic distance) 引入与思考 函数与原理 Jaccard distance Kullback-Leibler divergence Hellin ...

  5. 【带你玩转主题模型Topic Model】—— 之 利用sklearn 实现Latetnt Dirichlet Allocation(LDA)主题模型

    写在前面的话 9月份的第一篇blog, 正好是开学季,作为一个学生,hahahha, 我还是正正经经的贡献一篇认为比较干货的文章给大家吧. 我自己是花了很多时间去看懂LDA的算法了,当然了这篇文章不涉 ...

  6. 双稀疏主题模型(Dual-Sparse Topic Model)编程实现中的细节 Java

    本文作者:合肥工业大学 电子商务研究所 钱洋 email:1563178220@qq.com . 内容可能有不到之处,欢迎交流. 未经本人允许禁止转载. 文章目录 论文来源 相关介绍 单稀疏模型 双稀 ...

  7. 主题模型简介(topic model)

    定义 对于一个文档集合来说,假如一篇文章是讲猫科动物的,那么可能会一部分讲猫, 一部分讲老虎, 一部分讲猎豹.那么讲猫的那一部分与猫有关的词语出现的频率应该高一些,比如"鱼",&q ...

  8. 神经主题模型及应用(Neural Topic Model)

    主题模型一般会从一组文档中抽取若干组关键词来表达文档的核心思想,即"主题".首先看看最经典的概率主题模型,LDA模型. Latent Dirichlet Allocation 具体 ...

  9. 2018 A Sparse Topic Model for Extracting Aspect-Specific Summaries from Online Reviews 稀疏主题模型学习笔记

    论文来源 文章介绍 模型及推理 关于源码 论文来源 Rakesh V, Ding W, Ahuja A, et al. A Sparse Topic Model for Extracting Aspe ...

  10. Pseudo-document-based Topic Model(基于伪文档的主题模型)的理解以及源码解读

    本文作者:合肥工业大学 管理学院 钱洋 email:1563178220@qq.com 内容可能有不到之处,欢迎交流. 未经本人允许禁止转载. 论文来源 Zuo Y, Wu J, Zhang H, e ...

最新文章

  1. RGB Color Codes Chart
  2. shareSDK 微信分享的时候只能分享jpg格式的图片大于32k时为什么分享不出去?
  3. hibernate中@Entity和@Table的区别
  4. P1989 无向图三元环计数 思维 + 建图
  5. ARM与x86–蝶变ARM
  6. nuget 下载nupkg 以及依赖包
  7. mt950报文解析_MT700报文升级内容逐项解析
  8. office文档转换为html,OFFICE 文档转换为html在线预览
  9. java提取姓名与手机号码_java – 从手机号码中提取国家/地区代码
  10. 2022年十大突破性技术,“无密码”排名第一
  11. DM 标准单机安装手册
  12. Win10笔记本触摸板在按键时或刚按完键后无法正常使用的问题
  13. java-Stringbuffer
  14. 一点一点解读紫书中例子--大整数类BigInteger (1)
  15. 虚拟机重启网络失败:Error:Failed to start LSB: Bring up/down networking的解决方法
  16. “看得破、忍不过;想得到,做不来​。”
  17. 微信小程序 - image 宽高自适应(图片无法自适应撑开标签)
  18. 关于AD如何按1:1比例保存成PDF的具体做法
  19. 计算机房应急灯标准,GB 51309 - 2018《消防应急照明和疏散指示系统技术标准》规范组官方答疑(四)...
  20. OpenBlas0.3.14编译

热门文章

  1. easyui form表单提交标准格式
  2. 经过多次试验后第一个成功地实现 HTTPService 与 MXML 之间传递数据,ArrayCollection 与DataGrid 之间成功绑定...
  3. nginx下Thinkphp 隐藏index.php
  4. Codeforces 1009G Allowed Letters 最大流转最小割 sosdp
  5. 笔记本外接显示器设置全屏壁纸
  6. cocos2d-html5 碰撞检测的几种方法
  7. System.Data.OracleClient.OracleConnection的类型初始值设定项引发异常.
  8. 灰度变换——反转,对数变换,伽马变换,灰度拉伸,灰度切割,位图切割
  9. iPhone 6/6 Plus 出现后,如何改进工作流以实现一份设计稿支持多个尺寸?
  10. 解决在IE下“JSON”未定义的有关问题