Authors:

Luo SiCarnegie Mellon University, Pittsburgh, PA

Jamie CallanCarnegie Mellon University, Pittsburgh, PA

Atlanta, Georgia, USA — October 05 - 10, 2001
ACM New York, NY, USA ©2001

数据不公开:  educational Web pages ,A total of 91 Web pages。Pages were grouped into three readability levels: KindergartenGrade2, Grade3-Grade5, and Grade6-Grade8

monosyllable 单音节词

2. READABILITY METRICS

第一个是个初级中级学习者

第二个会比别的给的难度分更高

第三个用的更广

3. STATISTICAL LANGUAGE MODELS

线性模型广泛用于模型的组合,EM算法用来寻找最佳参数

线性插值公式来组合语言模型和句子长度模型:前者用ngram,后者考虑句长

1)unigram语言模型假设生成一个词的概率适合上下文无关的。虽然unigram模型在人类语言上效果不好,但是它们适合很多应用,有可以在小数据上训练的优点。

2)通过看某个特征的值是否和难度成正比或反比,来判断特征重要与否,最后得出句长特征很重要,公式法中单音节不适合该数据集;然后假设符合正态分布

4 实验

KF这种公式法只能得出最终属于哪个等级,但是我们的数据集并不含有这些等级。我们统计的方法可以给出概率这种soft metric。

-------------------------

N-Gram是基于一个假设:
第n个词出现与前n-1个词相关,而与其他任何词不相关。(这也是隐马尔可夫当中的假设。)整个句子出现的概率就等于各个词出现的概率乘积。各个词的概率可以通过语料中统计计算得到。假设句子T是有词序列w1,w2,w3...wn组成,用公式表示N-Gram语言模型如下:

P(T)=P(w1)*p(w2)*p(w3)...p(wn)=p(w1)*p(w2|w1)*p(w3|w1w2)...p(wn|w1w2w3...)
一般常用的N-Gram模型是Bi-Gram和Tri-Gram。分别用公式表示如下:
Bi-Gram:P(T)=p(w1|begin)*p(w2|w1)*p(w3|w2)...p(wn|wn-1)
Tri-Gram:P(T)=p(w1|begin1,begin2)*p(w2|w1,begin1)*p(w3|w2w1)...p(wn|wn-1,wn-2)

https://github.com/lijingpeng/kaggle/blob/master/competitions/Bag_of_Words/bags_of_words.ipynb 包含贝叶斯、回归分类

A Statistical Model for Scientific Readability-paper的更多相关文章

  1. machine learning model(algorithm model) .vs. statistical model

    https://www.analyticsvidhya.com/blog/2015/07/difference-machine-learning-statistical-modeling/ http: ...

  2. Writing your first academic paper

    Writing your first academic paper If you are working in academics (and you are if you are working wi ...

  3. The Model Complexity Myth

    The Model Complexity Myth (or, Yes You Can Fit Models With More Parameters Than Data Points) An oft- ...

  4. A Statistical View of Deep Learning (II): Auto-encoders and Free Energy

    A Statistical View of Deep Learning (II): Auto-encoders and Free Energy With the success of discrimi ...

  5. [转]NLP Tasks

    Natural Language Processing Tasks and Selected References I've been working on several natural langu ...

  6. Targeted Learning R Packages for Causal Inference and Machine Learning(转)

    Targeted learning methods build machine-learning-based estimators of parameters defined as features ...

  7. 【RNN】资源汇总

    wesome Recurrent Neural Networks A curated list of resources dedicated to recurrent neural networks ...

  8. Lessons Learned from Developing a Data Product

    Lessons Learned from Developing a Data Product For an assignment I was asked to develop a visual ‘da ...

  9. CVPR 2015 papers

    CVPR2015 Papers震撼来袭! CVPR 2015的文章可以下载了,如果链接无法下载,可以在Google上通过搜索paper名字下载(友情提示:可以使用filetype:pdf命令). Go ...

随机推荐

  1. 网站基础html javascript jquery

    第二章HTML HBuilder的使用 边改边看模式 chrome浏览器看. HTML的基本格式 超文本标记语言 HyperText Markup Language HyperText 超文本 Mar ...

  2. G711 G723 G729线路占多少带宽问题

    G.711   G.711   也称为PCM(脉冲编码调制),是国际电信联盟订定出来的一套语音压缩标准,主要用于电话.它主要用脉冲编码调制对音频采样,采样率为8k每秒.它利用一个 64Kbps 未压缩 ...

  3. 如何设置openwrt在编译linux内核时不优化内核?

    答:修改openwrwt源码目录下rule.mk文件中的HOST_CFLAGS即可 将HOST_CFLAGS:=-O2 $(HOST_CPPFLAGS)改成: HOST_CFLAGS:=-O1 $(H ...

  4. linux基础之find

    linux上文件查找工具: locate, find locate: 依赖于事先构建的索引,索引的构建在系统较为空闲时自动进行(周期性任务),手动更新数据库(updatedb) 索引构建过程需要遍历整 ...

  5. jQuery学习笔记(一)

    jQuery 事件 - ready() 方法 实例 在文档加载后激活函数(): $(document).ready(function(){ $(".btn1").click(fun ...

  6. Lintcode105 Copy List with Random Pointer solution 题解

    [题目描述] A linked list is given such that each node contains an additional random pointer which could ...

  7. [Spring MVC] 表单提交日期转换问题,比如可能导致封装实体类时400错误

    三种格式的InitBinder @InitBinder//https://stackoverflow.com/questions/20616319/the-request-sent-by-the-cl ...

  8. Probability和Likelihood的区别

    Bayes for Beginners: Probability and Likelihood 好好看,非常有用. 以前死活都不理解Probability和Likelihood的区别,为什么这两个东西 ...

  9. C# 获取CPU序列号、网卡MAC地址、硬盘序列号封装类,用于软件绑定电脑

    using System.Management; namespace GLaLa { /// <summary> /// hardware_mac 的摘要说明. /// </summ ...

  10. 7.9 GRASP原则九: 隔离变化

    GRASP原则九: 隔离变化  Protected Variations  需求一定会变化的!如何做到以系统的局部变化为代价就可以应对这一点?4.1 GRASP rule9: Protected ...