Feuding Families and Former Friends: Unsupervised Learning for Dynamic Fictional Relationships-Naacl 2016-20160422

1、Information

publication：-Naacl 2016

2、What

根据小说中的人物描述，a)在每个时间段给出，人物关系的描述的概率分布，b)从时间轴上看出关系的变化轨迹,提出模型Relationship modeling network.(RMN)

3、Dataset

Project Gutenberg

4、How

input: 时间片段内的小说片段

output: RMN模型中的参数：关系描述的矩阵，关系描述的概率分布

目标函数：小说描述的embeding 尽可能与关系描述表示的向量接近。

5、Evaluation:baseline:目前没有工作做过，故本论文与topic model 作比较。时间轴上关系变化的评估，与HTMM(hidden topic Markov model )比较

a) 在评价关系描述的学习中，加入干扰项，model peecision.

b）在人物关系轨迹评估中：人工评价 RMN学习出的关系变化与HTMM学习出的哪个更好

6、Conclusion

近一年文章中Embeding 的方法用的非常多。

时间： 2024-10-13 21:31:45

Feuding Families and Former Friends: Unsupervised Learning for Dynamic Fictional Relationships-Naacl 2016-20160422的相关文章

Unsupervised Learning: Use Cases

Unsupervised Learning: Use Cases Contents Visualization K-Means Clustering Transfer Learning K-Nearest Neighbors The features learned by deep neural networks can be used for the purposes of classification, clustering and regression. Neural nets are s

PredNet --- Deep Predictive coding networks for video prediction and unsupervised learning --- 论文笔记

PredNet --- Deep Predictive coding networks for video prediction and unsupervised learning ICLR 2017 2017.03.12 Code and video examples can be found at: https://coxlab.github.io/prednet/ 摘要:基于监督训练的深度学习技术取得了非常大的成功,但是无监督问题仍然是一个未能解决的一大难题(从未标注的数据中学习到

Unsupervised Learning and Text Mining of Emotion Terms Using R

Unsupervised learning refers to data science approaches that involve learning without a prior knowledge about the classification of sample data. In Wikipedia, unsupervised learning has been described as "the task of inferring a function to describe h

Machine Learning Algorithms Study Notes(4)—无监督学习（unsupervised learning）

1 Unsupervised Learning 1.1 k-means clustering algorithm 1.1.1 算法思想 1.1.2 k-means的不足之处 1.1.3 如何选择K值 1.1.4 Spark MLlib 实现 k-means 算法 1.2 Mixture of Gaussians and the EM algorithm 1.3 The EM Algorithm 1.4 Principal Components

Coursera机器学习-第八周-Unsupervised Learning(K-Means)

Clustering K-means算法是硬聚类算法,是典型的基于原型的目标函数聚类方法的代表,它是数据点到原型的某种距离作为优化的目标函数,利用函数求极值的方法得到迭代运算的调整规则.K-means算法以欧式距离作为相似度测度,它是求对应某一初始聚类中心向量V最优分类,使得评价指标J最小.算法采用误差平方和准则函数作为聚类准则函数. Unsuperivised Learning:Intruduction 典型的Supervised Learning 有一组附标记(y(i))的训练数据集, 我们

Unsupervised learning, attention, and other mysteries

Unsupervised learning, attention, and other mysteries Get notified when our free report “Future of Machine Intelligence: Perspectives from Leading Practitioners” is available for download. The following interview is one of many that will be included

Stanford机器学习课程笔记(1) Supervised Learning and Unsupervised Learning

最近跟完了Andrew Ng的Machine Learning前三周的课,主要讲解了机器学习中的线性回归(Linear Regression)和逻辑回归(Logistic Regression)模型.在这里做一下记录. 另外推荐一本统计学习的书,<统计学习方法>李航,书短小精悍,才200多页,但是内容基本上覆盖了机器学习中的理论基础. 笔记<1> 主要了解一下监督学习和无监督学习机器学习:是关于计算机基于数据构建概率统计模型并运用模型对数据进行预测与分析的一门学科. 机器学习

Unsupervised Learning: Linear Dimension Reduction---无监督学习：线性降维

一 Unsupervised Learning 把Unsupervised Learning分为两大类: 化繁为简:有很多种input,进行抽象化处理,只有input没有output 无中生有:随机给一个input,自动画一张图,只有output没有input 二 Clustering 有一大堆image ,把他们分为几大类,给他们贴上标签,将不同的image用相同的 cluster表示. 也面临一个问题,要有多少种cluster呢? 有两种clustering的方法: 2.1 K-means(

131.007 Unsupervised Learning - Feature Selection | 非监督学习 - 特征选择

1 Why? Reason1 Knowledge Discovery (about human beings limitaitons) Reason2 Cause of Dimensionality (维度灾难) (about ML algorithm itself) 所需的数据量会根据你所拥有的特征数量以指数速度增长 2 NP-Hard Problem arbitrarily choose m features from n features (m≤n),don't know what m t