《Modeling Transferable Topics for Cross-Target Stance Detection》笔记 Modeling Transferable Topics for Cross-Target Stance Detection 论文:https://dl.acm.org/doi/abs/10.1145/3331184.3331367 INTRODUCTION 定向的立场检测方法主要在目标内进行,即使用同一目标的数据对模型进行训练和测试。当我们面对一个新的目标时,我们已经标记了与现有目标有关的数据 2021-04-21 论文笔记 Stance Detection Topic Model
《A Convolutional Approach for Misinformation Identification》笔记 A Convolutional Approach for Misinformation Identification 论文:http://114.215.172.152/research/Classification/A%20convolutional%20approach%20for%20misinformation%20identification.pdf Introduction GRU- 2021-03-30 论文笔记 Rumour Detection
《Detecting Rumors from Microblogs with Recurrent Neural Networks》笔记 Detecting Rumors with Recurrent Neural Networks 论文:https://ink.library.smu.edu.sg/sis_research/4630/ Introduction 社会心理学文献将谣言定义为一个真实性未经证实或故意造假的故事或陈述。现有的谣言检测模型使用了学习算法结合了从帖子的内容、用户特征和扩散模式中得到的特征,或者简单地利用正 2021-03-29 论文笔记 Rumour Detection
《All-in-one : Multi-task Learning for Rumour Verification》笔记 Rumour Verification 论文:https://arxiv.org/abs/1806.03713 Introduction 社交媒体作为关注事件和突发新闻的平台越来越受用户欢迎。然而,并不是所有在社交媒体上传播的信息都是准确的,不准确的信息会对社会造成严重的危害。科学界对开发验证社交媒体信息的工具越来越感兴趣,Facebook也投入了大量的努力来减轻错误信息造成的问题。 谣言分辨 2021-03-25 论文笔记 Rumour Verification Multi-task
《Learning Reporting Dynamics during Breaking News for Rumour Detection in Social Media》笔记 Rumour Detection And PHEME Dataset 论文:https://arxiv.org/abs/1610.07363 PHEME 数据集:https://figshare.com/articles/PHEME_dataset_of_rumours_and_nonrumours/4010619 Introduction 近年来,利用社交媒体来关注新闻已经变得很普遍。像Twi 2021-03-23 论文笔记 Rumour Detection
《Adaptive Learned Bloom Filter (Ada-BF) Efficient Utilization of the Classifier with Application to Real-Time Information Filtering on the Web》笔记 Adaptive Learned Bloom Filter 论文:https://par.nsf.gov/servlets/purl/10217248 Learned Bloom filter Learned Bloom filter (LBF) 在Bloom filter之前增加了机器学习模型作为预过滤器,对于每个查询的元素$x$,给出得分$s(x)$,$s(x)$通常与$x\in S$的概 2021-03-08 论文笔记 Bloom Filter
《Network Applications of Bloom Filters:A Survey》笔记 Bloom Filter 论文:https://www.tandfonline.com/doi/abs/10.1080/15427951.2004.10129096 Bloom filter是一种空间效率高的随机数据结构,采用位数组表示一个集合以提供成员查询。Bllom filter是容许错误判断的,即把不属于这个集合的元素误判断属于这一集合,这样的错误称作false positives。因此B 2021-03-06 论文笔记 Bloom Filter