Analysis of Reply-Tweets for Buzz Tweet Detection
Analysis of Reply-Tweets for Buzz Tweet Detection
复制标题
DOI：
--
复制
发表时间：
2019
期刊：
影响因子：
--
通讯作者：
Kazuyuki Matsumoto;Yuta Hada;Minoru Yoshida;K. Kita
中科院分区：
文献类型：
--
作者：
Kazuyuki Matsumoto;Yuta Hada;Minoru Yoshida;K. Kita
在这项研究中，我们提出了一种方法，通过检查其他人发布的推文回复来预测一条推文是否会在互联网上引起轰动。我们还通过分析特征量来研究回复蜂鸣推文的区别特征。我们提出的方法fi-rst使用单词分布式表示或其他向量化方法将每条回复推文转换为向量表达。然后，我们应用二进制分类fi阳离子的机器学习方法来确定回复是对嗡嗡声的推文还是非嗡嗡声的推文。我们通过比较Classi-fier产生的“嗡嗡”和“非嗡嗡”的总分，将目标推文分为“嗡嗡”和“非嗡嗡”两类。使用StarSpace的方法获得了93.1%的F1得分，而使用转发次数和支持次数(“赞”)的方法获得了77.8%的F1得分。我们还发现，有一些词是嗡嗡作响的推文回复的特征，也有一些词是非嗡嗡作响的推文回复的特征。
In this study, we propose a method for predicting whether a tweet will create a buzz on the Internet by examining tweeted replies posted by others. We also investigate the distinguishing characteristics of replies to buzz tweets by analyzing feature amounts. Our proposed method first converts each reply tweet into a vector expression using a word distributed representation or some other vectorization method. We then apply a machine learning method for binary classification to determine whether the reply is to a buzz tweet or a non-buzz tweet. We classify the target tweet into “buzz” or “non-buzz” categories by comparing the total “buzz” and “non-buzz” scores produced by the classi-fier. The proposed method using StarSpace achieved 93.1% F1-score, while an approach that used number of retweets and number of favors (“likes”) achieved 77.8% F1-score. We also found that there are a number of words that are characteristic of buzz tweet replies and a number of words that are characteristic of non-buzz tweet replies.