Popularity prediction method of Twitter topics based on evolution patterns

doi:10.11772/j.issn.1001-9081.2022010045

Journal of Computer Applications ›› 2022, Vol. 42 ›› Issue (11): 3364-3370.DOI: 10.11772/j.issn.1001-9081.2022010045

• CCF Bigdata 2021 • Previous Articles Next Articles

Popularity prediction method of Twitter topics based on evolution patterns

Weifan XIE¹^,², Yan GUO¹(), Guangsheng KUANG¹^,³, Zhihua YU¹, Yuanhai XUE¹, Huawei SHEN¹

^1.Data Intelligence System Research Center，Institute of Computing Technology，Chinese Academy of Sciences，Beijing 100190，China
^2.School of Computer Science and Technology，University of Chinese Academy of Sciences，Beijing 101408，China
^3.School of Artificial Intelligence，University of Chinese Academy of Sciences，Beijing 101408，China

Received:2022-01-17 Revised:2022-03-03 Accepted:2022-03-07 Online:2022-04-18 Published:2022-11-10
Contact: Yan GUO
About author:XIE Weifan， born in 1997， M. S. candidate. His research interests include popularity prediction.
GUO Yan， born in 1974， Ph. D.， senior engineer. Her research interests include network information acquisition， network content processing.
KUANG Guangsheng， born in 1995， M. S. candidate. His research interests include natural language processing， data fusion.
YU Zhihua， born in 1973， Ph. D.， chief senior engineer. His research interests include internet public opinion analysis.
XUE Yuanhai， born in 1987， Ph. D.， senior engineer. His research interests include information retrieval， big data.
SHEN Huawei， born in 1982， Ph. D.， research fellow. His research interests include social computing， data mining， machine learning.
Supported by:
National Natural Science Foundation of China(U21B2046)

基于演化模式的推特话题流行度预测方法

解伟凡¹^,², 郭岩¹(), 匡广生¹^,³, 余智华¹, 薛源海¹, 沈华伟¹

^1.中国科学院计算技术研究所数据智能系统研究中心, 北京 100190
^2.中国科学院大学计算机科学与技术学院, 北京 101408
^3.中国科学院大学人工智能学院, 北京 101408

通讯作者: 郭岩
作者简介:解伟凡（1997—），男，山西运城人，硕士研究生，主要研究方向：流行度预测
郭岩（1974—），女，陕西西安人，高级工程师，博士，主要研究方向：网络信息获取、网络内容处理 guoy@ict.ac.cn
匡广生（1995—），男，江西赣州人，硕士研究生，主要研究方向：自然语言处理、数据融合
余智华（1973—），男，江西吉安人，正高级工程师，博士，主要研究方向：网络舆情分析
薛源海（1987—），男，云南玉溪人，高级工程师，博士，主要研究方向：信息检索、大数据
沈华伟（1982—），男，河南周口人，研究员，博士，主要研究方向：社会计算、数据挖掘、机器学习。

Abstract

Abstract:

A popularity prediction method of Twitter topics based on evolution patterns was proposed to address the problem that the differences between evolution patterns and the time?effectiveness of prediction were not taken into account in previous popularity prediction methods. Firstly， the K?SC （K?Spectral Centroid） algorithm was used to cluster the popularity sequences of a large number of historical topics， and 6 evolution patterns were obtained. Then， a Fully Connected Network （FCN） was trained as the prediction model by using historical topic data of each evolution pattern. Finally， in order to select the prediction model for the topic to be predicted， Amplitude?Alignment Dynamic Time Warping （AADTW） algorithm was proposed to calculate the similarity between the known popularity sequence of the topic to be predicted and each evolution pattern， and the prediction model of the evolution pattern with the highest similarity was selected to predict the popularity. In the task of predicting the popularity of the next 5 hours based on the known popularity of the first 20 hours， the Mean Absolute Percentage Error （MAPE） of the prediction results of the proposed method was reduced by 58.2% and 31.0% respectively， compared with those of the Auto?Regressive Integrated Moving Average （ARIMA） method and method using a single fully connected network. Experimental results show that the model group based on the evolution patterns can predict the popularity of Twitter topic more accurately than single model.

Key words: Twitter topic, evolution pattern, popularity prediction, social network, time series

摘要：

针对以往流行度预测方法未利用演化模式之间的差异和忽略预测时效性的问题，提出了一种基于演化模式的推特话题流行度预测方法。首先，基于K?SC算法对大量历史话题的流行度序列进行聚类，并得到6类演化模式；然后，使用各类演化模式下的历史话题数据分别训练全连接网络（FCN）作为预测模型；最后，为选择待预测话题的预测模型，提出幅度对齐的动态时间规整（AADTW）算法来计算待预测话题的已知流行度序列与各演化模式的相似度，并选取相似度最高的演化模式的预测模型进行流行度预测。在根据已知前20 h的流行度预测后5 h的流行度的任务中，与差分整合移动平均自回归（ARIMA）方法以及使用单一的全连接网络进行预测的方法相比，所提方法的预测结果的平均绝对百分比误差（MAPE）分别降低了58.2%和31.0%。实验结果表明，基于演化模式得到的模型群相较于单一模型能更加准确地预测推特话题流行度。

关键词: 推特话题, 演化模式, 流行度预测, 社交网络, 时间序列

CLC Number:

TP181

Weifan XIE, Yan GUO, Guangsheng KUANG, Zhihua YU, Yuanhai XUE, Huawei SHEN. Popularity prediction method of Twitter topics based on evolution patterns[J]. Journal of Computer Applications, 2022, 42(11): 3364-3370.

解伟凡, 郭岩, 匡广生, 余智华, 薛源海, 沈华伟. 基于演化模式的推特话题流行度预测方法[J]. 《计算机应用》唯一官方网站, 2022, 42(11): 3364-3370.

Figures/Tables 4

References 19

1	YANG J， LESKOVEC J. Patterns of temporal variation in online media［C］// Proceedings of the 4th ACM International Conference on Web Search and Web Data Mining. New York： ACM， 2011：177-186. 10.1145/1935826.1935863
2	BERNDT D J， CLIFFORD J. Using dynamic time warping to find patterns in time series［C］// Proceedings of the 1994 AAAI Conference on Knowledge Discovery and Data Mining. Menlo Park， CA： AAAI Press， 1994：359-370.
3	SZABO G， HUBERMAN B A. Predicting the popularity of online content［J］. Communication of the ACM， 2010， 53（8）：80-88. 10.1145/1787234.1787254
4	朱海龙，云晓春，韩志帅. 基于传播加速度的微博流行度预测方法［J］. 计算机研究与发展， 2018， 55（6）：1282-1293. 10.7544/issn1000-1239.2018.20161057
	ZHU H L， YUN X C， HAN Z S. Weibo popularity prediction method based on propagation acceleration［J］. Journal of Computer Research and Development， 2018， 55（6）：1282-1293. 10.7544/issn1000-1239.2018.20161057
5	BAO P， SHEN H W， HUANG J M， et al. Popularity prediction in microblogging network： a case study on Sina Weibo［C］// Proceedings of the 22nd International Conference on World Wide Web. New York： ACM， 2013：177-178. 10.1145/2487788.2487877
6	高金华，沈华伟，程学旗，等. 基于相似消息的流行度预测方法［J］. 中文信息学报， 2018， 32（11）：79-85. 10.3969/j.issn.1003-0077.2018.11.011
	GAO J H， SHEN H W， CHENG X Q， et al. Popularity prediction method based on similar historical tweets［J］ Journal of Chinese Information Processing， 2018， 32（11）：79-85. 10.3969/j.issn.1003-0077.2018.11.011
7	WANG X M， FANG B X， ZHANG H L， et al. Predicting the popularity of news based on competitive matrix［C］// Proceedings of the IEEE 2nd International Conference on Data Science in Cyberspace. Piscataway： IEEE， 2017：151-155. 10.1109/dsc.2017.88
8	AHMED M， SPAGNA S， HUICI F， et al. A peek into the future： predicting the evolution of popularity in user generated content［C］// Proceedings of the 6th ACM International Conference on Web Search and Data Mining. New York： ACM， 2013：607-616. 10.1145/2433396.2433473
9	LYMPEROPOULOS I N. RC‑Tweet： modeling and predicting the popularity of tweets through the dynamics of a capacitor［J］. Expert Systems with Applications， 2021， 163： No.113785. 10.1016/j.eswa.2020.113785
10	TSAGKIAS M， WEERKAMP W， DE RIJKE M. Predicting the volume of comments on online news stories［C］// Proceedings of the 18th ACM Conference on Information and Knowledge Management. New York： ACM， 2009：1765-1768. 10.1145/1645953.1646225
11	FIGUEIREDO F， BENEVENUTO F， ALMEIDA J. The tube over time： characterizing popularity growth of YouTube videos［C］// Proceedings of the 4th ACM International Conference on Web Search and Web Data Mining. New York： ACM， 2011：745-754. 10.1145/1935826.1935925
12	钟志豪，肖井华，吴晔，等. 基于抖音平台的在线短视频流行度建模研究［J］. 电子科技大学学报， 2021， 50（5）：774-781. 10.12178/1001-0548.2021035
	ZHONG Z H， XIAO J H， WU Y， et al. Modeling dynamics of online short video popularity based on Douyin platform［J］. Journal of University of Electronic Science and Technology of China， 2021， 50（5）：774-781. 10.12178/1001-0548.2021035
13	GAO X F， ZHENG Z W， CHU Q Q， et al. Popularity prediction for single tweet based on heterogeneous Bass model［J］. IEEE Transactions on Knowledge and Data Engineering， 2021， 33（5）：2165-2178.
14	HU Y， HU C J， FU S S， et al. Predicting the popularity of viral topics based on time series forecasting［J］. Neurocomputing， 2016， 210：55-65. 10.1016/j.neucom.2015.10.143
15	LI J N， GAO Y R， GAO X F， et al. SENTI2POP： sentiment‑ aware topic popularity prediction on social media［C］// Proceedings of the 2019 IEEE International Conference on Data Mining. Piscataway： IEEE， 2019： 1174-1179. 10.1109/icdm.2019.00143
16	WANG X， WANG C， DING Z Y， et al. Predicting the popularity of topics based on user sentiment in microblogging websites［J］. Journal of Intelligent Information Systems， 2018， 51（1）： 97-114. 10.1007/s10844-017-0486-z
17	GUPTA M， GAO J， ZHAI C X， et al. Predicting future popularity trend of events in microblogging platforms［J］. Proceedings of the American Society for Information Science and Technology， 2012， 49（1）：1-10. 10.1002/meet.14504901207
18	王新乐，杨文峰，廖华明，等. 基于多维度特征的主题标签流行度预测［J］. 山东大学学报（理学版）， 2020， 55（1）：94-101.
	WANG X L， YANG W F， LIAO H M， et al. Popularity prediction of hashtags based on multi‑dimensional features［J］. Journal of Shandong University （Natural Science）， 2020， 55（1）：94-101.
19	MATSUBARA Y， SAKURAI Y， PRAKASH B A， et al. Rise and fall patterns of information diffusion： model and implications［C］// Proceedings of the 18th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York： ACM， 2012： 6-14. 10.1145/2339530.2339537

方法	观测窗口/h
方法	20	30	40
ARIMA	1.443 9	1.403 3	1.393 2
FCN	0.874 4	0.801 7	0.775 6
K‑SC+FCN	0.826 3	0.627 9	0.492 2
K‑SC+DTW+FCN	0.652 4	0.528 4	0.471 5
K‑SC+AADTW+FCN	0.603 1	0.496 6	0.463 8

方法	观测窗口/h
方法	20	30	40
ARIMA	1.443 9	1.403 3	1.393 2
FCN	0.874 4	0.801 7	0.775 6
K‑SC+FCN	0.826 3	0.627 9	0.492 2
K‑SC+DTW+FCN	0.652 4	0.528 4	0.471 5
K‑SC+AADTW+FCN	0.603 1	0.496 6	0.463 8

演化模式	观测窗口/h
演化模式	20	30	40
模式1	0.605 9	0.496 9	0.469 7
模式2	0.610 3	0.502 2	0.465 2
模式3	0.598 8	0.504 9	0.454 1
模式4	0.604 5	0.485 3	0.462 2
模式5	0.589 2	0.494 7	0.463 2
模式6	0.593 7	0.489 2	0.462 3

演化模式	观测窗口/h
演化模式	20	30	40
模式1	0.605 9	0.496 9	0.469 7
模式2	0.610 3	0.502 2	0.465 2
模式3	0.598 8	0.504 9	0.454 1
模式4	0.604 5	0.485 3	0.462 2
模式5	0.589 2	0.494 7	0.463 2
模式6	0.593 7	0.489 2	0.462 3

[1]	Yiping LANG, Wentao MAO, Tiejun LUO, Lilin FAN, Yingying REN, Xia LIU. Predictability evaluation and joint forecasting method for intermittent time series [J]. Journal of Computer Applications, 2022, 42(9): 2722-2731.
[2]	Minghui WU, Guangjie ZHANG, Canghong JIN. Time series prediction model based on multimodal information fusion [J]. Journal of Computer Applications, 2022, 42(8): 2326-2332.
[3]	Yinglü XUAN, Yuan WAN, Jiahui CHEN. Time series classification by LSTM based on multi-scale convolution and attention mechanism [J]. Journal of Computer Applications, 2022, 42(8): 2343-2352.
[4]	Liqing QIU, Fushuai QU. Emotional map of emergency based on sentiment analysis and influence evaluation [J]. Journal of Computer Applications, 2022, 42(5): 1330-1338.
[5]	Yifei WANG, Lei YU, Fei TENG, Jiayu SONG, Yue YUAN. Resource load prediction model based on long-short time series feature fusion [J]. Journal of Computer Applications, 2022, 42(5): 1508-1515.
[6]	Hexuan HU, Huachao SUI, Qiang HU, Ye ZHANG, Zhenyun HU, Nengwu MA. Runoff forecast model based on graph attention network and dual-stage attention mechanism [J]. Journal of Computer Applications, 2022, 42(5): 1607-1615.
[7]	Jie YANG, Mingyang ZHANG, Xiaobin RUI, Zhixiao WANG. Influence maximization algorithm based on node coverage and structural hole [J]. Journal of Computer Applications, 2022, 42(4): 1155-1161.
[8]	Xiaojie LI, Chaoran CUI, Guangle SONG, Yaxi SU, Tianze WU, Chunyun ZHANG. Stock trend prediction method based on temporal hypergraph convolutional neural network [J]. Journal of Computer Applications, 2022, 42(3): 797-803.
[9]	Qingqing WU, Lihua ZHOU, Xuanyi CUN, Guowang DU, Yiting JIANG. Influence maximization algorithm based on directed acyclic graph in heterogeneous information networks [J]. Journal of Computer Applications, 2022, 42(3): 895-903.
[10]	Yuxi LIU, Yuqi LIU, Zonglin ZHANG, Zhihua WEI, Ran MIAO. News recommendation model with deep feature fusion injecting attention mechanism [J]. Journal of Computer Applications, 2022, 42(2): 426-432.
[11]	Meizi LI, Yifei MI, Qian ZHANG, Bo ZHANG. Opinion leader recognition algorithm based on K-core decomposition in social networks [J]. Journal of Computer Applications, 2022, 42(1): 26-35.
[12]	WU Wei, LI Zeping, YANG Huawei, LIN Chuan, WANG Zhongde. Deep attention video popularity prediction model fusing content features and temporal information [J]. Journal of Computer Applications, 2021, 41(7): 1878-1884.
[13]	ZHANG Meng, LI Weihua. Influence maximization algorithm based on user interactive representation [J]. Journal of Computer Applications, 2021, 41(7): 1964-1969.
[14]	LI Yangying, CHEN Zhijun, ZHANG Zihao, YOU Lan. Raw sugar demand forecasting model for sugar manufacturing enterprise based on modified Elman neural network [J]. Journal of Computer Applications, 2021, 41(7): 2113-2120.
[15]	LI Guorong, YE Jimin, ZHEN Yuanting. Time series clustering based on new robust similarity measure [J]. Journal of Computer Applications, 2021, 41(5): 1343-1347.

Popularity prediction method of Twitter topics based on evolution patterns

基于演化模式的推特话题流行度预测方法

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 4

References 19

Related Articles 15

Recommended Articles

Metrics