特徵為基礎之預測社交演化的雙RNN與模式學習模型

A Pattern-based Dual Learning for Social Evolution Prediction

梁紘銘(Hung-Ming Liang)

指導教授 ： 王英宏

淡江大學/工學院/資訊工程學系碩士班/碩士(2021年)

https://doi.org/10.6846/TKU.2021.00263

摘要

社群網路是近幾年十分蓬勃發展的一個科技，隨著他蓬勃發展，各式各樣預測社群網路的方法因應而生，這些方法大多是使用靜態社群網路來做預測，然而現實中的社群網路則大多是動態改變的，本研究以Mobile01論壇的資料為資料庫，利用EPMiner資料探勘演算法從數據中找出頻繁序列，利用長短期記憶（LSTM）模型預測社群網路關係。本研究提出了一個具有三個 LSTM 的兩層架構模型來進行預測。

關鍵字

機器學習 ； 社群網路 ； 資料探勘 ； 頻繁序列

並列摘要

Social networking is a technology that has been booming in recent years. With its vigorous development, various methods of predicting social networks have emerged. Most of these methods use static social networks to make predictions. However, social networks in reality are mostly dynamically changing. This research uses the data of the Mobile01 forum as the database, uses the EPMiner data mining algorithm to find frequent sequences from the data, and uses the long short-term memory (LSTM) model to predict social network relationships. This research proposes a two-tier architecture model with three-LSTM to make predictions.

並列關鍵字

Dynamic social network ； Long Short-Term Memory ； DNLP ； Deep learning

參考文獻

[1] L. A. Adamic and E. Adar, "Friends and neighbors on the web," SocialNetworks, vol. 25, no. 3, 2003, pp. 211–230.

[2] S. Asur, S.Parthasarathy, D.Ucar, "An event-based framework for characterizing the evolutionary behavior of interaction graphs," 13th ACM SIGKDD, 2007, pp. 913-921.

[3] J. F. Allen, "Maintaining knowledge about temporal intervals," Proceeding of the Communications of ACM 26(11), 1983, pp. 832–843.

[4] M. Berlingerio, F. Bonchi, B. Bringmann, A. Gionis, "Mining Graph Evolution Rules," Proceedings of the European Conference on Machine Learning and Knowledge Discovery in Databases (ECML PKDD’09), 2009, pp. 115-130.

[5] K. M. Borgwardt, H. Kriegel and P. Wackersreuther, "Pattern Mining in Frequent Dynamic Subgraphs," Sixth International Conference on Data Mining (ICDM'06), 2006, pp. 18-822.