arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1708.03381cs.LG
收藏

基于大规模数据的局部行为预测

Topical Behavior Prediction from Massive Logs

Qualcomm Inc.(高通公司)

Shih-Chieh Su

摘要:

本文研究了大尺度下的话题行为。我们使用网络日志,其中每个条目包含实体ID、时间戳和有关活动的Meta数据。结合递归神经网络(RNN)和卷积神经网络(CNN)的深度学习架构,探索行为的时间和空间关系。为了使行为数据适合于CNN中的空间学习,我们提出了几个简化步骤来形成主题度量,并将它们均匀地放置在图像中的像素中。实验结果表明,与多层感知器(MLP)网络相比,时间和空间的增益。引入了一种称为空间连接卷积网络(SCCN)的新学习框架,以更有效地预测主题度量。

英文摘要:

In this paper, we study the topical behavior in a large scale. We use the network logs where each entry contains the entity ID, the timestamp, and the meta data about the activity. Both the temporal and the spatial relationships of the behavior are explored with the deep learning architectures combing the recurrent neural network (RNN) and the convolutional neural network (CNN). To make the behavioral data appropriate for the spatial learning in the CNN, we propose several reduction steps to form the topical metrics and to place them homogeneously like pixels in the images. The experimental result shows both temporal and spatial gains when compared against a multilayer perceptron (MLP) network. A new learning framework called the spatially connected convolutional networks (SCCN) is introduced to predict the topical metrics more efficiently.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑