人工智能培训

搜索

论文代码开源:流媒体新闻的多语种聚类(Multilingual Clustering of Streaming News)

[复制链接]
admin 发表于 2018-9-8 09:12:21 | 显示全部楼层 |阅读模式
admin 2018-9-8 09:12:21 457 0 显示全部楼层
人工智能论文代码开源:流媒体新闻的多语种聚类(Multilingual Clustering of Streaming News)请注意该人工智能论文代码开源在github,大部分是python写的,框架可能是tensorflow或者pytorch。本文讨论了以下基本问题:给出了两个相同的图形布局,哪一个更美观?我们提出了一种基于神经网络的鉴别器模型,该模型在标记数据集上训练,该模型决定了两种布局中哪一种具有更高的美学质量。用作模型输入的特征向量基于已知的图绘制质量度量,经典统计,信息理论量和受凝聚态物理方法启发的双点统计。用于训练和测试的大型布线对使用力导向绘制算法和自然源于图生成过程的布局构建。使用数据增强技术进一步扩展。我们模型的平均预测准确率为95.70%,优于基于压力的判别准则和流行质量指标的线性组合,具有统计学上的显着差异。
This paper addresses the following basic question: given two layouts of thesame graph, which one is more aesthetically pleasing?We propose a neuralnetwork-based discriminator model trained on a labeled dataset that decideswhich of two layouts has a higher aesthetic quality.The feature vectors usedas inputs to the model are based on known graph drawing quality metrics,classical statistics, information-theoretical quantities, and two-pointstatistics inspired by methods of condensed matter physics.The large corpus oflayout pairs used for training and testing is constructed using force-directeddrawing algorithms and the layouts that naturally stem from the process ofgraph generation.It is further extended using data augmentation techniques.The mean prediction accuracy of our model is 95.70%, outperformingdiscriminators based on stress and on the linear combination of popular qualitymetrics by a statistically significant margin.论文代码开源:流媒体新闻的多语种聚类(Multilingual Clustering of Streaming News) FqvkV3Gz4H4P4yEh.jpg
URL地址:https://arxiv.org/abs/1809.01017v1     ----pdf下载地址:https://arxiv.org/pdf/1809.01017v1    ----         ----github下载地址:https://github.com/5gon12eder/msc-graphstudy    ----    论文代码开源:流媒体新闻的多语种聚类(Multilingual Clustering of Streaming News)请注意该人工智能论文代码开源在github,大部分是python写的,框架可能是tensorflow或者pytorch,keras,至于具体是哪一个没有完全测试。
回复

使用道具 举报

您需要登录后才可以回帖 登录 | 立即注册

本版积分规则 返回列表 发新帖

admin当前离线
管理员

查看:457 | 回复:0

快速回复 返回顶部 返回列表