arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1909.07523cs.CYcs.CLcs.LGcs.SI
收藏

发现差异特征:对抗性学习的信息可信度评估

Discovering Differential Features: Adversarial Learning for Information Credibility Evaluation

Xi’an Jiaotong University(西安交通大学)

Lianwei Wu, Yuan Rao, Ambreen Nazir, Haolin Jin

摘要:

一系列的深度学习方法提取了大量的可信度特征来检测互联网上的假新闻。然而,这些提取的特征仍然遭受许多不相关的和嘈杂的功能,严重限制了性能的方法。在本文中,我们提出了一种新的模型,基于对抗网络和共享私有模型(ANSP)的启发,其目的是减少共同的,不相关的信息可信度评估的特征提取的功能。具体而言,ANSP涉及两个任务:一是防止真假信息的二进制分类,以捕获依赖于强化学习引导的对抗网络的共同特征。另一种是从多种类型的可信度信息中提取可信度特征(以下称私密特征),并通过两种策略与公信度特征进行比较,正交性约束和KL-发散,以使私有特征更具差异性。首先在两个六标签LIAR和微博数据集上的实验表明,ANSP实现了最先进的性能,分别将准确率提高了2.1%,3.1%,然后在四标签Twitter 16上验证了模型的鲁棒性,性能提高了1.8%。

英文摘要:

A series of deep learning approaches extract a large number of credibility features to detect fake news on the Internet. However, these extracted features still suffer from many irrelevant and noisy features that restrict severely the performance of the approaches. In this paper, we propose a novel model based on Adversarial Networks and inspirited by the Shared-Private model (ANSP), which aims at reducing common, irrelevant features from the extracted features for information credibility evaluation. Specifically, ANSP involves two tasks: one is to prevent the binary classification of true and false information for capturing common features relying on adversarial networks guided by reinforcement learning. Another extracts credibility features (henceforth, private features) from multiple types of credibility information and compares with the common features through two strategies, i.e., orthogonality constraints and KL-divergence for making the private features more differential. Experiments first on two six-label LIAR and Weibo datasets demonstrate that ANSP achieves the state-of-the-art performance, boosting the accuracy by 2.1%, 3.1%, respectively and then on four-label Twitter16 validate the robustness of the model with 1.8% performance improvements.

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。

↑