Online news recommendation systems aim to address the information explosion of news and make personalized recommendations for users. The key problem of personalized news recommendation is to model users' interests...Online news recommendation systems aim to address the information explosion of news and make personalized recommendations for users. The key problem of personalized news recommendation is to model users' interests and track their changes. A common way to deal with the user modeling problem is to build user profiles from observed behavior. However, the majority of existing methods make static representations of user profiles and little research has focused on effective user modeling that could dynamically capture user interests in news topics. To address this problem, in this paper, we propose UP-TreeRec, a news recommendation framework based on a user profile tree(UP-Tree), which is a novel framework combining content-based and collaborative filtering techniques. First, by exploiting a novel topic model namely UILDA, we obtain the representation vectors for news content in a topic space as the fundamental bridge to associate user interests with news topics. Next, we design a decision tree with a dynamically changeable structure to construct a user interest profile from the user's feedback. Furthermore, we present a clustering-based multidimensional similarity computation method to select the nearest neighbor of the UP-Tree efficiently. We also provide a Map-Reduce framework-based implemen-tation that enables scaling our solution to real-world news recommendation problems. We conducted several experiments compared to the state-of-the-art approaches on real-world datasets and the experimental results demonstrate that our approach significantly improves accuracy and effectiveness in news recommendation.展开更多
随着社交媒体影响力的增加,带有误导性内容的虚假新闻大肆传播,严重威胁了网络空间安全以及政治、经济、社会等领域的正常秩序。近年来,基于自然语言处理技术构建虚假新闻检测模型取得了良好的效果,但是依然存在一些问题:一是在模型构...随着社交媒体影响力的增加,带有误导性内容的虚假新闻大肆传播,严重威胁了网络空间安全以及政治、经济、社会等领域的正常秩序。近年来,基于自然语言处理技术构建虚假新闻检测模型取得了良好的效果,但是依然存在一些问题:一是在模型构建时忽略了新闻发布者与用户的情感因素;二是没有考虑用户心理偏好对于虚假新闻传播的影响。针对上述问题,提出了基于情感增强和用户偏好感知的虚拟新闻检测模型(Fake News Detection Model with Emotion Enhancement and User Preference Awareness,EEUP-FD),首先对双重情感特征(社会情感和新闻内容情感)进行表征和捕获,然后利用用户的历史数据作为用户偏好嵌入,并融合双重情感特征与新闻传播图从而实现虚假新闻的检测。实验表明,EEUP-FD模型在真实数据集上表现出优于现有检测模型的性能。展开更多
基金supported by the Beijing Natural Science Foundation (No.4192008)the General Project of Beijing Municipal Education Commission (No. KM201710005023)
文摘Online news recommendation systems aim to address the information explosion of news and make personalized recommendations for users. The key problem of personalized news recommendation is to model users' interests and track their changes. A common way to deal with the user modeling problem is to build user profiles from observed behavior. However, the majority of existing methods make static representations of user profiles and little research has focused on effective user modeling that could dynamically capture user interests in news topics. To address this problem, in this paper, we propose UP-TreeRec, a news recommendation framework based on a user profile tree(UP-Tree), which is a novel framework combining content-based and collaborative filtering techniques. First, by exploiting a novel topic model namely UILDA, we obtain the representation vectors for news content in a topic space as the fundamental bridge to associate user interests with news topics. Next, we design a decision tree with a dynamically changeable structure to construct a user interest profile from the user's feedback. Furthermore, we present a clustering-based multidimensional similarity computation method to select the nearest neighbor of the UP-Tree efficiently. We also provide a Map-Reduce framework-based implemen-tation that enables scaling our solution to real-world news recommendation problems. We conducted several experiments compared to the state-of-the-art approaches on real-world datasets and the experimental results demonstrate that our approach significantly improves accuracy and effectiveness in news recommendation.
文摘随着社交媒体影响力的增加,带有误导性内容的虚假新闻大肆传播,严重威胁了网络空间安全以及政治、经济、社会等领域的正常秩序。近年来,基于自然语言处理技术构建虚假新闻检测模型取得了良好的效果,但是依然存在一些问题:一是在模型构建时忽略了新闻发布者与用户的情感因素;二是没有考虑用户心理偏好对于虚假新闻传播的影响。针对上述问题,提出了基于情感增强和用户偏好感知的虚拟新闻检测模型(Fake News Detection Model with Emotion Enhancement and User Preference Awareness,EEUP-FD),首先对双重情感特征(社会情感和新闻内容情感)进行表征和捕获,然后利用用户的历史数据作为用户偏好嵌入,并融合双重情感特征与新闻传播图从而实现虚假新闻的检测。实验表明,EEUP-FD模型在真实数据集上表现出优于现有检测模型的性能。
