ID 原文 译文
56798 实验显示,本文提出的文本相关度计算方法要远好于传统的文本相关度计算方法. Experimental results indicate that our methodoutperforms conventional text correlation calculation methods.
56799 景点推荐系统可以帮助游客过滤大量的无关信息,还能辅助商家发掘潜在的顾客. The attraction recommendation systems not only filter out overwhelming irrelevant information forvisitors but also identify potential customers for service providers.
56800 然而,现有的基于传统方法的推荐系统,如基于内容的推荐或协同过滤系统,虽推荐过程相对透明直观,但由于数据稀疏性的存在,推荐结果往往不够准确; However, the current attraction recommenda?tion methods such as content-based methods, collaborative filtering, or deep learning-based methods are eitherinaccurate due to data sparsity, or lack of interpretability, which results in the users’ suspicion on the recommen?dation results.
56801 基于深度学习的推荐方法,虽在一定程度上提高了推荐结果的精度,但由于缺乏可解释性和透明度,难以满足部分用户理解推荐依据的愿望,也阻碍了此类方法的推广应用. To address the limitations of the current methods, we introduce a novel framework for preferencepropagation on knowledge graphs (KGs), which utilizes lots of parameters to capture the abundant semantics ofexisting KGs more comprehensively, and meanwhile explains the results through reasoning the link paths fromuser’s history to candidates on KGs.
56802 为了解决当前方法所存在的局限,本文引入基于知识图谱的景点推荐框架,将推荐过程与知识图谱嵌入相结合,推断用户兴趣在知识图谱上的传播路径,以此作为推荐依据. With a multi-view spatiotemporal analysis on real-world travel data, weinvestigate the geographical characteristics of human tour activities and build a tourism-oriented KG based onopen web resources.
56803 此外,本文通过对真实旅游数据的多角度时空分析,探究旅游活动的时空规律,并将其应用于景点推荐框架中,提出一种面向旅游的基于知识图谱的可解释推荐方法——Geo-RippleNet,并通过构建基于开放网络资源的旅游知识图谱,对Geo-RippleNet进行了全面的实验验证. Then, we propose a KG-aware attraction recommendation method named Geo-RippleNetand implement it with extensive experiments on large-scale datasets.
56804 结果表明,本文提出的基于知识图谱的景点推荐方法,不仅可以最大限度地吸收知识图谱丰富的语义信息,从而实现可观的性能提升,还能充分利用图谱的关系知识,推理兴趣传播路径,以增强推荐结果的可解释性. It is argued that the framework for prefer?ence propagation on KGs not only absorb rich semantic information to achieve substantial performance gains inthe attraction recommendation scenario but also enhance the interpretability of recommendation results with thesupport of abundant relational knowledge.
56805 此外,将旅游活动的时空规律融入到上述推荐框架中,能够还原用户出游和决策的时空过程,进一步提高方法的性能表现. Moreover, incorporating the spatiotemporal characteristics of humantour activities into the framework for preference propagation further makes the recommendation performancemore aligned with the potential interests of visitors.
56806 近年来, SARS病毒和MERS冠状病毒等全球传染性疫情频繁发生, 2019年12月至2020年5月7日,新冠肺炎(COVID-19)疫情已造成世界范围内210多个国家的370多万人感染,其中超过26万人死亡. In recent years, SARS virus, MERS coronavirus, and other global pandemics have occurred frequently. The outbreak of COVID-19 has caused more than 3. 7 million infected people and 260000 deaths over 210 countriesaround the world from December 2019 to May 7, 2020, which has seriously affected the safety of people, thestability of social order, and the development of the economy.
56807 此次疫情严重影响了人民生命安全、社会正常秩序和经济稳定发展,因而探索和实施面向重大传染病的新型防控方案迫在眉睫. Therefore, it is urgent to explore and implementnew prevention and control programs for such major pandemics.