位置感知的大型复杂网络数据分析与应用

---

面对空间社会网络、交通网络等位置感知网络数据日益增长的挑战，网络数据处理技术在数据收集、清理、组织、解释、分析、利用和可视化等各个阶段都经历着革命性的变化。这些变化导致大数据框架、网络分析建模、链接或路径预测、推荐系统趋同成为全球瞩目的发展趋势。本特刊旨在提供一个论坛，介绍关于大型复杂网络数据的融合研究的最新进展。挑战包括城市中的实时事件检测、交通网络中的拥堵发现、社交用户的位置预测、物理世界中的社交用户行为识别以及处理多维复杂网络数据的统一系统。稳健的解决方案要求在机器学习、遗传算法、混沌、遗传算法、细胞自动机、神经网络和进化博弈论等领域采用高度创新的技术。

这期特刊已经吸引了来自机器学习、人工智能、数据挖掘、自然语言处理、数据和网络挖掘等领域的全球研究人员的高质量投稿，和大数据管理，利用他们的专业知识和匹配的挑战，开发更有效和实用的算法或模型，从日常产生的宝贵的社交网络数据和交通网络数据获取智能知识。总共有40份意见书。经过至少两名评审员的单盲同行评审，19篇论文最终被接受发表。合格率为47.5%。每篇论文的作者平均为4.2人。作者的附属机构来自中国、澳大利亚、法国、印度、沙特阿拉伯和加拿大。这些被接受的论文可以被组织在不同的组中。第一组文章的重点是基于位置的社交网络研究。S. Yang等人的论文《位置感知社会网络中的目标影响节点选择》提出了一种识别位置感知社会网络中基于目标的影响节点的有效解决方案。 The paper titled “Discovering Travel Community for POI Recommendation on Location-Based Social Networks” by L. Tang et al. improved the community detection method for high-quality point-of-interest recommendation. The other three papers—“A Joint Deep Recommendation Framework for Location-Based Social Networks” by O. Tal and Y. Liu, “Optimal Proxy Selection for Socioeconomic Status Inference on Twitter” by J. L. Abitbol et al., and “Incremental Bilateral Preference Stable Planning over Event Based Social Networks” by B. Li et al.—invented the new machine learning technologies to explore the useful annotations for users and recommendation. The focus of the second group of articles is traffic based road network research. The papers titled “An Effective Algorithm for Video-Based Parking and Drop Event Detection” by G. Li et al., “Location-Aware Web Service Composition Based on the Mixture Rank of Web Services and Web Service Requests” by J. Lu et al., and “Predicting Quality of Service via Leveraging Location Information” by L. Chen et al. provided the effective solution to discover the important location of delivering high-quality service in smart city environment. The other three papers titled “Semantic-Aware Top-k Multirequest Optimal Route” by S. Wang et al., “A Novel Index Method for K Nearest Object Query over Time-Dependent Road Networks” by Y. Yang et al., and “A Destination Prediction Network Based on Spatiotemporal Data for Bike-Sharing” by J. Jiang et al. proposed efficient algorithms and models to choose the optimal route planning. The focus of the third group of articles is deep learning. The papers titled “Sign Prediction on Unlabeled Social Networks Using Branch and Bound Optimized Transfer Learning” by W. Yuan et al., “Cognitive Driven Multilayer Self-Paced Learning with Misclassified Samples” by Q. Zhu et al., and “A Block Object Detection Method Based on Feature Fusion Networks for Autonomous Vehicles” by Q. Meng et al. investigated the deep learning technologies to optimize the entity recognition in social network data. The other five papers titled “pSPARQL: A Querying Language for Probabilistic RDF Data” by H. Fang, “Edge Computing in an IoT Base Station System: Reprogramming and Real-Time Tasks” by H. Wu et al., “Finding the Shortest Path with Vertex Constraint over Large Graphs” by Y. Yang et al., “Evaluation of Residential Housing Prices on the Internet: Data Pitfalls” by M. Li et al., and “Promoting Geospatial Service from Information to Knowledge with Spatiotemporal Semantics” by J. Geng et al. contributed a diverse range of topics using the semantic information. The solutions proposed in these research works include transfer learning, edge computing, data index structure, self-space learning, video analysis, feature fusion networks, multilayer classification, spatiotemporal pattern recognition, and probabilistic resource description framework data analysis. The effectiveness of the proposed solutions to the targeted problems has been reported based on empirical study and/or analysis.

综上所述，研究论文涵盖了广泛的应用领域，包括地理空间服务、基于位置的社会网络推荐、社会网络影响节点选择、住宅房价预测的数据陷阱检测、社会经济状况推断、停车检测、自动驾驶汽车目标检测、服务质量预测、共享单车、社交网络标签预测、时变道路网络k近邻查询、物联网基站系统优化。

的利益冲突

作者声明他们没有利益冲突。

致谢

这项工作得到了澳大利亚研究理事会发现计划的资助。DP160102114。

李所
克邓
鑫黄
佳洁徐

版权

版权所有©2019李建新等人。这是一篇发布在知识共享署名许可协议，允许在任何媒介上不受限制地使用、传播和复制，但必须正确引用原作。