Twitter位置数据集

Twitter位置数据集是由新加坡国立大学计算机学院创建的，包含了2006年至2010年间超过100万Twitter用户的数据，其中约20万条记录包含地理位置信息，主要集中在北美地区。该数据集用于研究在保持差分隐私的前提下发布位置数据集的方法。数据集的创建过程涉及使用局部保持映射将数据点映射到一维空间，并通过添加拉普拉斯噪声来实现隐私保护。该数据集的应用领域主要集中在隐私保护的数据发布和分析，旨在解决如何在保护个人隐私的同时，允许公众灵活地分析和探索数据的问题。

The Twitter Location Dataset was created by the School of Computing, National University of Singapore. It contains data of over 1 million Twitter users spanning 2006 to 2010, with approximately 200,000 records including geographic location information, primarily concentrated in North America. This dataset is utilized to investigate methods for publishing location datasets while maintaining differential privacy. The creation process of the dataset employs Locality Preserving Mapping to map data points into a one-dimensional space, and achieves privacy protection by adding Laplace noise. The main application areas of this dataset are privacy-preserving data publishing and analysis, aiming to address the challenge of enabling the public to flexibly analyze and explore data while safeguarding personal privacy.