Generation of individual daily trajectories by GPT-2
Generation of individual daily trajectories by GPT-2
复制标题
DOI：
10.3389/fphy.2022.1021176
复制
发表时间：
2022-11
期刊：
影响因子：
--
通讯作者：
T. Mizuno;Shouji Fujimoto -Shouji-Fujimoto -2190114964;Atushi Ishikawa -Atushi-Ishikawa -2190114580
中科院分区：
文献类型：
--
作者：
T. Mizuno;Shouji Fujimoto -Shouji-Fujimoto -2190114964;Atushi Ishikawa -Atushi-Ishikawa -2190114580
我们提出了一种新的方法，将个人的日常轨迹转换为令牌时间序列的tokenizer“SentencePiece”的地理空间划分使用日本区域网格代码“JIS X 0，410”。此外，我们通过使用神经语言模型GPT-2学习令牌时间序列，构建了一个高度准确的个人每日轨迹生成器。模型生成的个人每日轨迹再现了五个现实属性：1）轨迹的每小时移动距离分布具有遵循对数函数的厚尾，2）移动距离的自相关函数表现出短时记忆，3）长距离移动中在移动1小时的方向上存在正的自相关，4）在每个个体的日常轨迹中，最终位置通常靠近初始位置，以及5）人的扩散取决于他们移动的时间尺度。
We propose a new method to convert individual daily trajectories into token time series by applying the tokenizer “SentencePiece” to a geographic space divided using the Japan regional grid code “JIS X0,410.” Furthermore, we build a highly accurate generator of individual daily trajectories by learning the token time series with the neural language model GPT-2. The model-generated individual daily trajectories reproduce five realistic properties: 1) the distribution of the hourly moving distance of the trajectories has a fat tail that follows a logarithmic function, 2) the autocorrelation function of the moving distance exhibits short-time memory, 3) a positive autocorrelation exists in the direction of moving for one hour in long-distance moving, 4) the final location is often near the initial location in each individual daily trajectory, and 5) the diffusion of people depends on the time scale of their moving.