文中的可视化图均由tableau绘制。
1.1 数据展示
数据是来自美国纽约2016年5月份包括207个地铁站点在不同时间节点以及不同天气下的4.2649万条人流量数据,共包含21个字段。
1.2 数据说明
UNIT: Remote unit that collects turnstile information. Can collect from multiple banks of turnstiles. Large subway stations can have more than one unit.
DATEn: Date in “yyyymmdd” (20110521) format.
TIMEn: Time in “hh:mm:ss” (08:05:02) format.
ENTRIESn: Raw reading of cummulative turnstile entries from the remote unit.
EXITSn:Raw reading of cummulative turnstile exits from the remote unit.
ENTRIESn_hourly:Difference in ENTRIES from the previous REGULAR reading.
EXITSn_hourly:Difference in EXITS from the previous REGULAR reading.
datetime: Date and time in “yyyymmdd hh:mm:ss” format (20110501 00:00:00).
hour: Hour of the timestamp from TIMEn. Truncated rather than rounded.
day_week: Integer (0