Developing a streaming data processing workflow for querying space–time activities from geotagged tweets

• This paper proposes a new streaming data processing workflow for querying space-time activities (STA).
• The approach allows exploring geotagged tweets to discover STA patterns of daily life reducing participant’s mistakes.
• Different tasks have been implemented using cloud resources for handling six months of geotagged tweets from Canada.
• STA patterns have revealed activity choices that might be attributable to personal motivations for communicating an activity.

The critical dimensions in describing space–time activities are “what“, “where”, “when”, and “who”, which are frequently applied to collect data about basic functions people perform in space in the course of a day. Collecting data about these dimensions using activity-based surveys has presented researchers with a number of technical and social limitations, ranging from the restricted period of time participants have to record their activities to the level of accuracy with which participants complete a survey. This paper proposes a new streaming data processing workflow for querying space–time activities (STA) as a by-product of microblogging communication. It allows exploring a large volume of geotagged tweets to discover STA patterns of daily life in a systematic manner. A sequence of tasks have been implemented using different cloud-based computing resources for handling over one million of daily geotagged tweets from Canada for a period of six months. The STA patterns have revealed activity choices that might be attributable to personal motivations for communicating an activity in social networks.