Passive Visual Analytics of Social Media Data for Detection of Unusual Events

摘要: Now that social media sites have gained substantial traction, huge amounts of un-analyzed valuable data are being generated. Posts containing images and text spatiotemporal attached as well, having immense value for increasing situational awareness local events, providing insights investigations understanding the extent incidents, their severity, consequences, well time-evolving nature. However, large volume unstructured hinders exploration examination. To analyze such data, S.M.A.R.T system provides analyst with an interactive visual analysis spatial decision support environment assists in evacuation planning disaster management. fetches from various sources arranges them a perceivable manner, which is visually appealing. This turn aid finding abnormal events. Introducing passive mode makes tool more efficient, where it automatically detects idle time gives summary all anomalies encountered inactive period soon resumes monitoring. Using tool, can first extract major topics set selected messages rank probabilistically. The case studies past show improved by using methods mentioned before.