Файл:Real-time detection of operational military information in social media (IA realtimedetectio1094547261).pdf

Access to information has never been easier and people’s eagerness and ability to publish information on social media platforms has never been higher. The growing mountain of information has presented an opportunity and a significant challenge for data scientists. The military in particular can benefit from the ability to use public information to gain an awareness of its current vulnerabilities as well as learning about its adversaries. This thesis explores methods for collecting public information from social media that may be revealing operational military movements. This research demonstrates that it is possible to train a machine to search for and find military members in social media by using publicly available information distributed by the military. The postings of military members, once identified, can then be ingested and processed in real time, allowing the timely detection of possible military information that had been posted in social media.

Subjects: Twitter; tweet; social media; machine learning; support vector machine; big data; data mining; Python; text classification