Recently I was invited to an event by NUS’s Lee Kuan Yew School of Public Policy ( Singapore ) for a seminar on ‘Digital Frontiers Series – Assessing the rationality of Political Online Space : Man and Machine’. What an eye-opening session it was !
This is what LKY School did : They had a Cloudera Server and build algorithms to go through and crawl the different social media posts to aggregate political content. On this aggregated data, they sought to derive intelligence on the relevance of the posts to the actual events, how rational the posts were, whether any actionable intelligence could be derived out of the aggregated data.
They specifically looked for content pertaining to the following :
- National Library Board’s book ban
- PR Lee versus Roy Ngerng
- Homosexuality, LGBT, PinkDot
- Religion\Education, P1 Registration
- CPF, Minimum Sum, MediShield Life
- Foreigners and Immigration
- Miscellaneous topics
And on the content pertaining to these, analysis was on the below and the interrelation between the following:
- Author’s Identity
- Journalistic Objectivity
- Emotionality
- Partisanship
What an enlightening experience it was ! Having been an analytics person by profession, I was amazed at the level of detail to which the data scientist’s team in NUS and SMU had gone to perform various analyses like Sentiment Analysis on social media.
NUS’s report on the proceedings here.
Amaruvi's Aphorisms 