A white paper exploring how bias in algorithms and data affect development problems, especially when they interact with socio-legal systems
When teaching the “low politics” of international relations, I used to begin with infectious diseases and the need for the WHO. Diseases do not respect borders; their control therefore cannot be limited to what goes on within national borders. Nation states need to cooperate. Therefore the justification for WHO. In its peculiar way, COVID-19 highlighted how connected the world has become.
I find myself a little defensive when I bring up the needs to access data quickly for policy-relevant research in gatherings dominated by fans of GDPR. Does not stop me, but I keep wondering what they think of me. But reading Siddhartha Mukherjee, a doctor deeply engaged in the fight against COVID-19 makes me feel much better.the System designers and lawyers have forgotten the original purpose of healthcare records: to help cure the patient. Finally, we need to acknowledge that our E.
This tour d’horizon examines the possible of uses of data to help stop or slow the spread of COVID-19 directly. It gives weight to what can be done in the short term.
In light of the lessons emerging from international experience, it is important to avoid local-government authorities from being tempted to sign exclusive agreements before becoming fully informed of the implications. What positive contributions can be made by higher levels of government? What network and facility sharing will be allowed? Is there value in providing general guidelines and model contracts, while allowing for normal negotiations to take place, perhaps backed up by some forms of low-cost dispute resolution mechanisms? When lamp posts and similar public fixtures become sophisticated sensing devices that pull in massive amounts of data, questions of who has access to the data under what terms will become important.
A research paper exploring an alternative approach to address the concern of privacy in sharing big data datasets by generating privacy-preserving artificial call detail records (CDRs) in accordance with the desired macro features of the dataset.
A whitepaper outlining the development of an alternative socioeconomic index for Sri Lanka, using principal component analysis (PCA) and publicly available census data
This involves working up from specific use cases, not only working down from principles.
An extended research abstract which identifies several criteria that can be used to identify mobile network call detail records (CDRs) affected by load sharing and establishes why that is a prevalent issue, especially in urban areas.
A research brief exploring the possibility of using remote sensing and neural networks to estimate the paddy crop extent in Sri Lanka
The group is an expansion of the UN Global Pulse Data Privacy Advisory Group which was established in 2014, and of which Samarajiva was a member since inception.
CEO Helani Galpaya was invited to speak at the "Asian Liveable Cities Forum: Digital Solutions for Livable Cities" conference held in San Francisco, 12-14 November 2019
LIRNEasia's comments on the Framework for a Proposed Data Protection Legislation for Sri Lanka of June 2019
This document is intended to understand the extant policy context in relation to healthcare data protection, providing international comparisons, and raise important questions for Sri Lanka to consider in relation to data protection, albeit within a narrow sector specific scope.
The biggest barrier to policing social media is language. Based on a draft LIRNEasia white paper on neural language processing. Published in Foreign Policy.
A whitepaper distilling LIRNEasia's current thoughts on the possibilities and issues with the computation extraction of syntactic and semantic language from digital text.