prediction


What is not bias?

Posted by on February 21, 2018  /  0 Comments

Bias is an important topic in general. It is of special significance to a research organization. Issues of bias being built into models that are beginning to play significant roles in society and economy are coming to the forefront of public discourse. So we decided to talk about this topic at a Journal Club. Colleagues from University of Moratuwa’s DataSearch also attended.
For more than a year, we at LIRNEasia have been working on the analysis of images. The NYT story on Stanford researchers working on Google Street View describes the potential well. For computers, as for humans, reading and observation are two distinct ways to understand the world, Mr. Lieberman Aiden said. In that sense, he said, “computers don’t have one hand tied behind their backs anymore.
Yesterday I listened sporadically to a live streamed conference on Big Data. Sporadic was not intentional. I am in Dili, Timor Leste, where most connectivity is via satellite with latencies in the 700ms range. Anyway, the focus was not on big data per se. They talked about all sorts of things, mostly open data (in the parts I heard) and crowd-sourced data.
I keep telling people that I (and the people associated with reforms in South Asia in the late 1990s) had no expectation that we would achieve the levels of voice and data connectivity that we have now achieved. But here is the proof, from a piece I gave to UNESCO while still Director General of Telecommunications back in 1999. In most countries of the South, basic telecommunication connectivity is still a distant goal, leave alone the advanced Internet infrastructure that provides the basis for electronic commerce. Electronic transactions between companies and organizations, particularly those involved in worldwide commercial relationships, do take place within the context of Electronic Data Interchange (EDI). However, these closed user systems do not fall within the commonly understood meaning of electronic commerce that involves transactions with consumers in an open system such as the Internet.

Predictive apps

Posted by on July 31, 2013  /  0 Comments

It was one thing for Gmail to ask “did you intend to attach a document to this email?” based on your use of the word “attached” in the email. But it moves things to a whole new level when an app analyzes your digital bread crumbs and tells you stuff that you haven’t even thought about. The services guess what you want to know based on the digital breadcrumbs you leave, like calendar entries, e-mails, social network activity and the places you take your phone. Many use outside services for things like coupons, news and traffic.
I know one thing about predictions: they are wrong. But we use them. They have their uses. Does the VNI have a good use? You can decide.
Lots of people talk about predicting earthquakes. Here’s the science. Why should we be in interested in earthquake prediction? Because we live in a bad neighborhood: there has been a tsunamigenic earthquake every year, except 2008, since 2004 in the Sunda Trench. Until the prediction issue is resolved all we can do is focus on warning and preparedness.