As my Google Summer of Code project my plan was to develop an Integrative data visualization environment for TCIA. Over the summer me and my mentor Prof. Ashish Sharma, Emory University, decided to make a more general purpose and a declarative framework. What we finally came up with is InDEx. Integrative: InDEx is integrative in the sense that it can take data from a variety of data sources, namely: REST APIs, flat files etc. These data sources might provide data in the form of JSON, CSV etc. Declarative: Dashboard authors can declare an InDEx dashboard using 4 configuration files: dataSource.json, dataDescription.json, interactiveFilters.json, visualization.json
Problems with Github gists: The Indian Government blocked 32 websites a few days back, Github being one of them. I used Github gitsts for hosting a lot of my code snippets that I embed in my blog posts. Unfortunately for Indian readers(including me -_-) the gists wont be visible unless ofcourse they're smart enough(which I hope they are) to change their DNS. Update: Ok Github is unblocked now 😀
Dataexperiments.net has been up for a couple of weeks now. Its a data science blog that I have started for journaling my data science related experiments and other interesting stuff pertaining to data science, machine learning, data visualization, big data etc. The post titled "The Goal of this Blog" clearly describes the ambitions and aims of the blog. I'll continue writing about other cool computer science related stuff here, but would have lesser data science related posts as dataexperiments would be their rightful place.
I successfully completed my Google Summer of Code project last week. It has been an enriching journey and I am both delighted and saddened upon the conclusion. My project started of as creating a dashboard for TCIA but ended up being much bigger: A Data Visualization Engine. The initial few weeks were spent mainly on prototyping and running benchmark tests to push dc.js and crossfilter to their limits. We soon realized that dc.js approach of doing everything: loading, filtering and rendering on the client side would not work for the scales of data that we needed. We shifted the loading