Group items tagged

From basic statistics to machine learning and new ways to think about visualization, the Data Science Starter Kit gives you the tools you need to get started with data. If you haven't yet taken the leap, why wait? And if you're already experienced with data, the Starter Kit will push you further. The package includes (8) titles on R, basic statistics and data analysis, Python, machine learning, and visualization. This kit includes everything you need from analysis, visualization, to management.

Kovas Boguta, the head of analytics at Weebly and a guest author on ReadWriteWeb, has created another powerful data visualization, this time of the "the pro-democracy movement in Egypt and across the Middle East." The visualization drew from Twitter use by Egyptians and influential others around the #jan25 uprising. Those writing in Arabic only are represented in red, only in English are in blue and overlap by various shades of purple. Influence, in terms of follows, are represented by lines and those who influence each other are located in proximity.

Stefan Sagmeister recently introduced me to a unique forthcoming book, "a worthwhile endeavor, showing scientists the fundamentals of how to design a complex graphic," he said. Incidentally, he is the designer and a consultant. The title may be a mouthful, Visual Strategies: A Practical Guide to Graphics for Scientists and Engineers by Felice Frankel and Angela DePace (Yale University Press), yet it is more than worthwhile, it is an essential guide to literacy for fields that are essential to all our lives.

A number of instructors have been using NodeXL to help teaching social network analysis. It is relatively easy to use compared to many other network analysis and visualization tools, while still providing a rich set of metrics and visualization features. It is also free and integrates with Excel 2007, which many users are familiar with. This page is meant to collect information that can be useful in teaching with NodeXL.

The JavaScript InfoVis Toolkit provides tools for creating Interactive Data Visualizations for the Web.

Open data initiatives such as Linked Data have begun to see adoption (e.g. NYT), with the objective of opening up and creating relationships between databases. This presents opportunities to break down the barriers that separate individual websites and their content, creating unprecedented opportunities for the reuse and combination of different data sources. A common critique of apps has been the walled-garden approach to storing data and communicating with other apps-open data promises to change that. As a consequence we will begin to see more apps that aggregate a multitude of sources, rather than remaining tethered to any particular set of content (though those will likely continue to exist). These will be apps designed for a particular scenario and user experience-their primary point of differentiation. They will use whatever data sources are available in order to best satisfy the goals of whatever experience or scenario is being designed.

Great news browser and filter, but very limited network visualizations

A non-profit tech company that develops free and open source software for information collection, visualization and interactive mapping.

Recorded Future offers robust tools for temporal and predictive analysis including advanced visualizations, data for predictive modelling, and fine-grain Future oriented alerts. Customers around the world are using Recorded Future Premium to monitor media on the web, understand news and forecast a variety events, track sentiment surrounding events and more.

"Natural Earth is a public domain map dataset available at 1:10m, 1:50m, and 1:110 million scales. Featuring tightly integrated vector and raster data, with Natural Earth you can make a variety of visually pleasing, well-crafted maps with cartography or GIS software."

See relationships among data points: Network Diagram; Scatterplot; Matrix Chart Compare a set of values: Bar Chart; Block Histogram; Bubble Chart Track rises and falls over time: Line Graph; Stack Graph; Stack Graph for Categories See the parts of a whole: Pie Chart; Treemap; Treemap for Comparisons Analyze a text: Word Tree; Tag Cloud; Word Cloud Generator; Phrase Net See the world: Massachusetts Map; World Map; US County Map; New Jersey Map

"A Seattle data technology startup is taking the phrase "know your customer" to a whole new level. Versium Analytics today introduced a new analytics platform called LifeData. Led by former InfoSpace executives Chris Matty and Kevin Marcus, the 12-month-old company is trying to help organizations learn more about their customers by using a patent-pending technology that verifies and cross-indexes large sets of data."

Time-series network data to a dynamic animation using Visone (www.leydesdorff.net/visone).

cool EPA mapping application of pollution sites

"In a post-fact era, you can be an elected official and have a remarkably flexible relationship with the truth. Take climate science: more than 97% of scientists agree that climate change is a man-made phenomenon, but conservative politicians--and more than 65% of Republicans in Congress--outdo one another to demonstrate just how little they believe in science."

NodeXL is an open-source template for Excel 2007 and 2010 which provides the utility to create a hierarchical network edge list. With NodeXL, you can create a customizable hierarchical network graph and zoom, scale, or pan it for a more refined look of added data. Networks can also be imported or exported to a variety of file formats. For example, you can create a graph displaying a company's chain of command and structure of subordinate employees. NodeXL provides the convenience of eluding difficult applications, arcane file formats and programming language for creating such graphs. You can even import a list of your Twitter followers or YouTube and Flickr connections, to create graphs representing your social network.

The advantage of NodeXL, particularly for graphing Twitter communities, is it has built-in features for grabbing the data for you. Not only that the coding is clever enough to handle the data collection for mere mortals, so when you hit your rate limit NodeXL waits until it should be able to get more data. NodeXL also has "built-in connections for getting networks from Flickr, YouTube, and your local email. Additional importers for Exchange Email, Facebook, and Hyperlink networks are available".

Luka Potkonjak has contributed a new feature to NodeXL: edge bundling.  Following on the heels of the release of curved edges in version. 175, we now provide a way to gather parallel edges together into "bundles" that travel together until they diverge close to their destinations.

Rioting and related intra-crowd dynamics are significant human processes, but we know less about the basic behavioral science and subsequent processes that drive and shape rioting than we would like to. This is due, in large part, to the difficulty in studying riots on the ground and to the sheer complexity of riot phenomena. We know even less about the geographical dynamics of rioting, even though there is a dedicated (but only general) appreciation that geography is important. Existing work has, for the most part, adopted the most straightforward path to discovery, by examining coarse (city-scale) geographies of rioting, or in the few instances where intra-crowd riot dynamics are considered they have focused on stylized abstractions of behavior. Because of the difficulties of using standard social science inquiry to study riots (surveys, ethnographic analysis, interviews), many researchers have turned to computer modeling to create synthetic riots that can be configured, sampled, and experimented with. But, building models of something as bewilderingly complex as rioting is really quite difficult and so many short-cuts are taken. In particular, models are usually cellular-based in form (where rasters represent people and their local environment) and founded on physical interactions between relatively "dumb" particle-people (where continuum mechanics, random walks, or particle-particle forces serve as a substitute for socio-spatial interaction and behavior).