Upstream Data Mining And Data Viz Go Hand In Hand

18/11/2011 2:05:14 PM
by Sebastian Schweigert
A worthwhile read by Enrico Bertini on "Why Visualization Cannot Afford Ignoring Data Mining and Vice Versa".

datamining-(1).jpg

Here are a few notable exerpts that we really enjoyed from his article:

- Data is full of rubbish: I repeated it several times in this blog. Data never comes for free, you have to manipulate it in order to accommodate the needs you have for your project. The most classical things you will need to deal with are: missing values, outliers detection, normalization, aggregation, sampling, etc., but every project comes with its own bag of necessary data wrangling. Each one of these requires robust and solid techniques, it is not something you can improvise. And no matter how skilled a data visualization expert you are, you will need to borrow solid techniques from dataminers, otherwise you are an amateur.
- Humans don’t scale, machines do: There is no way to visualize a billion items. really believe me, there’s no way to do that effectively. If you assign every item to one single pixel (known as pixel-based visualization), which is the maximum scalability available, you will need either a huge screen or very tiny pixels. In both cases our body has limitations. With a huge screen your perception is hampered by the maximum field of view, that is, there’s no way to embrace the whole screen with your eyes. With tiny pixels the human eye is limited by its maximum resolution. On the other hand machines do scale and can crunch monstrous amounts of data. Add a number of machines to your cluster and you have more power.
 
- You cannot trust black boxes. The issue of trust is very well known among dataminers: the models data mining algorithms build are often arcane and even if something seems to work, there’s no way to really understand why and how it works. Visualization has the power to shorten this gap and help model builders gain better confidence on the babies they build.
- There’s no right answer. Data Mining has a long tradition for providing tools to build models that give clear cut answers automatically: “should I give the loan to this customer or not?“. This is fine and useful and it’s been a very successful model for data mining so far. But many of the modern inquiries on data are not so clear-cut. Data analysis is often exploratory and and there’s no right answer. When mining is used for this purpose it necessarily needs a certain level of flexibility: ask a question, produce some initial results, visualize them, understand better the problem, change the parameters, use another algorithm, compare alternative results etc … and how do you do that without visualization?
 
 

Well worth the time for a full reading:



20 Visualizations to Help You Understand Crime

17/11/2011 5:02:56 PM
by Sebastian Schweigert
If you work with crime data chances are you spend a lot of time trying to find crime trends for your self or for those you report to.  With a good visualization finding these patterns and sharing this information becomes much easier. Here Flowing Data showcases 20 different ways you can visualize your data to help you understand your data.  You'll notice that in this set mapping visualizations and statistical visualizations seem to be quite popular.  When selecting a visualization method keep in mind what you would like to show with your data.  For example, which of these 20 visualizations are better for finding patterns versus reporting trends?  Maybe you're trying to reach a much broader audience and are simply trying to entertain as opposed to inform. Each visualization is has its own purpose.  To help you decide which visualization is right for you we will periodically post blog articles about the strengths and weaknesses of differeint visualization methods. 

sf-crimemap-545x448.jpg
Click on the image to open up the gallery. 

 

Need help in choosing a visualization right now?  We would be glad to give you a hand. Please do not hesitate to contact info@geotime.com with any questions you may have. 



Investigating Money Laundering Cases: How The Who/Where/When Make All The Difference

17/11/2011 3:01:16 PM
by Global Administrator
A brief overview of the importance of the "Who/Where/When" when investigating money laundering cases. By looking at when someone is moving money around, patterns of illicit activity are clearly visible. It comes down to the face that techniques such as smurfing involve geographically diverse deposit locations and can be spread out over time, all to avoid detection. By seeing these geo-temporal patterns, it is possible to quickly identify new suspects by mapping the time and location of deposits to ATM video survaillance footage. 


Money Laundering Investigation Using GeoTime

This short video shows what financial transactions look like over time and geography, which is key to being able to spot patterns. The next step is to show the relationship between the accounts and the associated individuals to understand how the money moves through a network of organized criminals. 


Mapping the Mexican Drug War

17/11/2011 10:24:57 AM
by Sebastian Schweigert
The LA Times has put together a visualization summarizing the Mexican drug war.  You'll notice that they have included information on key players, number of deaths (by week and location), and indicated regional hotspots.  There are also some great interactive components within this graphic.  You can adjust the timeline, associate each key player with a location, and filter by topic or location.  If you were to use this visualization as a tool to summarize the drug war in Mexico, what other information would you want to see?

Click on the image below for the fully interactive graphic. 

LA-Times.jpg


Choosing Clarity over Style: World Food Prices

17/11/2011 9:45:27 AM
by Sebastian Schweigert
 One of the most important considerations to make when visualizating data is the complexity of your visualization.  We all want the "wow factor" in our data, but there are times when a simpler approach to visualization tells a better story. Your aim should be to inform rather than to entertain. For example, look at the increase in world food prices as visualized in Wolfram Alpha.  By using a simple line chart we can clearly see when the price of coffee has increased over the last 20 years.  With few visual elements its easy to see the price spikes in lat 1994, 1998 and 2011.  Below you will also see a collection of tables which give you the average current world price of different foods.  Here a table works just fine becaue we are comparing the price of one food item at one point in time.  It is easy to get the relative value of one item to another. 

World-Coffee-Prices.jpg


A Periodic Table of Visualization Methods

17/11/2011 9:16:35 AM
by Sebastian Schweigert
Visual-Literacy.org has provided this great interactive visualization of visualizations! In its Periodic Table of Visualization, they have categorized a multitude of visualization styles. They've also provided some suggestions on what type of visualizations are best for certain types of information. Roll over each visual element to view an example. Many of these examples would work well when you're trying to convey this information to others in presentations and reports.  If you want to analyze data, you may want to consider making parts of these visualizations more dynamic.  For example, with a Data Map it would be useful to have a dynamic animation in order to see changes in regional data over time.  With a Semantic Network, it would be great to have filters which highlight important nodes within the network. 

Click on the image below to open the interactive visualization. 
 A-Periodic-Table-of-Visualization-Methods---Google-Chrome_2011-11-17_10-11-47.jpg



7 Billion: How Did we Get So Big So fast - Dynamic Visualizations and Metaphors

16/11/2011 3:09:44 PM
by Sebastian Schweigert
Sometimes a static graphic alone can't tell the complex visual story hidden in your data.  NPR has provided a great example in this video of how to using moving and animated elements to tell a story of how the world's population has grown to 7 billion people. Watch as they use the flow of water to indicate births and deaths and color to separate each continent.  You'll notice they also shift between different visualization techniques to emphasise different elements of their story. For example, they use an Area Chart in order show the worlds total population grow over time.  When you're visualization your data, keep in mind that you may need separate visual representations for the different types of data you wish to share. 


Click on the Image below top view the video. 
Visualizing-How-A-Population-Grows-To-7-Billion--NPR-png.jpg


Announcing the new GeoTme(s) Blog

14/11/2011 3:05:32 PM
by Global Administrator
Today we are happy to launch the GeoTime(s) Blog, a new resource for all things related to movement analysis, data presentation and visual analytics. We will be using our new blog as place to share analysis techniques, customer stories, research work and interesting articles. It will be a collaborative effort between the GeoTime Team and our users, who are always coming up with new ways to use visual analysis tools to analyze and present complex data. As we are a company obsessed with all things visual, we will no doubt be sharing a lots of video content with you here on our new blog.