When Will Data.gov be Useful?

Posted by on Oct 27, 2014 in News | 2 comments

Data.gov is the top level data search system for the US, with references to over 130,000 datasets from federal and state agencies. And yet, I’ve never successfully used it for finding data. Here is an example search for “Diabetes Rates”:

Search for “Diabetes Rates” on Data.gov

 

So, we look for diabetes, and get births, 22 year old mortality data from the US Geological Survey, and quality of service data as the first three hits. The first link at least points to the right agency, but you still have to click three times to get there.

Here is the same search on Google:

 

Search for Diabetes Rates on Google.

Search for Diabetes Rates on Google.

 

Not only do I get links to real primary organizations, the fourth hit is the original source of the data, and Google helpfully gives us quick stats before the hits. Even better,  if you search for “diabetes rates data” you are one click away from the primary data source for US diabetes rates.

The really poor quality of search results on Data.gov has been a problem for its entire existence; I’ve never done a data search on Data.gov that returned what I wanted. I’ve always had better results with Google or browsing the website of the agency that produces the data.

Data.gov has been around for about 5 years, and despite human curation is still isn’t as useful as Google’s automatic index.  At some point, I’d like to stop being excited by its possibilities, and start being excited by it utility.

 

 

Read More

Learn Data Analysis Techniques

Posted by on Jun 5, 2014 in News | 0 comments

Once you’ve got the basic skills in programming and statistics, the best way to learn data analysis is to do it. So, we’re developing a practical experience program for aspiring data analysts. The program is in the pilot phase with a small set of students, but you can read about how it works on the Internship Program’s wiki page.

The goal of the program is to develop more experience with answering questions with data in San Diego and to make that experience available to non profits, government agencies, journalists and other organizations that have questions that could be answered with data, but don’t have the time or skills to do it.

We’re currently recruiting participants, mentors and projects. So, if you’d like to develop data skills, teach data skills, or have a data project, let us know.

 

 

Read More

The Cost of Cleaning

Posted by on May 7, 2014 in Commentary | 0 comments

We’ve frequently mentioned that people who work on data projects tell us that frequently, 80% of their projects are consumed by data preparation and cleaning, so it is interesting to get this data point from Kaggle:

(2) How long is a typical project?
When working with a top 0.5% data scientist, projects take just eight to 40 hours ($3k to $12k). Projects are finished in closer to eight hours for clean data and closer to 40 hours when the data requires cleaning.

So, in this anecdote, with some squinty-eyed interpretation,  data cleaning requires 32 out of 40 hours. Dead on. And, by the way, that’s 32 hours at $300 per hour.

Fortunately, the library has a plan to reduce the cost of data cleaning and preparation.

 

Read More

Burglary Rhythm Maps

Posted by on Apr 18, 2014 in Analysis | 0 comments

A Rhythm Map is a heat map that displays time in the X and Y dimensions. They are an excellent way to visualize repeating patterns in time, such as how crimes occur by hour and data of week. Here we look at some interesting patterns in burglaries in the City of San Diego.

First, here is the map for a range of crime types in San Diego, compiled from the type, time and date of about 400K crime incidents in the City of San Diego from 2006 to 2012.

Rhythm Map, All crimes in San Diego

 

Each square is a crime type. The vertical axis is the hour of the day, and the  horizontal axis is the day of the week, with Sunday being the cell between 0 and 1.  Darker red means there are more crimes than lighter red and yellow. The colors are not comparable across squares, only within the cell. So, the dark red cell at 5:00PM on Friday in the Burglary square may represent a very different number of crime incidents than then dark red cell at 12:00AM on Thursday in Sex Crimes.  Also note that these views combine citations, arrests and reported crimes, and there may be different patters when the maps are broken out on that factor.

There are a lot of interesting patterns here, but we’ll focus on Burglary. The first thing to notice is there are two time ranges, groups of darker red cells,  when burglaries occur: during the work hours on weekdays and on Friday evenings. ( The strong line at noon is most likely an artifact of crimes for which the time is not known being given that value arbitrarily.  )

What accounts for the two separate time ranges? First, let’s break it out by community. This chart uses Clarinova Place Codes for the community names.

 

Here we see that some communities exhibit one pattern or the other, and sometimes both. Downtown ( SanDOW ),  La Jolla ( SanLAJ ) and Mira Mesa ( SanMIR ) show the Friday pattern, while Southeastern ( SanSOT),  Greater North Park ( SanGRE) and Midtown ( SanMID ) show the week day pattern.

Community distinctions may explain some of the differences in the patterns, but there is a factor that is probably more important: residential vs commercial crime. So, let’s split out the maps on that factor.

Here is where the distinctions become the strongest. In Otay Mesa ( SanOAT ), Mira Mesa ( SanMIR ) University ( SanUNV ) and others, the Friday evening pattern completely splits from the weekday pattern. However, we also see a new weekday pattern in the commercial burglaries in Claremont ( SanCLA ), Uptown, Midtown, with commercial burglaries occurring across the weekday evenings.

Those features are consistent with exactly what you’d expect from burglary: the burglaries occur when the business and homes are unoccupied. But it doesn’t explain why in many communities the commercial crimes would occur more frequently on Friday evenings.  Another unusual pattern is that in Pacific Beach ( SanPCF ) there is a residential burglary cluster on Friday and Saturday evenings, with a similar but weaker pattern occurring in Uptown and College.

Rhythms are a powerful way to look for patterns in time-structured data, because they take advantage of the ways that human brains most quickly process visual information. However,  they aren’t  a complete solution; they are just a start. Before making any recommendations based on the data, we’d want to do a few statistical tests, and at least, look at the absolute number of incidents per cell in the areas exhibiting patterns.

 

Read More