Profile cover photo
Profile photo
Micah Norman
199 followers
199 followers
About
Micah's posts

Post has attachment
After a breif break from classes, but not a break from R, I have started #5 and #6.  I sent the below message to all members in those classes. Everyone in this group should be a moderator.  If you notice any new requests come in, please approve and promote the requester.

I would like to get this group started up.  Any interest in a study session this week?

-------------------------------Begin Message--------------------------------
In the first class we attempted to start a study group here in Chicago that never really got off of the ground.  As the content gets more difficult and expansive, I think it is a good idea to try and restart this group.  If anyone is interested, I have created a google group at https://plus.google.com/communities/108918450770950636710 .  I am not certain about the googleplus group url format, so if the above does not work, the name of the group is " Chicago Coursera Data Science and Kaggle Team ".

Post has attachment
After 27 million ptr lookups (sorry 4.2.2.2) the below are the top 15 responding spam sources from spamhouse.
Photo

Post has attachment
who ever said data scientists have no sense of humor! 

Post has attachment
Yesterday's alien alienvault world distribution of ip blacklists entries.
Photo

Post has attachment
From yesterday's spamhouse blocklist, the top twenty domain senders of spam. 
Photo

Post has attachment
Photo

Post has attachment
US PM25 numbers split into 5 distinct groups by color based on quantity.

url<-"http://www.epa.gov/ttn/airs/airsaqs/detaildata/501files/RD_501_88101_2012.zip"

filename<-unlist(strsplit(url,'/'))[NROW(unlist(strsplit(url,'/')))]
fname<-paste0(datafolder,filename)
if(!file.exists(fname)){

download.file(url,paste0(datafolder,filename),method='curl' )
}
#unzip  and change fname variable
fname<-'c:\\location\\of\\data.csv
pollution<-read.csv(fname, colClasses=c('numeric','character','factor','numeric','numeric'))

pollution$pm25group<-cut(pollution$pm25, breaks=quantile(pollution$pm25, probs=c(0.00,0.25,0.50,0.75,1.00)))

plot(y=pollution$latitude, x=pollution$longitude, col=pollution$pm25group, cex=.4, pch=20)
Photo

Post has attachment
As a Chicago native in the data science certificate, I pulled and cleaned all Chicago crime data Sive 1/1/2001 and created the following map. Each point is a crime (red for violent, green for drug, blue for sexual) at .005 alpha using ggplot2. The image below is the result along with individual images per crime type. 
PhotoPhotoPhotoPhoto
8/30/14
4 Photos - View album

Post has attachment
PhotoPhotoPhotoPhotoPhoto
2014-07-02
37 Photos - View album

You know, I don't know why I didn't think of this before, but I can host at my office. There is a Starbucks right across the street (hence my drinking problem) and tons of conference room. It is in the west loop at Jefferson and lake with a parking lot depending on the time of day and plenty of street parking. 
Wait while more posts are being loaded