Tag: Analytics

Analytics 2012 Conference

A nice conference from the grand old institution of Analytics,  SAS  Institute’s annual analytic pow-wow.

I especially like some of the trainings- and wonder if they could be stored as e-learning modules for students/academics to review

in SAS’s extensive and generous Online Education Program.

Sunday Morning Workshop

SAS Sentiment Analysis Studio: Introduction to Building Models

This course provides an introduction to SAS Sentiment Analysis Studio. It is designed for system designers, developers, analytical consultants and managers who want to understand techniques and approaches for identifying sentiment in textual documents.
View outline
Sunday, Oct. 7, 8:30a.m.-12p.m. – $250

Sunday Afternoon Workshops

Business Analytics Consulting Workshops

This workshop is designed for the analyst, statistician, or executive who wants to discuss best-practice approaches to solving specific business problems, in the context of analytics. The two-hour workshop will be customized to discuss your specific analytical needs and will be designed as a one-on-one session for you, including up to five individuals within your company sharing your analytical goal. This workshop is specifically geared for an expert tasked with solving a critical business problem who needs consultation for developing the analytical approach required. The workshop can be customized to meet your needs, from a deep-dive into modeling methods to a strategic plan for analytic initiatives. In addition to the two hours at the conference location, this workshop includes some advanced consulting time over the phone, making it a valuable investment at a bargain price.
View outline
Sunday, Oct. 7; 1-3 p.m. or 3:30-5:30 p.m. – $200

Demand-Driven Forecasting: Sensing Demand Signals, Shaping and Predicting Demand

This half-day lecture teaches students how to integrate demand-driven forecasting into the consensus forecasting process and how to make the current demand forecasting process more demand-driven.
View outline
Sunday, Oct. 7; 1-5 p.m.

Forecast Value Added Analysis

Forecast Value Added (FVA) is the change in a forecasting performance metric (such as MAPE or bias) that can be attributed to a particular step or participant in the forecasting process. FVA analysis is used to identify those process activities that are failing to make the forecast any better (or might even be making it worse). This course provides step-by-step guidelines for conducting FVA analysis – to identify and eliminate the waste, inefficiency, and worst practices from your forecasting process. The result can be better forecasts, with fewer resources and less management time spent on forecasting.
View outline
Sunday, Oct. 7; 1-5 p.m.

SAS Enterprise Content Categorization: An Introduction

This course gives an introduction to methods of unstructured data analysis, document classification and document content identification. The course also uses examples as the basis for constructing parse expressions and resulting entities.
View outline
Sunday, Oct. 7; 1-5 p.m.

 

 
You can see more on this yourself at –

http://www.sas.com/events/analytics/us/

 

 

 

 

 

 

 

 

 

 

 

Software Review- BigML.com – Machine Learning meets the Cloud

I had a chance to dekko the new startup BigML https://bigml.com/ and was suitably impressed by the briefing and my own puttering around the site. Here is my review-

1) The website is very intutively designed- You can create a dataset from an uploaded file in one click and you can create a Decision Tree model in one click as well. I wish other cloud computing websites like  Google Prediction API make design so intutive and easy to understand. Also unlike Google Prediction API, the models are not black box models, but have a description which can be understood.

2) It includes some well known data sources for people trying it out. They were kind enough to offer 5 invite codes for readers of Decisionstats ( if you want to check it yourself, use the codes below the post, note they are one time only , so the first five get the invites.

BigML is still invite only but plan to get into open release soon.

3) Data Sources can only be by uploading files (csv) but they plan to change this hopefully to get data from buckets (s3? or Google?) and from URLs.

4) The one click operation to convert data source into a dataset shows a histogram (distribution) of individual variables.The back end is clojure , because the team explained it made the easiest sense and fit with Java. The good news (?) is you would never see the clojure code at the back end. You can read about it from http://clojure.org/

As cloud computing takes off (someday) I expect clojure popularity to take off as well.

Clojure is a dynamic programming language that targets the Java Virtual Machine (and the CLR, and JavaScript). It is designed to be a general-purpose language, combining the approachability and interactive development of a scripting language with an efficient and robust infrastructure for multithreaded programming. Clojure is a compiled language – it compiles directly to JVM bytecode, yet remains completely dynamic. Every feature supported by Clojure is supported at runtime. Clojure provides easy access to the Java frameworks, with optional type hints and type inference, to ensure that calls to Java can avoid reflection.

Clojure is a dialect of Lisp

 

5) As of now decision trees is the only distributed algol, but they expect to roll out other machine learning stuff soon. Hopefully this includes regression (as logit and linear) and k means clustering. The trees are created and pruned in real time which gives a slightly animated (and impressive effect). and yes model building is an one click operation.

The real time -live pruning is really impressive and I wonder why /how it can ever be replicated in other software based on desktop, because of the sheer interactive nature.

 

Making the model is just half the work. Creating predictions and scoring the model is what is really the money-earner. It is one click and customization is quite intuitive. It is not quite PMML compliant yet so I hope some Zemanta like functionality can be added so huge amounts of models can be applied to predictions or score data in real time.

 

If you are a developer/data hacker, you should check out this section too- it is quite impressive that the designers of BigML have planned for API access so early.

https://bigml.com/developers

BigML.io gives you:

  • Secure programmatic access to all your BigML resources.
  • Fully white-box access to your datasets and models.
  • Asynchronous creation of datasets and models.
  • Near real-time predictions.

 

Note: For your convenience, some of the snippets below include your real username and API key.

Please keep them secret.

REST API

BigML.io conforms to the design principles of Representational State Transfer (REST). BigML.io is enterely HTTP-based.

BigML.io gives you access to four basic resources: Source, Dataset, Model and Prediction. You cancreate, read, update, and delete resources using the respective standard HTTP methods: POST, GET,PUT and DELETE.

All communication with BigML.io is JSON formatted except for source creation. Source creation is handled with a HTTP PUT using the “multipart/form-data” content-type

HTTPS

All access to BigML.io must be performed over HTTPS

and https://bigml.com/developers/quick_start ( In think an R package which uses JSON ,RCurl  would further help in enhancing ease of usage).

 

Summary-

Overall a welcome addition to make software in the real of cloud computing and statistical computation/business analytics both easy to use and easy to deploy with fail safe mechanisms built in.

Check out https://bigml.com/ for yourself to see.

The invite codes are here -one time use only- first five get the invites- so click and try your luck, machine learning on the cloud.

If you dont get an invite (or it is already used, just leave your email there and wait a couple of days to get approval)

  1. https://bigml.com/accounts/register/?code=E1FE7
  2. https://bigml.com/accounts/register/?code=09991
  3. https://bigml.com/accounts/register/?code=5367D
  4. https://bigml.com/accounts/register/?code=76EEF
  5. https://bigml.com/accounts/register/?code=742FD

Facebook Search- The fall of the machines

Increasingly I am beginning to search more and more on Facebook. This is for the following reasons-

1) Facebook is walled off to Google (mostly). While within Facebook , I get both people results and content results (from Bing).

Bing is an okay alternative , though not as fast as Google Instant.

2) Cleaner Web Results When Facebook increases the number of results from 3 top links to say 10 top links, there should be more outbound traffic from FB search to websites.For some reason Google continues to show 14 pages of results… Why? Why not limit to just one page.

3) Better People Search than  Pipl and Google. But not much (or any) image search. This is curious and I am hoping the Instagram results would be added to search results.

4) I am hoping for any company Facebook or Microsoft to challenge Adsense . Adwords already has rivals. Adsense is a de facto monopoly and my experiences in advertising show that content creators can make much more money from a better Adsense (especially ) if Adsense and Adwords do not have a conflict of interest from same advertisers.

Adwords should have been a special case of Adsense for Google.com but it is not.

5) Machine learning can only get you from tau to delta tau. When ad click behavior is inherently dependent on humans who behave mostly on chaotic , or genetic models than linear CPC models. I find FB has an inherent advantage in the quantity and quality of data collected on people behavior rather than click behavior. They are also more aggressive and less apologetic about behavorially targeted  ads.

Additional point- Analytics for Google Analytics is not as rich as analytics from Facebook pages in terms of demographic variables. This can be tested by anyone.

 

send email by R

For automated report delivery I have often used send email options in BASE SAS. For R, for scheduling tasks and sending me automated mails on completion of tasks I have two R options and 1 Windows OS scheduling option. Note red font denotes the parameters that should be changed. Anything else should NOT be changed.

Option 1-

Use the mail package at

http://cran.r-project.org/web/packages/mail/mail.pdf

> library(mail)

Attaching package: ‘mail’

The following object(s) are masked from ‘package:sendmailR’:

sendmail

>
> sendmail(“ohri2007@gmail.com“, subject=”Notification from R“,message=“Calculation finished!”, password=”rmail”)
[1] “Message was sent to ohri2007@gmail.com! You have 19 messages left.”

Disadvantage- Only 20 email messages by IP address per day. (but thats ok!)

Option 2-

use sendmailR package at http://cran.r-project.org/web/packages/sendmailR/sendmailR.pdf

install.packages()
library(sendmailR)
from <- sprintf(“<sendmailR@%s>”, Sys.info()[4])
to <- “<ohri2007@gmail.com>”
subject <- “Hello from R”
body <- list(“It works!”, mime_part(iris))
sendmail(from, to, subject, body,control=list(smtpServer=”ASPMX.L.GOOGLE.COM”))

 

 

BiocInstaller version 1.2.1, ?biocLite for help
> install.packages(“sendmailR”)
Installing package(s) into ‘/home/ubuntu/R/library’
(as ‘lib’ is unspecified)
also installing the dependency ‘base64’

trying URL ‘http://cran.at.r-project.org/src/contrib/base64_1.1.tar.gz&#8217;
Content type ‘application/x-gzip’ length 61109 bytes (59 Kb)
opened URL
==================================================
downloaded 59 Kb

trying URL ‘http://cran.at.r-project.org/src/contrib/sendmailR_1.1-1.tar.gz&#8217;
Content type ‘application/x-gzip’ length 6399 bytes
opened URL
==================================================
downloaded 6399 bytes

BiocInstaller version 1.2.1, ?biocLite for help
* installing *source* package ‘base64’ …
** package ‘base64’ successfully unpacked and MD5 sums checked
** libs
gcc -std=gnu99 -I/usr/local/lib64/R/include -I/usr/local/include -fpic -g -O2 -c base64.c -o base64.o
gcc -std=gnu99 -shared -L/usr/local/lib64 -o base64.so base64.o -L/usr/local/lib64/R/lib -lR
installing to /home/ubuntu/R/library/base64/libs
** R
** preparing package for lazy loading
** help
*** installing help indices
** building package indices …
** testing if installed package can be loaded
BiocInstaller version 1.2.1, ?biocLite for help

* DONE (base64)
BiocInstaller version 1.2.1, ?biocLite for help
* installing *source* package ‘sendmailR’ …
** package ‘sendmailR’ successfully unpacked and MD5 sums checked
** R
** preparing package for lazy loading
** help
*** installing help indices
** building package indices …
** testing if installed package can be loaded
BiocInstaller version 1.2.1, ?biocLite for help

* DONE (sendmailR)

The downloaded packages are in
‘/tmp/RtmpsM222s/downloaded_packages’
> library(sendmailR)
Loading required package: base64
> from <- sprintf(“<sendmailR@%s>”, Sys.info()[4])
> to <- “<ohri2007@gmail.com>”
> subject <- “Hello from R”
> body <- list(“It works!”, mime_part(iris))
> sendmail(from, to, subject, body,
+ control=list(smtpServer=”ASPMX.L.GOOGLE.COM”))
$code
[1] “221”

$msg
[1] “2.0.0 closing connection ff2si17226764qab.40”

Disadvantage-This worked when I used the Amazon Cloud using the BioConductor AMI (for free 2 hours) at http://www.bioconductor.org/help/cloud/

It did NOT work when I tried it use it from my Windows 7 Home Premium PC from my Indian ISP (!!) .

It gave me this error

or in wait_for(250) :
SMTP Error: 5.7.1 [180.215.172.252] The IP you’re using to send mail is not authorized

 

PAUSE–

ps Why do this (send email by R)?

Note you can add either of the two programs of the end of the code that you want to be notified automatically. (like daily tasks)

This is mostly done for repeated business analytics tasks (like reports and analysis that need to be run at specific periods of time)

pps- What else can I do with this?

Can be modified to include sms or tweets  or even blog by email by modifying the   “to”  location appropriately.

3) Using Windows Task Scheduler to run R codes automatically (either the above)

or just sending an email

got to Start>  All Programs > Accessories >System Tools > Task Scheduler ( or by default C:Windowssystem32taskschd.msc)

Create a basic task

Now you can use this to run your daily/or scheduled R code  or you can send yourself email as well.

and modify the parameters- note the SMTP server (you can use the ones for google in example 2 at ASPMX.L.GOOGLE.COM)

and check if it works!

 

Related

 Geeky Things , Bro

–

Configuring IIS on your Windows 7 Home Edition-

note path to do this is-

Control Panel>All Control Panel Items> Program and Features>Turn Windows features on or off> Internet Information Services

and

http://stackoverflow.com/questions/709635/sending-mail-from-batch-file

 

R for Predictive Modeling- PAW Toronto

A nice workshop on using R for Predictive Modeling by Max Kuhn Director, Nonclinical Statistics, Pfizer is on at PAW Toronto.

Workshop

Monday, April 23, 2012 in Toronto
Full-day: 9:00am – 4:30pm

R for Predictive Modeling:
A Hands-On Introduction

Intended Audience: Practitioners who wish to learn how to execute on predictive analytics by way of the R language; anyone who wants “to turn ideas into software, quickly and faithfully.”

Knowledge Level: Either hands-on experience with predictive modeling (without R) or hands-on familiarity with any programming language (other than R) is sufficient background and preparation to participate in this workshop.

What prior attendees have exclaimed

Workshop Description

This one-day session provides a hands-on introduction to R, the well-known open-source platform for data analysis. Real examples are employed in order to methodically expose attendees to best practices driving R and its rich set of predictive modeling packages, providing hands-on experience and know-how. R is compared to other data analysis platforms, and common pitfalls in using R are addressed.

The instructor, a leading R developer and the creator of CARET, a core R package that streamlines the process for creating predictive models, will guide attendees on hands-on execution with R, covering:

  • A working knowledge of the R system
  • The strengths and limitations of the R language
  • Preparing data with R, including splitting, resampling and variable creation
  • Developing predictive models with R, including decision trees, support vector machines and ensemble methods
  • Visualization: Exploratory Data Analysis (EDA), and tools that persuade
  • Evaluating predictive models, including viewing lift curves, variable importance and avoiding overfitting

Hardware: Bring Your Own Laptop
Each workshop participant is required to bring their own laptop running Windows or OS X. The software used during this training program, R, is free and readily available for download.

Attendees receive an electronic copy of the course materials and related R code at the conclusion of the workshop.

Schedule

  • Workshop starts at 9:00am
  • Morning Coffee Break at 10:30am – 11:00am
  • Lunch provided at 12:30 – 1:15pm
  • Afternoon Coffee Break at 2:30pm – 3:00pm
  • End of the Workshop: 4:30pm

Instructor

Max Kuhn, Director, Nonclinical Statistics, Pfizer

Max Kuhn is a Director of Nonclinical Statistics at Pfizer Global R&D in Connecticut. He has been applying models in the pharmaceutical industries for over 15 years.

He is a leading R developer and the author of several R packages including the CARET package that provides a simple and consistent interface to over 100 predictive models available in R.

Mr. Kuhn has taught courses on modeling within Pfizer and externally, including a class for the India Ministry of Information Technology.

Source-

http://www.predictiveanalyticsworld.com/toronto/2012/r_for_predictive_modeling.php

Random Sampling a Dataset in R

A common example in business  analytics data is to take a random sample of a very large dataset, to test your analytics code. Note most business analytics datasets are data.frame ( records as rows and variables as columns)  in structure or database bound.This is partly due to a legacy of traditional analytics software.

Here is how we do it in R-

• Refering to parts of data.frame rather than whole dataset.

Using square brackets to reference variable columns and rows

The notation dataset[i,k] refers to element in the ith row and jth column.

The notation dataset[i,] refers to all elements in the ith row .or a record for a data.frame

The notation dataset[,j] refers to all elements in the jth column- or a variable for a data.frame.

For a data.frame dataset

> nrow(dataset) #This gives number of rows

> ncol(dataset) #This gives number of columns

An example for corelation between only a few variables in a data.frame.

> cor(dataset1[,4:6])

Splitting a dataset into test and control.

ts.test=dataset2[1:200] #First 200 rows

ts.control=dataset2[201:275] #Next 75 rows

• Sampling

Random sampling enables us to work on a smaller size of the whole dataset.

use sample to create a random permutation of the vector x.

Suppose we want to take a 5% sample of a data frame with no replacement.

Let us create a dataset ajay of random numbers

ajay=matrix( round(rnorm(200, 5,15)), ncol=10)

#This is the kind of code line that frightens most MBAs!!

Note we use the round function to round off values.

ajay=as.data.frame(ajay)

 nrow(ajay)

[1] 20

> ncol(ajay)

[1] 10

This is a typical business data scenario when we want to select only a few records to do our analysis (or test our code), but have all the columns for those records. Let  us assume we want to sample only 5% of the whole data so we can run our code on it

Then the number of rows in the new object will be 0.05*nrow(ajay).That will be the size of the sample.

The new object can be referenced to choose only a sample of all rows in original object using the size parameter.

We also use the replace=FALSE or F , to not the same row again and again. The new_rows is thus a 5% sample of the existing rows.

Then using the square backets and ajay[new_rows,] to get-

b=ajay[sample(nrow(ajay),replace=F,size=0.05*nrow(ajay)),]

 

You can change the percentage from 5 % to whatever you want accordingly.

Using Google Analytics API with R:dimensions and metrics

I modified the query I wrote earlier  at http://www.decisionstats.com/using-google-analytics-with-r/to get multiple dimensions and metrics from the Google Analytics API, like hour of day,day of week to get cyclical parameters.We are adding the dimensions, and metrics to bring more depth in our analysis.Basically we are trying to do a time series analysis for forecasting web analytics data( which is basically time -stamped and rich in details ).

Basically I am modifying the dimensions and metrics parameters of the query code using the list at

http://code.google.com/apis/analytics/docs/gdata/dimsmets/dimsmets.html

 

query <- QueryBuilder()
query$Init(start.date = "2011-08-20",
                   end.date = "2012-08-25",
                   dimensions = c("ga:date","ga:hour","ga:dayOfWeek"),
                   metrics = c("ga:visitors","ga:visits","ga:pageviews","ga:timeOnSite"),
                   sort = c("ga:date","ga:hour","ga:dayOfWeek"),
                   table.id = paste(profiles$profile[3,3]))

#5. Make a request to get the data from the API 

ga.data <- ga$GetReportData(query)

#6. Look at the returned data 

str(ga.data)

head(ga.data$data) 




and we need the lubridate package to create a ymd:hour (time stamp)    since GA gives data aggregated at a hourly level at most. Also we need to smoothen the effect of weekend on web analytics data.




#Using package lubridate to convert character dates into time








library(lubridate)
ga.data$data[,1]=ymd(ga.data$data[,1])
ls()
dataset1=ga.data$data
names(dataset1) <- make.names(names(dataset1))
str(dataset1)
head(dataset1)






To be continued-