Tag: Document

Text Mining Barack Obama using R #rstats

  • We copy and paste President Barack Obama’s “Yes We Can” speech in a text document and read it in. For a word cloud we need a dataframe with two columns, one with words and the the other with frequency.We read in the transcript from http://www.nytimes.com/2008/01/08/us/politics/08text-obama.html?pagewanted=all&_r=0  and paste in the file located in the local directory- /home/ajay/Desktop/new. Note tm is a powerful package and will read ALL the text documents within the particular folder

library(tm)

library(wordcloud)

txt2=”/home/ajay/Desktop/new”

b=Corpus(DirSource(txt2), readerControl = list(language = “eng”))

> b b b tdm m1 v1 d1 wordcloud(d1$word,d1$freq)

Now it seems we need to remove some of the very commonly occuring words like “the” and “and”. We are not using the standard stopwords in english (the tm package provides that see Chapter 13 Text Mining case studies), as the words “we” and “can” are also included .

> b tdm m1 v1 d1 wordcloud(d1$word,d1$freq)

But let’s see how the wordcloud changes if we remove all English Stopwords.

> b tdm m1 v1 d1 wordcloud(d1$word,d1$freq)

and you can draw your own conclusions from the content of this famous speech based on your political preferences.

Politicians can give interesting speeches but they may be full of simple sounding words…..

Citation-

1. Ingo Feinerer (2012). tm: Text Mining Package. R package version0.5-7.1.

Ingo Feinerer, Kurt Hornik, and David Meyer (2008). Text Mining
Infrastructure in R. Journal of Statistical Software 25/5. URL:
http://www.jstatsoft.org/v25/i05/

2. Ian Fellows (2012). wordcloud: Word Clouds. R package version 2.0.

http://CRAN.R-project.org/package=wordcloud

3. You can see more than 100 of Obama’s speeches at http://obamaspeeches.com/

Quote- numbers dont lie, people do.

.

Cyber Cold War

I try to write on cyber conflict without getting into the politics of why someone is hacking someone else. I always get beaten by someone in the comments thread when I write on politics.

But recent events have forced me to update my usual “how-to” cyber conflict to “why” cyber conflict. This is because of a terrorist attack in my hometown Delhi.

(updated-

http://www.nytimes.com/2012/02/14/world/middleeast/israeli-embassy-officials-attacked-in-india-and-georgia.html?_r=1&hp

Iran allegedly tried  (as per Israel) to assassinate the wife of Israeli Defence Attache in Delhi using a magnetic bomb, India as she went to school to pick up her kids, somebody else put a grenade in Israeli embassy car in Georgia which was found in time. 

Based on reports , initial work suggests the bomb was much more sophisticated than local terrorists, but the terrorists seemed to have some local recce work done.

India has 0 history of antisemitism but this is the second time Israelis have been targeted since 26/11 Mumbai attacks. India buys 12 % of oil annually from Iran (and refuses to join the oil embargo called by US and Europe)

Cyber Conflict is less painful than conflict, which is inevitable as long as mankind exists. Also the Western hemisphere needs a moon shot (cyber conflict could be the Sputnik like moment) and with declining and aging populations but better technology, Western Hemisphere govts need cyber conflict as they are running out of humans to fight their wars. Eastern govt. are even more obnoxious in using children for conflict propaganda, and corruption.

Last week CIA.gov website went down

This week Iranian govt is allegedly blocking https traffic on eve of Annual Revolution Day (what a coincidence!)

 

Some resources to help Internet users in Iran (or maybe this could be a dummy test for the big one – hacking the great firewall of China)

News from Hacker News-

http://news.ycombinator.com/item?id=3575029

 

I’m writing this to report the serious troubles we have regarding accessing Internet in Iran at the moment. Since Thursday Iranian government has shutted down the https protocol which has caused almost all google services (gmail, and google.com itself) to become inaccessible. Almost all websites that reply on Google APIs (like wolfram alpha) won’t work. Accessing to any website that replies on https (just imaging how many websites use this protocol, from Arch Wiki to bank websites). Also accessing many proxies is also impossible. There are almost no official reports on this and with many websites and my email accounts restricted I can just confirm this based on my own and friends experience. I have just found one report here:

Iran Shut Down Gmail , Google , Yahoo and sites using “Https” Protocol

The reason for this horrible shutdown is that the Iranian regime celebrates 1979 Islamic revolution tomorrow.

I just wanted to let you guys know about this. If you have any solution regarding bypassing this restriction please help!

 

The boys at Tor think they can help-

but its not so elegant, as I prefer creating a  batch file rather than explain coding to newbies. 

this is still getting to better and easier interfaces

https://www.torproject.org/projects/obfsproxy-instructions.html.en

Obfsproxy Instructions

client torrc

Step 1: Install dependencies, obfsproxy, and Tor

 

You will need a C compiler (gcc), the autoconf and autotools build system, the git revision control system, pkg-config andlibtool, libevent-2 and its headers, and the development headers of OpenSSL.

On Debian testing or Ubuntu oneiric, you could do:
# apt-get install autoconf autotools-dev gcc git pkg-config libtool libevent-2.0-5 libevent-dev libevent-openssl-2.0-5 libssl-dev

If you’re on a more stable Linux, you can either try our experimental backport libevent2 debs or build libevent2 from source.

Clone obfsproxy from its git repository:
$ git clone https://git.torproject.org/obfsproxy.git
The above command should create and populate a directory named ‘obfsproxy’ in your current directory.

Compile obfsproxy:
$ cd obfsproxy
$ ./autogen.sh && ./configure && make

Optionally, as root install obfsproxy in your system:
# make install

If you prefer not to install obfsproxy as root, you can instead just modify the Transport lines in your torrc file (explained below) to point to your obfsproxy binary.

You will need Tor 0.2.3.11-alpha or later.

Step 2a: If you’re the client…

 

First, you need to learn the address of a bridge that supports obfsproxy. If you don’t know any, try asking a friend to set one up for you. Then the appropriate lines to your tor configuration file:

UseBridges 1
Bridge obfs2 128.31.0.34:1051
ClientTransportPlugin obfs2 exec /usr/local/bin/obfsproxy --managed

Don’t forget to replace 128.31.0.34:1051 with the IP address and port that the bridge’s obfsproxy is listening on.
 Congratulations! Your traffic should now be obfuscated by obfsproxy. You are done! You can now start using Tor.

For old fashioned tunnel creation under Seas of English Channel-

http://dag.wieers.com/howto/ssh-http-tunneling/

Tunneling SSH over HTTP(S)
This document explains how to set up an Apache server and SSH client to allow tunneling SSH over HTTP(S). This can be useful on restricted networks that either firewall everything except HTTP traffic (tcp/80,tcp/443) or require users to use a local (HTTP) proxy.
A lot of people asked why doing it like this if you can just make sshd listen on port 443. Well, that might work if your environment is not hardened like I have seen at several companies, but this setup has a few advantages.

  • You can proxy to anywhere (see the Proxy directive in Apache) based on names
  • You can proxy to any port you like (see the AllowCONNECT directive in Apache)
  • It works even when there is a layer-7 protocol firewall
  • If you enable proxytunnel ssl support, it is indistinguishable from real SSL traffic
  • You can come up with nice hostnames like ‘downloads.yourdomain.com’ and ‘pictures.yourdomain.com’ and for normal users these will look like normal websites when visited.
  • There are many possibilities for doing authentication further along the path
  • You can do proxy-bouncing to the n-th degree to mask where you’re coming from or going to (however this requires more changes to proxytunnel, currently I only added support for one remote proxy)
  • You do not have to dedicate an IP-address for sshd, you can still run an HTTPS site

Related-

http://opensourceandhackystuff.blogspot.in/2012/02/captive-portal-security-part-1.html

and some crypto for young people

http://users.telenet.be/d.rijmenants/en/onetimepad.htm

 

Me- What am I doing about it? I am just writing poems on hacking at http://poemsforkush.com

Topic Models

Some stuff on Topic Models-

http://en.wikipedia.org/wiki/Topic_model

In machine learning and natural language processing, a topic model is a type of statistical model for discovering the abstract “topics” that occur in a collection of documents. An early topic model was probabilistic latent semantic indexing (PLSI), created by Thomas Hofmann in 1999.[1] Latent Dirichlet allocation (LDA), perhaps the most common topic model currently in use, is a generalization of PLSI developed by David Blei, Andrew Ng, and Michael Jordan in 2002, allowing documents to have a mixture of topics.[2] Other topic models are generally extensions on LDA, such as Pachinko allocation, which improves on LDA by modeling correlations between topics in addition to the word correlations which constitute topics. Although topic models were first described and implemented in the context of natural language processing, they have applications in other fields such as bioinformatics.

http://en.wikipedia.org/wiki/Latent_Dirichlet_allocation

In statistics, latent Dirichlet allocation (LDA) is a generative model that allows sets of observations to be explained by unobserved groups that explain why some parts of the data are similar. For example, if observations are words collected into documents, it posits that each document is a mixture of a small number of topics and that each word’s creation is attributable to one of the document’s topics. LDA is an example of a topic model

David M Blei’s page on Topic Models-

http://www.cs.princeton.edu/~blei/topicmodeling.html

The topic models mailing list is a good forum for discussing topic modeling.

In R,

Some resources I compiled on Slideshare based on the above- Continue reading “Topic Models”

Data Documentation Initiative

Here is a nice initiative in standardizing data documentation for social sciences (which can be quite a relief to legions of analysts)

http://www.ddialliance.org/what

 

 

 

 

Benefits of DDI

The DDI facilitates:

  • Interoperability. Codebooks marked up using the DDI specification can be exchanged and transported seamlessly, and applications can be written to work with these homogeneous documents.
  • Richer content. The DDI was designed to encourage the use of a comprehensive set of elements to describe social science datasets as completely and as thoroughly as possible, thereby providing the potential data analyst with broader knowledge about a given collection.
  • Single document – multiple purposes. A DDI codebook contains all of the information necessary to produce several different types of output, including, for example, a traditional social science codebook, a bibliographic record, or SAS/SPSS/Stata data definition statements. Thus, the document may be repurposed for different needs and applications. Changes made to the core document will be passed along to any output generated.
  • On-line subsetting and analysis. Because the DDI markup extends down to the variable level and provides a standard uniform structure and content for variables, DDI documents are easily imported into on-line analysis systems, rendering datasets more readily usable for a wider audience.
  • Precision in searching. Since each of the elements in a DDI-compliant codebook is tagged in a specific way, field-specific searches across documents and studies are enabled. For example, a library of DDI codebooks could be searched to identify datasets covering protest demonstrations during the 1960s in specific states or countries.
Also see-
  1. http://www.ddialliance.org/Specification/DDI-Codebook/2.1/DTD/Documentation/DDI2-1-tree.html
  2. http://www.ddialliance.org/Specification/DDI-Lifecycle/3.1/

 

#Rstats Credit Scoring using R

I came across a nice, lucid and very readable document at the http://cran.r-project.org/doc/contrib/Sharma-CreditScoring.pdf

Credit Scoring is really a bread and butter activity at many analytics shopfloors, and I really liked the way Credit Scoring is explained and executed by the author- which can be used by any user regardless of experience.
Sharma-CreditScoringhttp://www.scribd.com/embeds/74139509/content?start_page=1&view_mode=list&access_key=key-ttkkmxe3hkmq3ic746c//

 

LibreOffice – Extensions and Templates

Just an announcement from The Document Foundation (which has notable supporters including Google etc at http://www.documentfoundation.org/supporters/)

With both Google Docs and Libre Office – it seems like a flank attack on Office productivity software (from the cloud and from the PC/tablet ground)- however Microsoft’s Sharepoint is much better in collobration compared to the Google Docs and it has huge number of templates (more than the 38 extensions and 13 templates right now at the links below (just like WordPress has huge number of themes compared to Blogger)

Anyways, check out- it is an interesting start

http://extensions.libreoffice.org/

Extension Releases 

Extensions for all program modules
Gallery Contents for all program modules
Language Tools for all program modules
Dictionaries of different languages for all program modules
Writer-Extensions
Calc-Extensions
Impress-Extensions
Draw-Extensions
Base-Extensions
Math-Extensions 

….

and http://templates.libreoffice.org/

Template Releases

Accounting -Templates
Agenda-Templates
Arts-Templates
Book-Templates
Brochure/Pamphlet-Templates
Budget-Templates
Business-Templates
Business POS-Templates
Business Shipping-Templates
Calendar-Templates
Card-Templates
Curriculum/Resume-Templates
CD/DVD-Templates
Certificate-Templates
Checkbook-Templates
Christmas-Templates
Computer-Templates
Conference-Templates
E-book-Templates
Education-Templates
Academia-Templates
Elementary/Secondary School-Templates
Envelope-Templates
Fax-Templates
Genealogy-Templates
Grocery-Templates
Invoice-Templates
Labels-Templates
Letter-Templates
Magazine-Templates
Media-Templates
Memo-Templates
Music-Templates
Newsletter-Templates
Notes-Templates
Paper-Templates
Presentation-Templates
Recipe-Templates
Science-Templates
Sports-Templates
Timeline-Templates
Timesheet-Templates
Trades-Templates
To Do List-Templates
Writer-Templates

 

The all new Blogging in Blogger

I had given up on Blogspot ever having a makeover in favor of the nice themes at

wordpress, but man, the new CEO at google is really shaking some stuff here.

Check out the nice features for customizing the themes at Blogspot

Continue reading “The all new Blogging in Blogger”