R Elastic MapReduce Hive and Distributed Text Mining Toolbox.... its a whirlpool

After having learned about MapReduce for solving a computer vision problem for my recent internship at Google, I felt like an enlightened person. I had been using R for text mining already. I wanted to use R for large scale text processing and for experiments with topic modeling. I started reading tutorials and working on some of those. I will now put down my understanding of each of the tools:

1) R text mining toolbox: Meant for local (client side) text processing and it uses the XML library
2) Hive: Hadoop interative, provides the framework to call map/reduce and also provides the DFS interface for storing files on the DFS.
3) RHIPE: R Hadoop integrated environment
4) Elastic MapReduce with R: a MapReduce framework for those who do not have their own clusters
5) Distributed Text Mining with R: An attempt to make seamless move form local to server side processing, from R-tm to R-distributed-tm

I have the following questions and confusions about the above packages

1) Hive and RHIPE and the distributed text mining toolbox need you to have your own clusters. Right?

2) If I have just one computer how would DFS work in case of HIVE

3) Are we facing with the problem of duplication of effort with the above packages?

I am hoping to get insights on the above questions in the next few days

Wireless troubles on Ubuntu

I had been struggling with wireless issues on Ubuntu 10.04 on my new dell studio laptop. I tried rfkill but all in vain. The wireless was disabled and hard-blocked. I was unable to enable it. I have now finally found a fix, thanks to this website.

The fix is to use this command
rmmod dell_laptop

Parsing custom XML files using R

I have been trying to learn R, especially because I need to start using the text mining package in R to do loads of analysis on text data. A couple of datasets are already indexed in R but I wanted to index my own datasets written in custom XML formats. The tutorial is very useful. And I was able to get some success by writing my own XMLreaders and XMLsources. The end goal is to have a corpus object (say myCorpus) that has "PlainText Documents" so if you type the following command

> class(myCorpus)
TextDocument "character"

The XMLsource is something that reads all the files and extracts the XMLNode list which is then processed by XMLreader (that works on each XMLNode in the list) to return a List of "TextDocument"

Here is the custom XML file

library(XML);
library(tm);
mySource <- function(x, encoding = "UTF-8") XMLSource(x, function(tree) ;
xmlChildren(tree[1]]$children$corpus) myXMLReader, encoding);
myXMLReader <- readXML( spec = list(Content = list("node", "/DOC"), id = list("node", "/DOC/DOCNO")), doc = PlainTextDocument());
myCorpus <- Corpus(mySource("/home/shivani/research/mycode/R/query.xml"));

class(mycorpus[[1]]) The things that I did not like about this method of created a XMLSource is that it is extremely tailor made. See xmlChildren(tree[[1]]$children$corpus. This was done by trail and error, trying different functions till you find one that gives you a XMLNode list in return. Also, even after all this monkey-ing around I have not been able to get rid of the docid in my parsed document. Since there is a mix of data in the finally extracted documents.

If anybody could suggest a way to extract the text in the document without extracting a document number that would be great.

<?xml version="1.0"?>
<corpus>
<DOC>
<DOCNO>1</DOCNO>
emission obama dioxide
</DOC>
<DOC>
<DOCNO>2</DOCNO>
wall street market stock dollar
</DOC>
<DOC>
<DOCNO>3</DOCNO>
global greenhouse pollutants greenhouse dioxide
</DOC>
<DOC>
<DOCNO>4</DOCNO>
soldier field combat field
</DOC>
<DOC>
<DOCNO>5</DOCNO>
student student community
</DOC>
<DOC>
<DOCNO>6</DOCNO>
emission obama battle obama afghanistan field obama soldier soldier volumes emission
</DOC>
<DOC>
<DOCNO>7</DOCNO>
afghanistan traders field dollar iraq dollar iraq
</DOC>
<DOC>
<DOCNO>8</DOCNO>
emission obama soldier warming greenhouse obama carbon
</DOC>
<DOC>
<DOCNO>9</DOCNO>
dollar market dollar america
</DOC>
<DOC>
<DOCNO>10</DOCNO>
global dioxide
</DOC>
</corpus>

Dreams of summer

So I got accepted to Google Inc Mountainview CA for a summer internship. Not only am I excited but I am also encouraged. My last claim to fame happened in Aug 2008. The long dry spell was killing me. Although, this means a lot of work it also means I am worth it :)

I was also thinking about the things I need to occupy my summer with and here they are

1. Google Internship: Do a good job at it
2. Develop C/C++ version of the Gibbs Sampler and look at implementation issues in a large-large database
3. Look into usability of MapReduce for Unigram/VSM/LDA-combined representation to come up with a really neat representation
4. Push out the IEEE-T-SE paper on comparitive study ... May 18
5. Push out the IEEE-T-KDE paper on Unigram + LDA ... May 18
6. Look at my Lemur-toolkit extensions and get back in touch with that code
7. Develop a GUI for retrieval engine and document existing matlab/c/c++ code

Cant wait for life to begin!!!

Perl: Detecting two words concatenated into one sepearted by uppercase character

If you are given a word 'textArea' and want to be able to separate them into 'text Area', then here is the perl code to help you achieve it

   $item = 'textAreaBuffer';
   if($item =~ m/.*?[a-z].*?[A-Z]/){
        $count=0;
        @rem=();
        while($item =~ m/([a-z][A-Z])/g){
         $rem[$count]=pos($item);$count=$count+1;
        
        }
       
        for($count=0;$count<@rem;$count++){          
           if($count==0){           
                print FILE2 substr($item,0,$rem[$count]-1);           
                print FILE2 " ";            }
           else{           
                print FILE2 substr($item,$rem[$count-1]-1,$rem[$count]-$rem[$count-1]);           
                print FILE2 " ";          
                 }
           }
       print FILE2 substr($item,$rem[@rem-1]-1);         
       print FILE2 " ";
 } 

Vocabulary handling for software reuse

Vocabulary mismatch is even more of a pronounced problem when dealing with corpus made of source code. There are several tools used in the past and here are some of them that I will need to research and learn about

1) Soundex: Phonetic variations of the words are captured
2) Lexical affinity: A sliding window (grep-like) tool that calculates how close are two identifier names
3) Separate words using "_": counter_activity -> counter activity
4) Separate words using case sensitivity: LoadBuffer -> Load buffer

One could additionally use WORDNET and spelling correction or missing character handling to improve upon the existing techniques

Combining discriminative and Generative training

This blog is a continuation of the previous post Targeting discrimination as opposed to similarity. In model learning, given labeled data there is always a choice to make, discriminative or generative. There are pros and cons of either methods and usually a two-stage approach is used to combine the power of the two. I found a very recent paper, Multi-Conditional Learning published in AAAI 2006, on combining the power of the two at the time of learning. The likelihood function is now composed of the generative likelihood and the discriminative likelihood. This is possible only because we have labeled data. The discriminative-generative pair used for illustration is primarily the CRF-MRF. The MRF is used to capture topic co-occurrences. The word co-occurrences are captured by the latent topic model. However, by using a MCL regularizer (discriminative component) in the learning algorithm, the model being learnt is actually a combination of MRF and CRF, thereby combining the power of the two.The authors claim huge performance gains in document retrieval on the 20newsgroups dataset.