2009 August | nandeshwar.info

Tag Cloud of Data Mining Jobs

August 20th, 2009 No comments

Here’s what I did to get a cool looking tag cloud of data mining jobs:

Used Yahoo Pipes (I created mine, but this one has more feeds)– this pipe aggregates feeds from different job web-sites, and gives the user unique job listing that you can subscribe via RSS: Job Feed Aggregator by Sean Dolan
Subscribed to the RSS feed for the keyword “data mining”
Copied the job descriptions and requirements of many jobs, and saved the text file
Got the python stemmer
Applied the python stemmer to the text file. Stemmer truncates words to their roots, so that we can combine variants of a word into a single word. (First or second step in text mining)
Created a tag cloud using the services of http://www.wordle.net/ . They use “stop words,” so I didn’t have to apply those. Stop words are common words, which necessarily don’t add any value for categorization, of a language.

Data Mining Jobs Tag Cloud

The most frequent word is: experience. Companies want people with experience in different data mining techniques. You’ll see that some other big words are: SAS (stemmed as sa), Excel, SQL, analytical skills, statistics, and quantitative skills.

And how do you master these skills, you ask?

Get a graduate degree in statistics, economics, mathematics, computer science, financial engineering, or industrial engineering with emphasis on databases, data mining, and marketing.
Successfully complete data mining projects using free, open-source data mining tools, such as Weka, R, Orange, Rapid-Miner.
Participate in data mining competitions. SAS’s data mining conference has a data mining competition every year.

Have a look at a detailed study by Pejic Bach, M: Creating profile of data mining specialist