Note this is what I was trying to write in previous post. Also credit is due to http://lixiaoxu.lxxm.com/rwebfriend/
Month: May 2009
Spreading Content using Social Media
The concept of social media is different for different people- some view it as a revolution enabling people to communicate world wide, others view it as a new marketing media, while some of us wonder what the big fuss is all about.
Here are some ways to spread content/marketing communication using Social Media-
1) RSS – Really Simple Syndication has effectively proven a challenge to content creators as well as search engines.Using it effectively can make it a very nice and user friendly way to spread the words you write.
- Newsletters- I use both Feedblitz ( for historical reasons) and my default feed is Google Feedburner. Google feedburner is useful especially for content leaching and security. I also like to track how people read the newsletter and Feddblitz is very good in that, but if the RSS feed is used by a RSS reader then Feedburner does a better job of tracking IMO.Note Linkedin now has a good feature for Announcements that does the same thing.
- Linkedin News – In the Linkedin Group created for your Blog ( if not, do it.now) There is an option for News – add your RSS feed there for people/visitors to read it.
- Facebook- Use the Facebook application called Facebook Notes for spreading your content to all your connections everytime you wrte a post.
- Linkedin WordPress Application- This updates your Linkedin profile for latest posts. Very useful for link seekers like me who have 8000+ Linkedin connections ( Note I have got 4 contracts and multiple nowledge opportunities through the links there- In a hyper Linked world every one needs as many links they can get)
2) Thid Party Sites- Create a Page to Drop Links or Create a Community to Exchange Comments/Views
- Create a Facebook Page for your site and also add similar pages for Favourites- Like a FB analogy for Blogroll( Example –
Join ours at http://www.facebook.com/pages/DecisionStats/191421035186
- Create a Linkedin Group ( if not done so already)
Join Decisionstats at http://www.linkedin.com/groups?gid=54257
- Create a Community on either of these ( Ning.com, Self Hosted- Wordframe.com or Buddypress.org, Yahoo Bloglog or Google Friend Connect, Facebook Friend Connect)
- The biggest analytics community on Ning.com is www.analyticbridge.com
- The biggest data specific community on a Wordframe Instal is www.smartdatacollective.com
- Join me on Yahoo MyBlogLog at http://www.mybloglog.com/buzz/community/decisionstats/
- Join me on Google Connect on the right margin
- I am yet to test out Facebook Connect
3) Twitter- Use the RSS in Step 1 ( Preferably use Feedburner than the raw xml feed to help you track) and add it to www.twitterfeed.com for automated posting of your content.
4) Use Customized Widgets- I recommend the one at Add to Any WordPress Plugin to help your readers share it on – and you CAN tweak the colors here –
http://www.addtoany.com/buttons/share_save/menu_style/wordpress
5) WordPress- The latest version of WordPress is available at both Simple Scripts ( for Self Hosted) and at www.wordpress.org
Use it for a better looking website/blog. Using custom themes and Akismet with Bad Behaviour plugins guarantees security. For more tips on WordPress security see www.dullest.com
6) Use it or Lose it-
- Using Facebook has helped me maintain contacts with SCHOOL Friends
- Linkedin has helped me like n Online Rolodex
- Twitter – I am still trying to figure it out but it is fun.
Hope that helps. Have a great Memorial Day weekend in the US- those vets have helped the world stay free in Europe, Africa and Asia as well
Ubuntu Analytics for Beginners (like me)
So I finally gave in, said Goodbye to my Windows with it’s virus warnings, viruses, anti viruses and security updates and got an Ubuntu.
It is like learning to speak a new language at the age of >30.
Here are some tips for configuring Internet on Ubuntu and Configuring R on Ubuntu
Using R from Ubuntu
Finding All packages
Go to Applications -Accessories-Terminal
Type the following
$apt-cache search r-cran-
To install any package simply
Type
sudo apt-get install r-cran-rcmdr r-cran-xml

After this to get other packages not listed here, invoke R by typing R at the prompt
Load installed packages by Command
>library(packagename)
Configuring Internet (specific to Indian Broadband) on an Ubuntu Laptop
Connect the Internet Wire and Switch on the ADSL modem
Go to Applications -Accesories-Terminal
Type the following at Terminal ( Note the $ sign represents Terminal- so type only ifconfig in following command)
$ ifconfig
Check the IP address from this value ( among a big dump of stuff you get by typing ifconfig)
inet addr:XXX.XX.XX.XX
Go to System- Administration- Network
Click Wired Connection- Unlock ( Use the admin password given to you- same as your machine login password)
Select Wired Connection -Properties
Configuration to Automatic Configuration (DHCP)
Go to Host
Select username-laptop Properties
Change Ip address to one gotten above
Close
(To be continued)
For Enthusiastic Beginners in Ubuntu – Note I use the Ubuntu Long Term version 8.10. The latest version is 9.04 but it is not a Long Term Version Yet. Also recommended is the Cannonical Shop for Ubuntu Merchandise (see a free screenshot in the post)
Google Custom Search
Here is a revised version of the Custom Search Engine that I first talked of last year- this year it now includes Business Intelligence Sites.
Try it out and let me know of you want to help create a customized Data Mining Engine- Note it already has 800 plus analytics and Business Intelligence Sites.
I got much better results than Google when searching for R, but thats to be expected 🙂
Building KXEN Models on Ubuntu
Doing analytics on Linux sometimes seems user unfriendly but the reality is itis not so- and it is actually cheaper for you as you can focus on the analytical software rather than the Operating System licensing costs.
![]()

![]()
Note: The software used in this were KXEN Linux version 2.4 and Ubuntu Hardy Heron.
Using KXEN on an Ubuntu Linux proved surprisingly even more easy. Thanks to some excellent help provided by the KXEN support team and some discussions with KXEN’s head of Research,Bertrand the following 5 step procedure should help you start building models in KXEN right away.
Using Ubuntu has the added advantages of security, low costs as well as all the ease of a Graphical User Interface-
1)Backward Compatability
$ sudo apt-get install libstdc++5
(it will then ask the password).
2)Installing Java
$ sudo apt-get install sun-java6-jre
3) Download and Unzipping the Software
Download the zipped folder from the KXEN Download Site.
Unzip the Linux Version of KXEN system from the Download Site- This creates the master folder (example Kxen_X86-Linux-2.4.21-4.Elsmp_v5_0_3 )
4)Licensing
Run the KXEN Node Generator in KxNodeCodeGenerator folder within the master folder above.
The new file KXEN Node.txt is then sent back to support team and they send the License_nl.cfg
5)Installation ( for stand alone client)
Install a JVM 1.4.2 and export the java/bin directory in the $PATH environment variable.
The exe is located in the folder KJWizardJNI- The following commands
$ cd KJWizardJNI
$ PATH=/opt/j2sdk1.4.2_10/bin:$PATH ; export PATH
$ ./KJWizardJNI.sh
Run KXEN Models Happily ever after !!!
Note KXEN offers the ability to export models in a variety of formats including PMML, SAS, SQL and other languages
Disclaimer- I am a consultant on social media to KXEN
Building KXEN Models on Ubuntu
Doing analytics on Linux sometimes seems user unfriendly but the reality is itis not so- and it is actually cheaper for you as you can focus on the analytical software rather than the Operating System licensing costs.
![]()

![]()
Note: The software used in this were KXEN Linux version 2.4 and Ubuntu Hardy Heron.
Using KXEN on an Ubuntu Linux proved surprisingly even more easy. Thanks to some excellent help provided by the KXEN support team and some discussions with KXEN’s head of Research,Bertrand the following 5 step procedure should help you start building models in KXEN right away.
Using Ubuntu has the added advantages of security, low costs as well as all the ease of a Graphical User Interface-
1)Backward Compatibility
$ sudo apt-get install libstdc++5
(it will then ask the password).
2)Installing Java
$ sudo apt-get install sun-java6-jre
3) Download and Unzipping the Software
Download the zipped folder from the KXEN Download Site.
Unzip the Linux Version of KXEN system from the Download Site- This creates the master folder (example Kxen_X86-Linux-2.4.21-4.Elsmp_v5_0_3 )
4)Licensing
Run the KXEN Node Generator in KxNodeCodeGenerator folder within the master folder above.
The new file KXEN Node.txt is then sent back to support team and they send the License_nl.cfg
5)Installation ( for stand alone client)
Install a JVM 1.4.2 and export the java/bin directory in the $PATH environment variable.
The exe is located in the folder KJWizardJNI- The following commands
$ cd KJWizardJNI
$ PATH=/opt/j2sdk1.4.2_10/bin:$PATH ; export PATH
$ ./KJWizardJNI.sh
Run KXEN Models Happily ever after !!!
Note KXEN offers the ability to export models in a variety of formats including PMML, SAS, SQL and other languages
Disclaimer- I am a consultant on social media to KXEN
Google Custom Search
Here is a revised version of the Custom Search Engine that I first talked of last year- this year it now includes Business Intelligence Sites.
Try it out and let me know of you want to help create a customized Data Mining Engine- Note it already has 800 plus analytics and Business Intelligence Sites.
I got much better results than Google when searching for R, but thats to be expected 🙂