Comments on: Topic extraction using Wikipedia data

By: knine

knine — Mon, 09 Nov 2009 08:06:26 +0000

Hello..I have some troubles with the conversion of movie to iPhone. I bought an iPhone last week and yesterday when I put a movie onto iPhone and click play, the phone give me a warning that he cannot play video in that format. I think I need to convert the movies to iPhone supported format first. So how can I deal with the conversion? Can any one recommend a powerful and free converter for me ? ________________ unlock iphone 3g

By: Tagz | "Topic extraction using Wikipedia data | Prashanth Ellina" | Comments

Sat, 16 May 2009 17:02:26 +0000

[...] [upmod] [downmod] Topic extraction using Wikipedia data | Prashanth Ellina (blog.prashanthellina.com) 1 points posted 3 weeks, 1 day ago by jeethu tags datamining [...]

By: Jacob

Jacob — Sat, 21 Mar 2009 09:48:52 +0000

hey id like to talk with you about your blog. please email me – thanks.

By: Hugo Leung

Hugo Leung — Tue, 16 Dec 2008 06:10:20 +0000

Hi,

I am a pharmacy student doing research on Aspirin, and oddly enough, I could really use your expertise. I am studying the interconnectedness of papers on aspirin in order to visualize the most relevant papers which have studied people over 75 years old. I am currently using graphviz to visual this web of papers. I love the graph outputs you’ve created! Can you tell me how I might be able to automate things so that the relevant nodes are darker and relevant edges thicker as you have done? As I am not a good at programming, i was going to do this manually!

By: prashanthellina

prashanthellina — Mon, 31 Mar 2008 16:52:26 +0000

Chris,

Your work is interesting! I tried out the page you gave me for a paragraph. I believe a constrained tagging vocabulary for all kinds of content is a great way to “semanticize” it. You can take a look at this presentation (http://www.prashanthellina.com/docs/interesting_ways_of_using_wikipedia_data.pdf) to get more details about my algorithm. You can contact me at prashanthellina AT gmail.com.

By: chris sizemore

chris sizemore — Fri, 28 Mar 2008 13:04:13 +0000

hi there, would have sent you an email, but haven’t noticed an obvious one after a very quick scan

i’m really impressed and intrigued by your research here… your approach and algorithm seems very complementary to what i’ve been experimenting with…

http://www.slideshare.net/guest2c797e/wikipedia-as-controlled-vocabulary/

http://sells.welcomebackstage.com:5000/item/submit

do let me know what you think. i wonder if there’s an easy way for me to try your node interconnectedness approach out as an additional filter in my approach?

best–

–chris sizemore

By: prashanthellina

prashanthellina — Tue, 01 Jan 2008 14:05:19 +0000

I did try neato and fdp and in both the cases the layout was not good as nodes were overlapping and edges would cut through nodes. I will however try out neato -Gmodel=subset and get back. Thanks for the tip.

By: robojiannis

robojiannis — Tue, 01 Jan 2008 13:47:50 +0000

Hey Stephen North just posted a comment on my blog about your work, which could be of interest for you.

“Hi. I’d suggest at least trying neato or fdp for layouts of these networks, particularly, neato -Gmodel=subset”

Hope it helps.

By: social media and green horses » Topic extraction in wikipedia

social media and green horses » Topic extraction in wikipedia — Sun, 30 Dec 2007 17:10:05 +0000

[...] Ellina made a great work extracting topics using wikipedia data. Using the Graphviz program, he shows “…the wealth of information (both as text and as [...]

By: shanmuganandh

shanmuganandh — Thu, 27 Dec 2007 02:31:22 +0000

cool…

By: prashanthellina

prashanthellina — Fri, 21 Dec 2007 12:59:41 +0000

I did but it was pushing down relevant nodes for some sample texts.

By: venkat

venkat — Fri, 21 Dec 2007 12:46:30 +0000

brilliant!
you amaze me, as ever.

did you try using the global interconnect count for filtering?