News for hackers. Stuff that matters.
Re:Show HN: Graph of wikipedia articles semantic similarity ... by lucamartinetti on Wednesday March 21, 2012 @10:05AM (#3734427)
Small experiment of visualization of wikipedia articles as a graph using d3.js.

Articles with more traffic are bigger. I computed the semantic similarity using LSI with python (gensim) You have to scroll down/right a bit!

http://similarityapi.appspot.com/graph/?title=blade%20runner

There is also a JSON api: http://similarityapi.appspot.com/api/v1/?limit=100&title...

All feedback is appreciated:

@lucamartinetti [email protected]

Re:Show HN: Graph of wikipedia articles semantic similarity ... by 3pt14159 on Wednesday March 21, 2012 @03:19PM (#3736066)
I've had much, much better results with LDA than LSI. Give that a shot if you have a chance, you'll be blown away. Stop word ratios are important, and make the max number of tokens 500,000.
Re:Show HN: Graph of wikipedia articles semantic similarity ... by viscanti on Wednesday March 21, 2012 @11:08AM (#3734705)
The JSON api should degrade gracefully if results aren't found. I.E. There should be a JSON message explaining that that item doesn't exist.
Re:Show HN: Graph of wikipedia articles semantic similarity ... by lucamartinetti on Wednesday March 21, 2012 @11:29AM (#3734791)
Right! It could use some input checking / normalization too. It expects the title parameter to be lower case now.
Re:Show HN: Graph of wikipedia articles semantic similarity ... by rplnt on Wednesday March 21, 2012 @12:04PM (#3734968)
Option to select language version could be a good feature (defaulting to en as now).
Re:Show HN: Graph of wikipedia articles semantic similarity ... by Radim on Wednesday March 21, 2012 @03:29PM (#3736133)
how much data did you use for the semantic analysis?
Re:Show HN: Graph of wikipedia articles semantic similarity ... by lucamartinetti on Wednesday March 21, 2012 @04:42PM (#3736507)
The whole text of all articles from wikipedia english (then filtered those with more the 1k views last month)