Math_Distance-0.3

Here's an updated and working version of Math_Distance. Below, the last message replied: > If you know an existing C lib that does metrics/clustering/similarity, then you > can use SWIG to generate the PHP wrapper functions to use that. I can give a > hand w/ the little I know the process if needed. Well, I really haven't used any of those libraries, but sure they exist. Actually you mentioned GSL, but I can't find clustering in GSL. Wrapping a library would surely speed up developement process, but there are some topics in which I think about: - Will these functions be available in any plattform? - They will sure behave faster than php code, but ¿will they run faster than if programmed directly in C as php extensions? - Will they be as flexible as they need to handle, for example, text data? If the balance of this answers if positive to wrapping, then let's wrap!!! Anyway, I have proposed myself to deploy Math_Distance and Clustering_KMeans, anything beyond that point is on the grace of numbers! > Yep, if there is a C lib for that, then the wrapper methods can decide on the > fly whether to use the PHP or the C implementation of the algorithm. Quite nice! > > for distance metrics. I think that Math_Distance would do it fine, as short. > +1 for that name > classes. I was thinking on only 3 metrics for vectors: cartesian distance and > the other 2 mentioned below. Ok. The eight metrics that I'm shipping are yet working fine! The last 4 (Jaccard Similarity, Sorensen Similarity, Simple Matching Coefficient and Baroni-Urbani Buser Coefficient) are meant to be used with binary individuals. At the time, binary individuals are represented by arrays of integers, beeing 0 = false and >0 = true. You can also see it in the accompanying example. Altough the software it's working, if anyone would like to test te accuracy of the results, I'll really appreciate. Greetings Alfredo http://alfredorahn.com

Attachment: [application/x-zip-compressed] Math_Distance-0.3.zip