mahout-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From thibaut <thibaut.gensol...@gmail.com>
Subject Matrix inversion
Date Thu, 05 May 2016 19:45:56 GMT
Hi there,

I am working in a group of research of the Michigan University (Mathematics), and we are thinking
to increase the speed of some algorithms we are using and developed here, by using distributed
systems.

We were thinking about using Spark, but I found recently Mahout and I read about it. We are
using a lot KNN and Minimal Spanning Tree here, and our main concern is about dealing with
the inversion of Matrix (really really big matrix)

I found this paper : https://web.njit.edu/~ansari/papers/16IEEEAccess.pdf <https://web.njit.edu/~ansari/papers/16IEEEAccess.pdf>
, Spark-based Large-scale Matrix Inversion for Big Data Processing, which provides a really
good method for dealing with the inversion issue.

My askings are: 
- Is it better for what we want to do to use Mahout, or Spark ? 
- I saw that you already have a distributed PCA. Do you have a really efficient matrix inversion
algorithm in Mahout ? 
- How good is the linear algebra library in compare to Matlab for example ?

Finally, our main concern for using Spark is about the linear algebra library that is used
with Spark. And we were wondering how good is the Mahout one ?

Thanking you in advance,

Best regards.
Thibaut
Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message