I have CSV files that contain data of cache performance of a source with different workloads for a particular time period. For each time interval, data is recorded, includeing columns like ReadHits, WriteHits, Cacheusage, ReadMiss, etc.

Example contents:


      1       ,  150  , 0   ,  15474 , 12

      2       ,   0    , 0   , 700375, 245  

      3       ,  15426 ,  1546 , 45121,195

Note: Each interval will be of same time period, e.g. 1 interval = 40 seconds.

In each column, data will be from 0 to 60k+; this fluctuates for each interval. Example:

     interval 7    8    9    10  11

     Readhits 0   240  1680   0  2001

Suppose I have data until 60 intervals. How can I predict data from intervals 61 to 70?

I have used ARIMA model, random forest, kmeans and different machine learning algorithms but have never been able to predict anything close to actual values!

Which algorithm will be better on this kind of data for predicting data of the next intervals? Is there anything useful that can be extracted from this information?


migrated from ai.stackexchange.com Apr 18 '17 at 19:20

This question came from our site for people interested in conceptual questions about life and challenges in a world where "cognitive" functions can be mimicked in purely digital environment.


I would encode your input values to 2d bitmaps (2d arrays), and feed it to something like DXNN or HyperNEAT. I'm using DXNN, started playing with it recently, more superior to HyperNEAT.

it is an neuroevolutionary algorithm that evolves topology and weights for a NN that in turn controls a HyperCube (substrate encoding).

It is written in Erlang, which is good, unless you aren't familiar with functional languages, in that case you're better off with HyperNEAT.

On the other hand if you can, get the book.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.