mruby-changefinder

ChangeFinder class detect change points via continuous outlier and smoothing.

See also the paper.

Sample

• input trend data and output change-point score

• output change-point score only

See also @kentaro 's blog entry (JA).

install by mrbgems

• add conf.gem line to build_config.rb

MRuby::Build.new do |conf|

    # ... (snip) ...

    conf.gem :github => 'matsumoto-r/mruby-changefinder'
end

example

#
# ChangeFinder.new sdar_order_for_outlier, outlier_discount_param, sdar_order_for_change_point, change_point_discount_param, smooth_term
#
> cf = ChangeFinder.new 5, 0.01, 10, 0.01, 7
 => #<ChangeFinder:0x7fad5c80be50 @ts_data_buffer=[], @change_point_analyze=#<ChangeFinder::SDAR:0x7fad5c80bb80>, @smooth_term=5, @outlier_analyze=#<ChangeFinder::SDAR:0x7fad5c80be20>>
> cf.learn [1,2,1,2,3,2,1,2,1]
 => [6.2017912433901, 1.3973555597559, 2.4211198000217, 2.3979400886673, 1.7835503570548, 1.4166612339939, 1.4837836144657, 1.2835583707215, 1.1556254255408]
> cf.score 1
 => 1.1044914205061
>

Support tools

• create simulation data using ar model, white noise and change points

cd misc/
make

# ./ar number_of_data ar1_coeff ar2_coeff change_point_time change_point_total change_point_decrease expected_value_for_white_noise variance_value_white_noise
./ar 10000 0.6 0.5 2000 5 1 0.0 1.0 > ar.tsv

• create graph from ar.tsv

• analyze ar.tsv by ChangeFinder

./mruby/bin/mruby example/cf.rb
ls example/result.tsv

• create graph from result.tsv

Simulate data automatically

rake check

• create graph from example/result.tsv including outlier point using like google docs

License

under the MIT License:

• see LICENSE file