sciruby 0.1.0
Sign up to get free protection for your applications and to get access to all the features.
- data/lib/sciruby.rb +31 -0
- data/lib/sciruby/recommend.rb +66 -0
- metadata +63 -0
data/lib/sciruby.rb
ADDED
@@ -0,0 +1,31 @@
|
|
1
|
+
# = sci_ruby.rb -
|
2
|
+
# SciRuby - Ruby scientific visualization and computation.
|
3
|
+
#
|
4
|
+
# Copyright (C) 2011 SciRuby Development Team
|
5
|
+
# * John O. Woods
|
6
|
+
# * John T. Prince
|
7
|
+
# * Claudio Bustos
|
8
|
+
# * others
|
9
|
+
#
|
10
|
+
# This program is free software; you can redistribute it and/or
|
11
|
+
# modify it under the terms of the GNU General Public License
|
12
|
+
# as published by the Free Software Foundation; either version 2
|
13
|
+
# of the License, or (at your option) any later version.
|
14
|
+
#
|
15
|
+
# This program is distributed in the hope that it will be useful,
|
16
|
+
# but WITHOUT ANY WARRANTY; without even the implied warranty of
|
17
|
+
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
18
|
+
# GNU General Public License for more details.
|
19
|
+
#
|
20
|
+
# You should have received a copy of the GNU General Public License
|
21
|
+
# along with this program; if not, write to the Free Software
|
22
|
+
# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
|
23
|
+
#
|
24
|
+
# Specific notices will be placed where they are appropriate.
|
25
|
+
#
|
26
|
+
|
27
|
+
module SciRuby
|
28
|
+
VERSION = '0.1.0'
|
29
|
+
|
30
|
+
autoload(:Recommend, 'sciruby/recommend')
|
31
|
+
end
|
@@ -0,0 +1,66 @@
|
|
1
|
+
require 'set'
|
2
|
+
require 'distribution' # for Hypergeometric
|
3
|
+
|
4
|
+
unless defined?(SortedSet) # Ruby 1.8
|
5
|
+
class SortedSet < Set
|
6
|
+
end
|
7
|
+
end
|
8
|
+
|
9
|
+
# Added by John O. Woods.
|
10
|
+
#
|
11
|
+
# Makes use of Distribution gem, formerly part of Statsample, by Claudio Bustos.
|
12
|
+
|
13
|
+
module SciRuby
|
14
|
+
# Methods and classes for expert recommendation systems.
|
15
|
+
module Recommend
|
16
|
+
# Set Distance functions: determine distances between sets.
|
17
|
+
#
|
18
|
+
# These functions may be useful for k-nearest neighbors searches and expert recommendation systems.
|
19
|
+
#
|
20
|
+
# Sets are used for systems where vectors are binary. For example, if you have a matrix of customers and products, a
|
21
|
+
# zero means the customer has not bought the product, and a one means the customer has. There is no concept of
|
22
|
+
# degree.
|
23
|
+
#
|
24
|
+
# Pearson is probably the function most people will want to use.
|
25
|
+
class SetDistance
|
26
|
+
|
27
|
+
# Create a new recommendation-by-set-distance object. This requires as arguments two sets (+a+ and +b+) and a
|
28
|
+
# +total+ size which indicates the number of items from which the sets are drawn.
|
29
|
+
#
|
30
|
+
# It also takes an optional +distance_function+ (e.g., :hypergeometric or :pearson). If this is given, the distance
|
31
|
+
# will be calculated immediately. Otherwise, you can use the various distance functions to calculate distance by
|
32
|
+
# a variety of metrics (e.g., distance_hypergeometric, distance_pearson, and so on).
|
33
|
+
def initialize a, b, total, distance_function = nil
|
34
|
+
@a = a.is_a?(Set) ? a : SortedSet.new(a)
|
35
|
+
@b = b.is_a?(Set) ? b : SortedSet.new(b)
|
36
|
+
@total = total
|
37
|
+
|
38
|
+
unless distance_function.nil? # Calculate immediately if a distance function is given.
|
39
|
+
@distance_function = "distance_#{distance_function.to_s}".to_sym
|
40
|
+
@distance = self.send @distance_function
|
41
|
+
end
|
42
|
+
end
|
43
|
+
|
44
|
+
attr_reader :a, :b, :total, :distance
|
45
|
+
|
46
|
+
def m; a.size; end
|
47
|
+
def n; b.size; end
|
48
|
+
def ab; @ab ||= a.intersection(b); end
|
49
|
+
def k; ab.size; end
|
50
|
+
alias :a_dot_b :k
|
51
|
+
|
52
|
+
# Calculate distance as the hypergeometric probability of seeing an intersection of +k+ or greater between two sets
|
53
|
+
# +a+ and +b+. This is basically the complement of cdf(+k-1+, +m+, +n+, +total+).
|
54
|
+
def distance_hypergeometric
|
55
|
+
@distance_hypergeometric ||= 1.0 - Distribution::Hypergeometric.cdf(k-1, m, n, total)
|
56
|
+
end
|
57
|
+
|
58
|
+
# The generalization of Pearson correlation coefficient using binary vectors.
|
59
|
+
def distance_pearson
|
60
|
+
@distance_pearson ||= 1.0 - (total * a_dot_b - m*n).abs / Math.sqrt( (total - m)*(total - n)*m*n )
|
61
|
+
end
|
62
|
+
|
63
|
+
end
|
64
|
+
|
65
|
+
end
|
66
|
+
end
|
metadata
ADDED
@@ -0,0 +1,63 @@
|
|
1
|
+
--- !ruby/object:Gem::Specification
|
2
|
+
name: sciruby
|
3
|
+
version: !ruby/object:Gem::Version
|
4
|
+
version: 0.1.0
|
5
|
+
prerelease:
|
6
|
+
platform: ruby
|
7
|
+
authors:
|
8
|
+
- SciRuby
|
9
|
+
autorequire:
|
10
|
+
bindir: bin
|
11
|
+
cert_chain: []
|
12
|
+
date: 2011-08-01 00:00:00.000000000 -04:00
|
13
|
+
default_executable:
|
14
|
+
dependencies: []
|
15
|
+
description: ! 'Ruby has for some time had no equivalent to the beautifully constructed
|
16
|
+
numpy, scipy, and matplotlib libraries for Python. We believe that the time for
|
17
|
+
a Ruby science and visualization package has come and gone. Sometimes when a solution
|
18
|
+
of sugar and water becomes super-saturated, from it precipitates a pure, delicious,
|
19
|
+
and diabetes-inducing crystal of sweetness, induced by no more than the tap of a
|
20
|
+
finger. So it is, we believe, with the need for numeric and visualization libraries
|
21
|
+
in Ruby.
|
22
|
+
|
23
|
+
|
24
|
+
We are not the first with this idea, but we are trying to bring it to life.
|
25
|
+
|
26
|
+
|
27
|
+
At this point, SciRuby has not much to offer. But if you install this gem, you''ll
|
28
|
+
get as dependencies all of the libraries that we plan to incorporate into SciRuby.
|
29
|
+
|
30
|
+
'
|
31
|
+
email: john.woods@marcottelab.org
|
32
|
+
executables: []
|
33
|
+
extensions: []
|
34
|
+
extra_rdoc_files: []
|
35
|
+
files:
|
36
|
+
- lib/sciruby.rb
|
37
|
+
- lib/sciruby/recommend.rb
|
38
|
+
has_rdoc: true
|
39
|
+
homepage: http://www.github.com/SciRuby/sciruby
|
40
|
+
licenses: []
|
41
|
+
post_install_message:
|
42
|
+
rdoc_options: []
|
43
|
+
require_paths:
|
44
|
+
- lib
|
45
|
+
required_ruby_version: !ruby/object:Gem::Requirement
|
46
|
+
none: false
|
47
|
+
requirements:
|
48
|
+
- - ! '>='
|
49
|
+
- !ruby/object:Gem::Version
|
50
|
+
version: '0'
|
51
|
+
required_rubygems_version: !ruby/object:Gem::Requirement
|
52
|
+
none: false
|
53
|
+
requirements:
|
54
|
+
- - ! '>='
|
55
|
+
- !ruby/object:Gem::Version
|
56
|
+
version: '0'
|
57
|
+
requirements: []
|
58
|
+
rubyforge_project: sciruby
|
59
|
+
rubygems_version: 1.6.2
|
60
|
+
signing_key:
|
61
|
+
specification_version: 3
|
62
|
+
summary: Tools for scientific computation in Ruby
|
63
|
+
test_files: []
|