Page Menu
Home
c4science
Search
Configure Global Search
Log In
Files
F121086803
cpp_ngram.cpp
No One
Temporary
Actions
Download File
Edit File
Delete File
View Transforms
Subscribe
Mute Notifications
Award Token
Subscribers
None
File Metadata
Details
File Info
Storage
Attached
Created
Tue, Jul 8, 15:03
Size
4 KB
Mime Type
text/x-c
Expires
Thu, Jul 10, 15:03 (1 d, 23 h)
Engine
blob
Format
Raw Data
Handle
27257728
Attached To
R3597 cpp_ngram
cpp_ngram.cpp
View Options
/*
Copyright (c) 2009 Noé Cuneo
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in
all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
THE SOFTWARE.
*/
#include <vector>
#include <string>
#include <map>
using
namespace
std
;
struct
NGCountStruct
{
// Stores the number of occurances of a n-gram in both strings
unsigned
int
counts
[
2
];
NGCountStruct
(){
counts
[
0
]
=
0
;
//number of occurances in the left string
counts
[
1
]
=
0
;
//number of occurances in the right string
}
};
template
<
class
stringType
>
class
NgramComparator
{
map
<
stringType
,
NGCountStruct
>
comp_map
;
stringType
theStrings
[
2
];
public
:
NgramComparator
(
const
stringType
&
s1
,
const
stringType
&
s2
,
unsigned
int
n
){
//Initialization with two strings
theStrings
[
0
]
=
s1
;
theStrings
[
1
]
=
s2
;
//For both strings... (w=0 for s0, w=1 for s1)
for
(
int
w
=
0
;
w
<=
1
;
w
++
){
int
size
=
theStrings
[
w
].
size
();
// for example: "bami" => "b", "ba", "bam", "ami", "mi", "i"
for
(
int
i
=
1
-
n
;
i
<=
size
-
1
;
i
++
){
int
start
=
i
;
int
stop
=
i
+
n
;
// Truncate if necessary
if
(
start
<
0
)
start
=
0
;
if
(
stop
>
size
)
stop
=
size
;
if
(
stop
>
start
){
// Build the key
stringType
key
=
theStrings
[
w
].
substr
(
start
,
stop
-
start
);
// Try to find the ngram key.
// "typename" is just here to let the compilator know that
// map<stringType, NGCountStruct>::const_iterator is a type!
typename
map
<
stringType
,
NGCountStruct
>::
const_iterator
it
=
comp_map
.
find
(
key
);
// If we do not have this key yet, we create it
if
(
it
==
comp_map
.
end
()){
// initialize a 0-0 NGCountStruct
comp_map
[
key
];
};
// Then increment it
comp_map
[
key
].
counts
[
w
]
+=
1
;
};
};
}
};
vector
<
int
>
getScoreArray
(){
// returns [left_ngrams, right_ngrams, common_ngrams]
vector
<
int
>
score
(
3
,
0
);
for
(
typename
map
<
stringType
,
NGCountStruct
>::
const_iterator
it
=
comp_map
.
begin
()
;
it
!=
comp_map
.
end
();
it
++
){
int
lefts
(
it
->
second
.
counts
[
0
]),
rights
(
it
->
second
.
counts
[
1
]);
score
[
0
]
+=
lefts
;
score
[
1
]
+=
rights
;
score
[
2
]
+=
min
(
lefts
,
rights
);
};
return
score
;
};
};
// Now we implement explicit methods for both types (string and wstring)
// so that we can expose them to Python with minimal object conversion...
// For Python scripts which need special calculations, we expose get_totals
vector
<
int
>
unicode_get_totals
(
wstring
s1
,
wstring
s2
,
unsigned
int
n
=
3
){
// returns [left_ngrams, right_ngrams, common_ngrams]
NgramComparator
<
wstring
>
ngc
(
s1
,
s2
,
n
);
return
ngc
.
getScoreArray
();
}
vector
<
int
>
ascii_get_totals
(
string
s1
,
string
s2
,
unsigned
int
n
=
3
){
NgramComparator
<
string
>
ngc
(
s1
,
s2
,
n
);
return
ngc
.
getScoreArray
();
}
// For standard ngram comparison, we expose the faster methods that follow
float
unicode_compare
(
wstring
s1
,
wstring
s2
,
unsigned
int
n
=
3
){
// returns common_ngrams/avg_ngrams_count
NgramComparator
<
wstring
>
ngc
(
s1
,
s2
,
n
);
vector
<
int
>
score
=
ngc
.
getScoreArray
();
return
float
(
score
[
2
])
*
2.0
/
(
float
(
score
[
0
]
+
score
[
1
]));
}
float
ascii_compare
(
string
s1
,
string
s2
,
unsigned
int
n
=
3
){
// returns common_ngrams/avg_ngrams_count
NgramComparator
<
string
>
ngc
(
s1
,
s2
,
n
);
vector
<
int
>
score
=
ngc
.
getScoreArray
();
return
float
(
score
[
2
])
*
2.0
/
(
float
(
score
[
0
]
+
score
[
1
]));
}
Event Timeline
Log In to Comment