Grouping By Similarity (using pg_trgm)?

Started by Cory Tuckeralmost 11 years ago5 messagesgeneral
Jump to latest
#1Cory Tucker
cory.tucker@gmail.com

[pg version 9.3 or 9.4]

Suppose I have a simple table:

create table data (
my_value TEXT NOT NULL
);
CREATE INDEX idx_my_value ON data USING gin(my_value gin_trgm_ops);

Now I would like to essentially do group by to get a count of all the
values that are sufficiently similar. I can do it using something like a
CROSS JOIN to join the table on itself, but then I still am getting all the
rows with duplicate counts.

Is there a way to do a group by query and only return a single "my_value"
column and a count of the number of times other values are similar while
also not returning the included similar values in the output, too?

#2David G. Johnston
david.g.johnston@gmail.com
In reply to: Cory Tucker (#1)
Re: Grouping By Similarity (using pg_trgm)?

On Thu, May 14, 2015 at 11:58 AM, Cory Tucker <cory.tucker@gmail.com> wrote:

[pg version 9.3 or 9.4]

Suppose I have a simple table:

create table data (
my_value TEXT NOT NULL
);
CREATE INDEX idx_my_value ON data USING gin(my_value gin_trgm_ops);

Now I would like to essentially do group by to get a count of all the
values that are sufficiently similar. I can do it using something like a
CROSS JOIN to join the table on itself, but then I still am getting all the
rows with duplicate counts.

Is there a way to do a group by query and only return a single "my_value"
column and a count of the number of times other values are similar while
also not returning the included similar values in the output, too?

​Concept below - not bothering to lookup the functions/operators for
pg_trgm:

SELECT my_value_src, count(*)
FROM (SELECT my_value AS my_value_src FROM data) src
JOIN (SELECT my_value AS my_value_compareto FROM data) comparedto
ON ( func(my_value_src, my_value_compareto) < # )
GROUP BY my_value_src

​David J.

#3Cory Tucker
cory.tucker@gmail.com
In reply to: David G. Johnston (#2)
Re: Grouping By Similarity (using pg_trgm)?

That produces pretty much the same results as the CROSS JOIN I was using
before. Because each "my_value" in the table are different, if I group on
just their value then I will always have the full result set and a bunch of
essentially duplicated results.

Any other ideas/options?

On Thu, May 14, 2015 at 12:08 PM David G. Johnston <
david.g.johnston@gmail.com> wrote:

Show quoted text

On Thu, May 14, 2015 at 11:58 AM, Cory Tucker <cory.tucker@gmail.com>
wrote:

[pg version 9.3 or 9.4]

Suppose I have a simple table:

create table data (
my_value TEXT NOT NULL
);
CREATE INDEX idx_my_value ON data USING gin(my_value gin_trgm_ops);

Now I would like to essentially do group by to get a count of all the
values that are sufficiently similar. I can do it using something like a
CROSS JOIN to join the table on itself, but then I still am getting all the
rows with duplicate counts.

Is there a way to do a group by query and only return a single "my_value"
column and a count of the number of times other values are similar while
also not returning the included similar values in the output, too?

​Concept below - not bothering to lookup the functions/operators for
pg_trgm:

SELECT my_value_src, count(*)
FROM (SELECT my_value AS my_value_src FROM data) src
JOIN (SELECT my_value AS my_value_compareto FROM data) comparedto
ON ( func(my_value_src, my_value_compareto) < # )
GROUP BY my_value_src

​David J.

#4David G. Johnston
david.g.johnston@gmail.com
In reply to: Cory Tucker (#3)
Re: Grouping By Similarity (using pg_trgm)?

On Thu, May 14, 2015 at 1:09 PM, Cory Tucker <cory.tucker@gmail.com> wrote:

That produces pretty much the same results as the CROSS JOIN I was using
before. Because each "my_value" in the table are different, if I group on
just their value then I will always have the full result set and a bunch of
essentially duplicated results.

Any other ideas/options?

​how do you want to solve the problem:

A is similar to B
B is similar to C
A IS NOT similar to C

​?

I'm not sure that PostgreSQL is the best tool to solve clustering
problems...though my experience with them is minimal.

David J.

#5Oleg Bartunov
oleg@sai.msu.su
In reply to: Cory Tucker (#1)
Re: Grouping By Similarity (using pg_trgm)?

Have you seen http://www.sai.msu.su/~megera/postgres/talks/pgcon-2012.pdf ?

On Thu, May 14, 2015 at 9:58 PM, Cory Tucker <cory.tucker@gmail.com> wrote:

Show quoted text

[pg version 9.3 or 9.4]

Suppose I have a simple table:

create table data (
my_value TEXT NOT NULL
);
CREATE INDEX idx_my_value ON data USING gin(my_value gin_trgm_ops);

Now I would like to essentially do group by to get a count of all the
values that are sufficiently similar. I can do it using something like a
CROSS JOIN to join the table on itself, but then I still am getting all the
rows with duplicate counts.

Is there a way to do a group by query and only return a single "my_value"
column and a count of the number of times other values are similar while
also not returning the included similar values in the output, too?