You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@mahout.apache.org by WangRamon <ra...@hotmail.com> on 2011/10/30 10:56:20 UTC
Get the most similar users with ItemSimilarityJob.
Hi All I have a use case which is trying to find the most similarity users. So the input data is still in format [USER_ID, ITEM_ID, PREF], I think I can use ItemSimilarityJob in Mahout 0.6, although the name is Item Similarity, by reading the code I find if I change the input to be [ITEM_ID, USER_ID, PREF], I should get the desired result which is the USER_ID and its similar user's ID, am I correct? Thanks in advance. Cheers Ramon
Re: Get the most similar users with ItemSimilarityJob.
Posted by Sebastian Schelter <ss...@apache.org>.
That would be great.
On 30.10.2011 11:03, Dan Brickley wrote:
> 2011/10/30 Sebastian Schelter <ss...@apache.org>:
>> Exactly.
>
> Wouldn't it be useful for every single Mahout job (the sub-commands of
> the bin/mahout script) to have a Wiki page to collect such gems? If
> so, should they be named after the Java class, or the commandline
> option? If they were named systematically then 'bin/mahout
> rowsimilarity --help' etc could even point users at such a page.
>
> The fate of the wiki isn't exactly clear, given
> https://issues.apache.org/jira/browse/MAHOUT-804 but if folk think it
> useful I'd put some time into this.
>
> Dan
>
>> --sebastian
>>
>> On 30.10.2011 10:56, WangRamon wrote:
>>>
>>>
>>>
>>>
>>> Hi All I have a use case which is trying to find the most similarity users. So the input data is still in format [USER_ID, ITEM_ID, PREF], I think I can use ItemSimilarityJob in Mahout 0.6, although the name is Item Similarity, by reading the code I find if I change the input to be [ITEM_ID, USER_ID, PREF], I should get the desired result which is the USER_ID and its similar user's ID, am I correct? Thanks in advance. Cheers Ramon
>>
>>
Re: Get the most similar users with ItemSimilarityJob.
Posted by Dan Brickley <da...@danbri.org>.
2011/10/30 Sebastian Schelter <ss...@apache.org>:
> Exactly.
Wouldn't it be useful for every single Mahout job (the sub-commands of
the bin/mahout script) to have a Wiki page to collect such gems? If
so, should they be named after the Java class, or the commandline
option? If they were named systematically then 'bin/mahout
rowsimilarity --help' etc could even point users at such a page.
The fate of the wiki isn't exactly clear, given
https://issues.apache.org/jira/browse/MAHOUT-804 but if folk think it
useful I'd put some time into this.
Dan
> --sebastian
>
> On 30.10.2011 10:56, WangRamon wrote:
>>
>>
>>
>>
>> Hi All I have a use case which is trying to find the most similarity users. So the input data is still in format [USER_ID, ITEM_ID, PREF], I think I can use ItemSimilarityJob in Mahout 0.6, although the name is Item Similarity, by reading the code I find if I change the input to be [ITEM_ID, USER_ID, PREF], I should get the desired result which is the USER_ID and its similar user's ID, am I correct? Thanks in advance. Cheers Ramon
>
>
Re: Get the most similar users with ItemSimilarityJob.
Posted by Sebastian Schelter <ss...@apache.org>.
Exactly.
--sebastian
On 30.10.2011 10:56, WangRamon wrote:
>
>
>
>
> Hi All I have a use case which is trying to find the most similarity users. So the input data is still in format [USER_ID, ITEM_ID, PREF], I think I can use ItemSimilarityJob in Mahout 0.6, although the name is Item Similarity, by reading the code I find if I change the input to be [ITEM_ID, USER_ID, PREF], I should get the desired result which is the USER_ID and its similar user's ID, am I correct? Thanks in advance. Cheers Ramon