You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@predictionio.apache.org by Vaghawan Ojha <va...@gmail.com> on 2017/03/28 18:26:56 UTC
The recommendation engine shows only one result for all users
Hi,
I've added more than 300000 events in the engine, and I'm using this
template (
https://github.com/apache/incubator-predictionio-template-recommender)
When I had small data set, the engine was Ok and it was showing different
result for different users but when I've trained 300000 of dataset it's
giving me the weird result.
For every users there is same product recommendation with just the
difference in the score.
Why this happened, I deleted the data once and tested again but still the
same result. I don't know what's being so wrong here.
Have anyone had experienced the same? Please let me know.
Thank you
Re: The recommendation engine shows only one result for all users
Posted by Vaghawan Ojha <va...@gmail.com>.
Hi,
I've approx : 1000000 event data, all the data looks exactly like this:
{"entityId":"00021938","entityType":"user","event":"buy","targetEntityType":"item","targetEntityId":"BM55432"}
{"entityId":"00040844","entityType":"user","event":"buy","targetEntityType":"item","targetEntityId":"BM59126"}
{"entityId":"00040844","entityType":"user","event":"buy","targetEntityType":"item","targetEntityId":"BM50258"}
{"entityId":"00040844","entityType":"user","event":"buy","targetEntityType":"item","targetEntityId":"BM8920"}
{"entityId":"00040844","entityType":"user","event":"buy","targetEntityType":"item","targetEntityId":"BM11408"}
{"entityId":"00040844","entityType":"user","event":"buy","targetEntityType":"item","targetEntityId":"BM63982"}
{"entityId":"00041181","entityType":"user","event":"buy","targetEntityType":"item","targetEntityId":"BM41244"}
{"entityId":"00041181","entityType":"user","event":"buy","targetEntityType":"item","targetEntityId":"BM31483"}
{"entityId":"00041181","entityType":"user","event":"buy","targetEntityType":"item","targetEntityId":"BM66411"}
{"entityId":"00041181","entityType":"user","event":"buy","targetEntityType":"item","targetEntityId":"BM9068"}
{"entityId":"00040765","entityType":"user","event":"buy","targetEntityType":"item","targetEntityId":"BM8928"}
{"entityId":"00040765","entityType":"user","event":"buy","targetEntityType":"item","targetEntityId":"BM8925"}
It's fine at the size of approx 50000 event, but not same with the large
data. Should you need anything else, let me know further.
Thanks
On Fri, Mar 31, 2017 at 7:17 PM, Donald Szeto <do...@apache.org> wrote:
> Hi,
>
> A statistics of your event data's distribution would be helpful.
>
> Regards,
> Donald
>
> On Tue, Mar 28, 2017 at 11:27 AM Vaghawan Ojha <va...@gmail.com>
> wrote:
>
>> Hi,
>>
>> I've added more than 300000 events in the engine, and I'm using this
>> template (https://github.com/apache/incubator-predictionio-
>> template-recommender)
>>
>> When I had small data set, the engine was Ok and it was showing different
>> result for different users but when I've trained 300000 of dataset it's
>> giving me the weird result.
>> For every users there is same product recommendation with just the
>> difference in the score.
>>
>> Why this happened, I deleted the data once and tested again but still the
>> same result. I don't know what's being so wrong here.
>>
>> Have anyone had experienced the same? Please let me know.
>>
>> Thank you
>>
>
Re: The recommendation engine shows only one result for all users
Posted by Donald Szeto <do...@apache.org>.
Hi,
A statistics of your event data's distribution would be helpful.
Regards,
Donald
On Tue, Mar 28, 2017 at 11:27 AM Vaghawan Ojha <va...@gmail.com>
wrote:
> Hi,
>
> I've added more than 300000 events in the engine, and I'm using this
> template (
> https://github.com/apache/incubator-predictionio-template-recommender)
>
> When I had small data set, the engine was Ok and it was showing different
> result for different users but when I've trained 300000 of dataset it's
> giving me the weird result.
> For every users there is same product recommendation with just the
> difference in the score.
>
> Why this happened, I deleted the data once and tested again but still the
> same result. I don't know what's being so wrong here.
>
> Have anyone had experienced the same? Please let me know.
>
> Thank you
>