Question on Naive Bayes Viewer

Hi, guys,

I encountered a very weird question on Naive Bayes viewer, that is : one of the attributes does not appeared in Naive Bayes viewer? The original attribute data type is int data type, but then within the mining structure, I change it to discrete with Text as its data type. But the problem is after I trained the model, on the naive bayes viewer, that attribute does not appear at all? Why is that?

I have set the dependency value to be very low to enable all attributes to appear. But only that attribute got the problem?

I am looking forward to hearing from you shortly and thanks a lot in advance.

With best regards,

Yours sincerely

When you change a column's content type or data type (e.g from Integer to Text), BI Developer Studio might mark the column as Ignorable in the mining models that use it.

Could this be the issue you are seeing?

|||

No, the column is still labelled as input.

|||Was there a specific reason you needed to change the data type to Text? As far as the model is concerned, there's no difference between text and int for discrete attributes.

|||

Hi, Raman,

Thank you very much.

I think I know what you mean now, set the content type to be discrete, but the data type remains as Int? As when I left the data type as Int and content type as Continuous, the model treated all the data as continous. Is that right? Thank you.

With best regards,

Yours sincerely,

|||Correct - if you want ints to be treated as discrete, you just need to make sure that the content type is set correctly (to Discrete or Discretized).

|||

Hi, Raman,

Thanks.

With best regards,

Yours sincerely,

Question on Naive Bayes Viewer

Hi, guys,

I encountered a very weird question on Naive Bayes viewer, that is : one of the attributes does not appeared in Naive Bayes viewer? The original attribute data type is int data type, but then within the mining structure, I change it to discrete with Text as its data type. But the problem is after I trained the model, on the naive bayes viewer, that attribute does not appear at all? Why is that?

I have set the dependency value to be very low to enable all attributes to appear. But only that attribute got the problem?

I am looking forward to hearing from you shortly and thanks a lot in advance.

With best regards,

Yours sincerely

When you change a column's content type or data type (e.g from Integer to Text), BI Developer Studio might mark the column as Ignorable in the mining models that use it.

Could this be the issue you are seeing?

|||

No, the column is still labelled as input.

|||Was there a specific reason you needed to change the data type to Text? As far as the model is concerned, there's no difference between text and int for discrete attributes.

|||

Hi, Raman,

Thank you very much.

I think I know what you mean now, set the content type to be discrete, but the data type remains as Int? As when I left the data type as Int and content type as Continuous, the model treated all the data as continous. Is that right? Thank you.

With best regards,

Yours sincerely,

|||Correct - if you want ints to be treated as discrete, you just need to make sure that the content type is set correctly (to Discrete or Discretized).

|||

Hi, Raman,

Thanks.

With best regards,

Yours sincerely,

Question on cube write back

Hi, all experts here,

I encountered a problem with cube writing back which contains measures with count aggregation rather than sum which said write back can not be allowed on measure groups with other aggregations rather than sum? Is it saying we can only enable write back on cubes with measure groups all based on sum aggregations?

But in my case, the measures needed to be counted rather than sum. What can I try to enable write back on this cube?

I am looking forward to hearing from you for your advices and thank you very much in advance.

With best regards,

Yours sincerely,

I'm having trouble thinking of a situation where you would need to writeback to a count measure, but that is beside the point.

One work around might be to create a new column in your fact table for this measure and pre-populate it with a value of 1 during ETL. Then you would be able to use a sum aggregation and get the same value as you would with a count measure. It would then also be possible to write back to this measure.

|||

Hi,

In my data, the facts are bit data, which only with values of 1 and 0 meaning different things respectively. Therefore I need to count these different situations happened in the fact table across different levels. There are no numeric data types at the moment which are for sum.

Thanks for your advices and I am looking forward to hearing from any of you here more ideas in this situation. Maybe you would give me some better ideas on how to deal with it.

With best regards,

Yours sincerely,

|||

Helen,

Explain me better what is the goal, I didnt understood... Explain me as I a dummy!

:-)

Regards!

|||

HI, PedroCGD,

The facts in the facts tables are with 0 or 1 values only. (0 and 1 therefore represent different meanings), therefore I dont see any point to sum these facts in any cube. Instead, count aggregation makes more sense which can count the numbers for the fact being 0 or 1.

Hope my explanation is clear for your help.

With best regards,

Yours sincerely,

|||It sounds like you have some flags which should modelled as separate attributes, not as measures. You would have one count measure and set either a single dimension or a combined dimension (also known as a 'junk' dimension). In this way you should be able to get the count by any combination of attributes.|||

Hi, Darren,

Thanks for your kind advices.

As you kindly advised, I will then need to update the underlying relational database (as the underlying data warehouse is designed by others)?Since at the moment, the data residing in the data warehouse does not have any junk dimensions at all. All those flags attributes are residing in the fact tables.

I would need to create single dimension for each flag attribute or combine some of the flag attributes together into one single dimension? But only one count measure for all these dimension attributes?

Will it be any good to complete these in data source view of the cube? Thanks again. And hope you can give me more further advices on it.

With best regards,

Yours sincerely,

Question on cube processing and deployment

Hi, all experts here,

Thank you very much for your kind attention.

I encountered a very strange problem processing and deploying the cube to the analysis services server. The fact table where the measure group coming from only got several hundred rows, while at the same time I have other 6 dimensions with many attributes. My system drive currently still has nearly 1 GB space left, but when I deployed the cube, it took ages to run and ended up with process failure because of running out of space on my system drive.

This problem never happened before and it is really frustrated and I dont know what is going on.

Would please any experts here give me any advices. (I have tried ages, but still nothing really changes).

I am looking forward to hearing from you and thanks a lot in advance.

With best regards,

Yours sincerely,

Any expers know what is the cause of that problem?

Thanks.

With best regards,

Yours sincerely,

Question on attribute keys of measure group

Hi, dear friends,

I encountered a problem processing my cube with the error message telling: processing measure_group_name (which only has a count as its only measure)failed as attribute keys can not be found? Why is that? I have no idea why is that? Does it mean I have to include all the keys of the fact table where the measure group from in the measure groups as measures? Otherwise the system failed to identify the distinct count of the system? Or whatever measures the measure group contains, it alwys have to include all the keys including the dimension keys of dimensions which the measure group is related to?

Thanks in advance for your help and advices and I am always looking forward to hearing from you shortly from you.

With best regards,

Yours sincerely,

This message means that a column used to join the measure group to a dimension contains a value not found in the dimension. This is usually caused by a referential integrity error - the FK does not existing in the PK of the lookup table. However, it can also be caused by the presence of a null in the FK as by default this null will be converted to zero or blank which probably does not exist as a key in the dimension. You can handle the situation of nulls in FK columns within the fact table by setting the NullProcessing option in the advanced tab of the measure group relationship dialog from the Dimension Usage tab of the cube editor.|||

Hi, Matt,

Thanks a lot. Yes, the problem is the matching of the PK and the FK between the tables.

However, the data source view logical key and the relationships between tables are really annoying then, as it is not able to check the integrity of the tables?

With best regards,

Yours sincerely,

|||

So it sounds like you would like to have an option on the Data Source View to validate logical keys and relationships. I'd recommend you file a feature request on http://connect.microsoft.com/SQLServer/Feedback. Such feature requests will be automatically added to our issue tracking system for consideration in future releases and those that originate from a customer such as yourself get extra consideration since we know it's a real world scenario. If you can describe the scenario it will also help as it will allow us to better identify which requests will help the most people and how to implement them so that they actually provide the help intended.

Thanks, Matt

|||

Hi, Matt,

Thanks a lot and this sounds really a good idea of it.

With best regards,

Yours sincerely,