question on training providers in Nj/NY area

Does anyone know of training orgs that offer reporting services training? I was given a brief overview,but want a more advanced class.
For those who are using this tool, Is there ad-hoc user reporting? how would you compare this tool to others, like Cognos,etc
any thoughts on this tool would helpful. we currently use crstal reports, but looking for something more sophisticated
bevfor the add-hoc report part in your question it will be supported in reporting service 2005 see the following link
http://www.devx.com/dbzone/Article/28047?trk=DXRSS_DB
if you arelooking for something more sophisticated the answer it is not the reporting service as it is simple , fast and easy to make a report but in the same time it has some limits so u need to knew what do you need exactly?? and choose your tool.|||

I think the level of sophistication is drill downs, etc. Also would be nice to be incorporated in an .net application, either web based or client server. the reporting services seems to have it's own URL, etc

I know cognos is nice, but very pricey. Does anyone have recommendations on similiar report apps? Or if you use cognos, is it worth the price, for the level of reporting, drill down,etc

thanks for any info you may have

bev

|||The SQL Server 2005 version will compare to Cognos but development skills required, with the release of Business Object 11 the SQL Server team is making the development easier because the offerings are similar but Business Object can be used by the executive. One main difference Business Object implemented Time Interval at time span level which will make Time series aggregation easier, I read recently SQL Server 2005 ANSI SQL Time Interval was taken out so fine grain time series aggregation will take more code. Hope this helps.

Question on large volume of training dataset

Hi, all experts here,

Thanks a lot for your kind attention.

I have a question on training large volume of datasets. In this case, the training will take a long while to complete, is there anything we can do to improve that? I know, we obviously cant split the training dataset into different smaller datasets. What we can do to improve that?

Hope my question is clear for your help.

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

With best regards,

Yours sincerely,

Generally, the performance of the training operation depends on the size of the training set and there is not much one can do about this. Sometimes, the accuracy of the model is not improved significantly by adding new data. you might try smaller samples first and see if you really need all the data.

Certain optimizations could be done, depending on the algorithm. For example, if you are using the Neural Network algorithm you might want to make sure that the continuous columns are treated as continuous and not discretized, particularly if the the column is predictable. Also, make sure the model does not include unnecessary columns and only the required columns are marked as Predictable.

All these would improve the performance, but not significantly

Alternately, if you are trying just to do some sort of data exploration, you might want to start with Naive Bayes, which takes little time for training.|||

Hi, Bogdan,

Thanks a lot for your kind advices.

With best regards,

Yours sincerely,

Question on dealing with missing values for training models

Hi, all,

Just really wonder what is the good idea to deal with missing values? Should we leave the missing values there in the traning data set ? Or replace it with other values?

What I am really concerned is that if we simply replace those missing values with other values, then how will it really affect the correctness of the training models?

I am looking forward to hearing from you for the above issue and it will be really great if we have any kind of best practices of dealing with this issue.

Thanks.

With best regards,

Yours sincerely,

I have the same question, what's the best practice to replace NULL values in the source data? Should they be left as NULL or replaced with some predetermined default value?|||

For most algorithms, NULL and NULL substituted with a value do not mean the same thing. NULL is usually ignored whereas a substituted NULL is a real state value. The correct modeling approach depends on your data. If NULL has some information in your scenario (e.g. NULL for State means customer from a different country), it's a good idea to sutstitute NULL with a default value before training. If however, NULL means absense of data and has no information, it should be left as is.

Hope this helps.

|||

Hi, Shuvro,

Thanks a lot. It's is quite clear to me now to have a clear idea on dealing with the Null values.

With best regards,

Yours sincerely,

Question on dealing with missing values for training models

Hi, all,

Just really wonder what is the good idea to deal with missing values? Should we leave the missing values there in the traning data set ? Or replace it with other values?

What I am really concerned is that if we simply replace those missing values with other values, then how will it really affect the correctness of the training models?

I am looking forward to hearing from you for the above issue and it will be really great if we have any kind of best practices of dealing with this issue.

Thanks.

With best regards,

Yours sincerely,

I have the same question, what's the best practice to replace NULL values in the source data? Should they be left as NULL or replaced with some predetermined default value?|||

For most algorithms, NULL and NULL substituted with a value do not mean the same thing. NULL is usually ignored whereas a substituted NULL is a real state value. The correct modeling approach depends on your data. If NULL has some information in your scenario (e.g. NULL for State means customer from a different country), it's a good idea to sutstitute NULL with a default value before training. If however, NULL means absense of data and has no information, it should be left as is.

Hope this helps.

|||

Hi, Shuvro,

Thanks a lot. It's is quite clear to me now to have a clear idea on dealing with the Null values.

With best regards,

Yours sincerely,

Question on attributes selection for un-supervised algorithms and supervised algorithms

Hi, all,

Thanks for your kind attention.

Just wonder is there any good idea for us to select attributes for training models? Both for non-supervised algorithms like Association Rules and Clustering etc. and supervised algorithms like decision tree etc.

It will be much interesting to hear from you for any best practices and popular methods of dealing with this issue.

I am looking forward to hearing from you and thanks for your advices.

With best regards,

Yours sincerely,

Hi,

I assume that you are trying to select a subset of all your attributes to train the models. SQL Server Data Mining Algorithms have built in feature selection methods. For example, the Microsoft Decision Trees support the following attribute scoring methods: Entropy, Bayesian with K2 Prior and Bayesian Dirichlet Equivalent with Uniform Prior (which is used by default). When feature selection is necessary, the algorithm calculates the scores for each attribute and only train trees with selected features (with top scores, of course). Other algorithms have similar feature selection mechanism.

Thanks,

Question on attributes selection for un-supervised algorithms and supervised algorithms

Hi, all,

Thanks for your kind attention.

Just wonder is there any good idea for us to select attributes for training models? Both for non-supervised algorithms like Association Rules and Clustering etc. and supervised algorithms like decision tree etc.

It will be much interesting to hear from you for any best practices and popular methods of dealing with this issue.

I am looking forward to hearing from you and thanks for your advices.

With best regards,

Yours sincerely,

Hi,

I assume that you are trying to select a subset of all your attributes to train the models. SQL Server Data Mining Algorithms have built in feature selection methods. For example, the Microsoft Decision Trees support the following attribute scoring methods: Entropy, Bayesian with K2 Prior and Bayesian Dirichlet Equivalent with Uniform Prior (which is used by default). When feature selection is necessary, the algorithm calculates the scores for each attribute and only train trees with selected features (with top scores, of course). Other algorithms have similar feature selection mechanism.

Thanks,