Please start any new threads on our new
site at https://forums.sqlteam.com. We've got lots of great SQL Server
experts to answer whatever question you can come up with.
| Author |
Topic |
|
Kirun
Yak Posting Veteran
60 Posts |
 Posted - 2002-04-27 : 12:10:59
|
| a newbie to this OLAP world. wanted to ask some questions.. what are the predicting fields which one can include in the table.. This predicting fiedls should be in facttable or in the main table...What is a tresh hold value and how can one set this value..and where where one manage the data mining algorithem and do we set this threshhold value in algoririthemwhat is trained data and how should we use this trained data... |
|
|
khurshed
Starting Member
1 Post |
Posted - 2002-05-12 : 16:19:45
|
| There are two types of analysis: Classification and prediction.For "prediction" we will have to define a predicted field. The predicted field is that field whose values will be determine in the analysis. To do it, we will have to define a predicted class. We can define it either in the dimension table or in facttable depend upon the sceniro but the predicted field must satisfy the demographic characterstics in the table. Trained data will be the data that will be used for the learning process. We can train the data by using the "data mining prediction task" in DTS. It will create a predicted column in the selected dimension table that will be furthur used for prediction. The data can be shown either in the table or by means of charts. |
 |
|
|
Kirun
Yak Posting Veteran
60 Posts |
Posted - 2002-05-13 : 00:15:30
|
| its good to see a pakistani here i m happy that you have helped me in clearing many things i will give that DTS data mining prediction a shot then i will get back to you...thanks in advance Salaam ( a new word to this forum) |
|
|
|
|
|
|
Pridected fields and mining models