can use Where in Dts query?

i create a model mining to know percent of product, which is like more by customer is have percent biger than other products that is don't like by customer.

i use DTS to get result to sql server.

select flattened

predict([modelName].[tablePredict],Include_statistic)

from

[modelName]

prediction join

.....

i have result with a table A in sql server:

productid $support ... $adjustedprobability

1 0.0 ...

1 0.0 ...

1 63.3 .....

how can i write in Dts so that table A not have record with value $support=0.0

- second: percent people like that product is knowned by $adjustedprobability or $support and value small or big

1:

SELECT FLATTENED
(SELECT * FROM Predict([modelName.tablePredict,INCLUDE_STATISTICS) WHERE $support > 0) as p
FROM [modelName]
PREDICTION JOIN ....

2: If you want descending, you can just use the column in the query - e.g.
SELECT FLATTENED Predict(model.table, 5, $AdjustedProbability) FROM ...

If you want bottom you need to use BottomCount(Predict(model.table),include_statistics), $support,5)

You should check my syntax - I just typed this from memory

Thx

|||

thank you very much.

you right when tell me check my syntax, there are some mistake.

thank you again

|||

hi, where can i put where in this query. it gives error when i try to use where clause.

i want 5 products but also their $AdjustedProbability should be > 0.36 ex.

SELECTFLATTENED

TopCount(Predict([Customer Products],

INCLUDE_STATISTICS) ,

$AdjustedProbability ,

5)

FROM [Basket Analysis]

NATURALPREDICTIONJOIN

( SELECT (

SELECT 'Product A' AS [Product]

)

AS [Customer Products] ) AS t