DMX处理挖掘结构错误

时间:2014-05-23 07:31:27

标签: sql-server sql-server-2012 ssms ssms-2012 dmx-ssas

我在处理挖掘结构时遇到问题......

这是我的挖掘结构

CREATE MINING
STRUCTURE [QUESTION1] (
    [ownerReviewID] LONG KEY,
    [makeName] TEXT DISCRETE,
    [modelName] TEXT DISCRETE,
    [price] LONG CONTINUOUS,
    [priceDisc] LONG DISCRETIZED(AUTOMATIC),
    [seriesYear] LONG CONTINUOUS,
    [seriesYearDisc] LONG DISCRETIZED(AUTOMATIC),
    [ownerName] TEXT DISCRETE,
    [ownedSinceNew] BOOLEAN DISCRETE,
    [ownedPeriod] LONG DISCRETE,
                   [feature_ratings] TABLE (
                                [transID] LONG KEY,
                                [ownerReviewID]
LONG DISCRETE,
                                [feature] TEXT DISCRETE,
                                [rating] LONG CONTINUOUS,
                                [ratingDisc] LONG DISCRETIZED(AUTOMATIC)
                           )

) WITH HOLDOUT
(30 PERCENT OR 10000 CASES);

我用Naive Bayes创建了一个挖掘模型:

ALTER MINING STRUCTURE [QUESTION1]

ADD MINING MODEL [PredictReBuy-NBayes]
( 
 [ownerReviewID],
 [makeName],
 [modelName],
 [priceDisc] AS [price],
 [seriesYearDisc] AS
[seriesYear],
 [ownerName],
 [ownedSinceNew],

    feature_ratings(
         transID,
         feature,
         [ratingDisc] AS
[rating]      PREDICT     )
) 
USING Microsoft_Naive_Bayes

到目前为止,上述2个代码工作正常,全部执行。

当我尝试执行处理代码时,底部显示的错误出现了。 这是我的处理代码:

 INSERT INTO MINING STRUCTURE [QUESTION1]

(
   [ownerReviewID],[makeName],[modelName],[price],[seriesYear],[ownerName],[ownedSinceNew],[ownedPeriod],  feature_ratings(SKIP, [transID])
)
SHAPE {
  OPENQUERY([Car Rating],'SELECT ownerReviewID, makeName, modelName, price, seriesYear, ownerName, ownedSinceNew, ownedPeriod FROM dbo.owner_reviews ORDER BY ownerReviewID')}
APPEND
( 
  {OPENQUERY([Car Rating],'SELECT transID, ownerReviewID, feature, rating FROM 
    dbo.feature_ratings ORDER BY transID')
}
RELATE ownerReviewID to ownerReviewID 
) AS [feature_ratings]

我目前收到此错误,我不明白如何解决......任何想法?谢谢!

Executing the query ...Error (Data mining): The count of the specified columns does not match the count of  

the mining model columns or the input rowset columns.Execution complete

1 个答案:

答案 0 :(得分:0)

我重新编写代码..结果是错误就是它所说的......它与列不匹配。