??SQL Server 2000 ???????? - PowerPoint PPT Presentation

1 / 48
About This Presentation
Title:

??SQL Server 2000 ????????

Description:

Title: ppt title Subject: event name Author: speaker Description: Template design: Silverfox Productions Graphic design: Polly Macomber Formatter: – PowerPoint PPT presentation

Number of Views:147
Avg rating:3.0/5.0
Slides: 49
Provided by: spea192
Category:

less

Transcript and Presenter's Notes

Title: ??SQL Server 2000 ????????


1
??SQL Server 2000????????
  • ZhaoHui Tang ???
  • Program Manager
  • SQL Server Analysis Services
  • Microsoft ??

2
??
  • Microsoft???????
  • OLE DB for DM ??????
  • ?????????????
  • ????
  • ???????
  • ?????????
  • Microsoft?????????
  • ???

3
SQL Server 2000???????
4
???
  • ??????????????
  • ????
  • ??????
  • ????
  • ?????????????????????
  • ????
  • ???????????
  • ????C4.5?C5?CART?Chaid

5
????????
IQ IQ IQ ????? ????? ???? ???? ?? ??
? ? ? ? ? ? ? ? ?
???? ? 300 500 200 700 300 400 600 500 500
???? ? 100 1000 900 400 1600 400 1600 1100 900
6
????
IQ IQ IQ ????? ????? ???? ???? ?? ??
? ? ? ? ? ? ? ? ?
???? ? 200 400 100 700 0 300 400 400 250
???? ? 50 250 100 400 0 100 300 250 150
7
Microsoft ???
  • ?????
  • ???????( Bayesian )??????( Entropy)
  • ????
  • ????????N??
  • ?????

8
????(EM )
  • ????????????????????????
  • ??
  • ???????
  • ????????????
  • ??????????????
  • ?????? ?????????
  • ??????

9
EM ??
10
Microsoft?????(???EM )
???? ?????
??/?? ??
?????
???
??
???? ? ????
????
11
OLE DB for Data Mining
12
OLE DB for DM
  • ?????????
  • ??????
  • SQL
  • OLE DB
  • DM???????
  • ??(Case)?????(Nested Case)
  • ????(Mining Model)
  • ????(Model Creation)
  • ????(Model Training)
  • ??(Prediction)
  • ??API???

13
???
?? ID ?? ?? ?? ??
1 ??? 85 ? ?
2 ?? 40 ? ?
3 ?? 90 ? ?
4 ?? 50 ? ?
5 ?? 45 ? ?

14
DM ????
Create Mining Model CreditRisk (CustomerID long
key, Gender text discrete, Income long
continuous, Profession text discrete, Risk text
discrete predict) Using Microsoft_Decision_Trees
Insert into CreditRisk (CustomerId, Gender,
Income, Profession, Risk) Select CustomerID,
Gender, Income, Profession,Risk From Customers
Select NewCustomers.CustomerID, CreditRisk.Risk,
PredictProbability(CreditRisk) From CreditRisk
Prediction Join NewCustomers On
CreditRisk.GenderNewCustomer.Gender And
CreditRisk.IncomeNewCustomer.Income And CreditRis
k.ProfessionNewCustomer.Profession
15
Schema Rowsets(????)
  • ????????????
  • OLE DB for DM ??Schema Rowsets??
  • Mining_Services
  • Mining_Service_Parameters
  • Mining_Models
  • Mining_Columns
  • Mining_Model_Contents
  • Model_Content_PMML

16
??????????
17
????????????
18
(No Transcript)
19
?????????
20
??
Customer
CustomerGuid
DayTimeOnLine
NightTimeOnLine
BrowserType
EmailTime
ChatTime
GeoLocation
WebClick
CustomerGuid
URLCategory
Time
Duration
ReferPage
21
Web????
22
Web ?????
23
????????????
Create Mining Model CustomerClustering (CustomerID
text key, DayTimeOnline long
continuous NightTimeOnline long
continuous, BrowserType text
discrete, ChatTime long continuous, EmailTime
long continuous, GeoLocation text
discrete ) Using Microsoft_Clustering
24
??Customer?WebClick????
Create Mining Model CustomerClustering (CustomerID
text key, DayTimeOnline long
continuous, NightTimeOnline long
continuous, BrowserType text
discrete, ChatTime long continuous, EmailTime l
ong continuous, GeoLocation text
discrete WebClick table ( UrlCategory text key
) )Using Microsoft_Clustering
25
MSFTies??????
26
Web???????
27
?? Microsoft?????????
28
?? Microsoft?????????
Insurance
Business
Loan
No Loan
Insurance
No Insurance
Loan
No Loan
Stock
No Stock
Stock
Loan
Business
No Business
No Stock
Stock
Shopping
No Shopping
Insurance
No Insurance
29
?????
30
?????
Create Mining Model SiteAffiliation (CustomerID
text key, WebClick table predict
( UrlCategory text key ) )Using
Microsoft_Decision_Trees
Insert into SiteAffiliation (CustomerID,WebClick
(skip, UrlCategory)OpenRowset(MSDataShape,
'data providerSQLOLEDBServermyserverUIDme
PWDmypass' , 'ShapeSelect CustomerID from
Customer Append ( Select customerid,
URLCategory from WebClick
relate CustomerID to CustomerID) as WebClick )
31
(No Transcript)
32
????
33
????
34
?????
Select Flattened Topcount((select URLCategory,
adjustedProbability as prob From Predict(Web
Click, INCLUDE_STATISTICS, EXCLUSIVE)), prob, 5)
From WebLog PREDICTION JOIN (select (select
'Business' as URLCategory) union (select
Telecom as URLCategory) as WebClick) as
input On WebLog.Web Click.URLCategory
input.WebClick.URLCategory
35
????
ADO/DSO
????
36
DM?????
37
DM?????
  • ???????Unisys?Microsoft??????
  • ??????????
  • ??????AS2k????????? DM??
  • ????DM?????????
  • ????????????
  • ?????????????SQL Server magazine ??????

38
DMM????
39
??????
40
???MDT???????
1 ???????????????????????(??)????
41
?????????????????
  • ??
  • 1 mm ???
  • 25??
  • 1??????

????
I/P ?? ????
10 4.08
20 7.27
50 31.54
100 40.55
200 129.35
42
???????????????
  • ??
  • 20 ???
  • 25 ??
  • 1 ??????

????
?? ????
10,000 0.38
1,000,000 11.32
5,000,000 34.19
10,000,000 100.53
43
??MDT???????
2 ???????????????????????????
44
????????????????????
  • ??
  • ???????25
  • ????200
  • ???????

????
???? ????
10,000 15.09
50,000 67.79
100,000 120.88
200,000 240.62
45
????????????????????
  • ??
  • 200000 ???
  • 1000?????

????
???? ????
10 85.26
25 120.82
50 172.96
100 281.65
46
????
  • DM URL
  • www.microsoft.com/data/oledb
  • www.microsoft.com/data/oledb/DMResKit.htm
  • ???
  • Microsoft.public.SQLserver.datamining
  • Communities.msn.com/AnalysisServicesDataMining
  • ???
  • ?????
  • www.unisys.com/windows2000/default-07.asp
  • www.microsoft.com/SQL/evaluation/compare/analysisd
    mwp.asp

47
????????,??????????????
  • ?????????
  • http//www.microsoft.com/china/community

48
(No Transcript)
Write a Comment
User Comments (0)
About PowerShow.com