IBM Switch 15 Manuale Utente

iIBM SPSS Modeler 15 User’s Guide

C Notices 249Index 252x

90Chapter 5Encrypting and Decrypting InformationWhen you save a stream, node, project , output file, or model nugget, you can encrypt it to preve ntits

91Building Streams Models palette (.gen) Nodes (.nod) Output (.cou) Projects (.cpj)Opening New FilesStreams can be loaded direc tly from the File

92Chapter 5Map to.This method starts with the node to be introduced to the stream. First, right-click the nodeto in troduce;then, using the Data Mappi

93Building StreamsStep 3: Replace the template source node.Using the Data Mapping option o n the pop-up menufor the template source node, click Select

94Chapter 5Mapping between StreamsSimilar to connecting nodes, this method of data mapping does not require you to set es sentialfields before hand. Wi

95Building StreamsTo Set Essential FieldsE Right-click the source node of th e template stream that will be replaced.E On the menu, click:Data Mapping

96Chapter 5Mapped.Lists the fields selected for mapping to template fields. These are the fields whose namesmay have to change to match the original field

97Building StreamsFigure 5-51ToolTip and custom node nameInsert values automatically into a CLEM expression.Using th e Expression Builder, accessible

98Chapter 5Figure 5-53Selecting the Demos folder from the list of recently-used directoriesMinimize output window clutter.You can close and delete ou

Chapter6666Handling Missing ValuesOverview of Missing ValuesDuring the Data Preparation phase of data mining, you will of ten want to replace missing

Chapter1111About IBM SPSS ModelerIBM® SPSS® Modeler is a set of data mining tools that enable you to quickly develop predictivemodels using business e

100Chapter 6Figure 6-1Specifying missing values for a continuous variableReading in mixed data.Note that when y ou are reading in fields with numeric s

101Handling Missing ValuesIn general terms, there are two approa ches you can follow: You can exclude fields or re cords with missing values You can

102Chapter 6Screening or Removing FieldsTo screen out fields with too many mis sing value s, you have seve r al options: You can use a Data Audit node

103Handling Missing ValuesThe @ functions can be used in conjunction with the @FIELD function to identify th e presen ceof blank or null values in one

104Chapter 6Note on Discarding RecordsWhen usin g a Select node to discard records, note that syntax uses three-valued logic andautomatically inc lude

Chapter7777Building CLEM ExpressionsAbout CLEMThe Control Language for Expression Manipulation (CLEM) is a powerful language for analyzingand m anipul

106Chapter 7Figure 7-1Derive node creating a new field based on a formulaCLEM expressions can also be used for global search and replace operations .

107Building CLEM ExpressionsFigure 7-2Filler node replacing system-missing values with 0More complex CLEM e xpressions can also be created. For exampl

108Chapter 7Figure 7-3Conditional Derive comparing values of one field to those of the field before itCLEM ExamplesTo illus trate c orrect syntax as w

109Building CLEM ExpressionsComplex ExpressionsExpress ions can also be lengthy and more complex. The following expression returns true if thevalue of

2Chapter 1IBM SPSS Modeler ServerSPSS Mod eler uses a client/ser ve r architecture to distribute requ ests for resource-intensiveoperations to powerfu

110Chapter 7Frequently, special functions are used in combination, which is a commonly used method offlagging blan ks in mor e than one field at a time.

111Building CLEM ExpressionsCharacters—Alwa ys use single backquotes like this ` . For example, note the character d inthe function stripchar(`d`,&qu

112Chapter 7If you want to ov erride precedence, or if you are in any doubt of the or der of evaluation, you ca nuse parentheses to make it explicit —

113Building CLEM Expressions Determining the length (number of c haracters) for a string variable—length(STRING). Checking the alphabe tical orderin

114Chapter 7Figure 7-4Filler node replacing system-missing values with 0For more information, see the topic Functions Handling Blanks and Null Values

115Building CLEM ExpressionsCalculating Time PassedYou can easily calculate the time passed from a baselin e date using a family of functions similar

116Chapter 7You can also use a number of counting functions to obtain counts of values that me et specificcriteria , even w he n th ose values are stor

117Building CLEM ExpressionsWorking with Multiple-Response DataA n umber of comparison functions can be used to analyze multiple-response da ta, inclu

118Chapter 7functions. In a ddition, the Builder controls automatically add th e proper quotes for fields andvalues, making it easier to create syntact

119Building CLEM ExpressionsAccessing the Expression BuilderThe Expression Builder is available in all nodes where CLEM e xp r essions are used, inclu

3About IBM SPSS Modelercan be shared by multiple users, or accessed from the thin-client applicat ion IBM SPSS ModelerAdvantage. You install the adapt

120Chapter 7E Double-click or click the yellow arrow bu tton to add the field or function to the expression field.E Use the operand buttons in the cente

121Building CLEM ExpressionsAfter you ha ve selected a group of functions, double - click to insert the func tions into theexpression field at the poin

122Chapter 7Viewing or Selecting ValuesField values can be viewed from a number of places in the system, includin g the Express ionBuilder, data audit

123Building CLEM ExpressionsChecking CLEM ExpressionsClick Check in the Expression Builder (lower right corner) to validate the expression. Expression

124Chapter 7Figure 7-12Find/Replace dialog boxE With th e cursor in a text area, pr ess Ctrl+F to access the Fi nd/Replace dialog box.E Enter the tex

125Building CLEM ExpressionsCharactersMatches\0nn The character wi th octal value 0nn (0 <= n <= 7)\0mnn The character with octal value 0mnn (0

126Chapter 7Boundary matchers Matches\Z The end of the input but for the final terminator, if any\z The end of the input

Chapter8888CLEM Language ReferenceCLEM Reference OverviewThis section describes the Control Language for Expression Manipulation (CLEM), which is apow

128Chapter 8For more information, see the topic Values a nd Data Types in Chapter 7 on p. 110. Additionally,these rules are cov ered in more detail in

129CLEM Language ReferenceStringsGenerally, you should enclos e strings in double quotation marks. Exampl es of strings are"c35product2" and

4Chapter 1IBM SPSS Modeler DocumentationDocumentation in onli ne help format is available from the Help menu of S PSS Modeler. Thisincludes documentat

130Chapter 8Format ExamplesDD/MM/YYYY 15/01/1963MM/DD/YY 01/15/63MM/DD/YYYY 01/15/1963DD-MM-YY15-01-63DD-MM-YYYY15-01-1963MM-DD-YY 01-15-63MM-DD-YYYY0

131CLEM Language ReferenceFormat ExamplesMM.SS55.58, 01.00(H)H.(M)M.(S)S12.1.12, 1.1.1, 22.12.12(H)H.(M)M12.23, 7.45, 22.7(M)M.(S)S55.58, 1.0CLEM Oper

132Chapter 8Operation Comments Precedence (see next section)&&Used between two integers. The re sultis the bitwise ‘and’ of the in t egers IN

133CLEM Language ReferenceFunctions ReferenceThe following CLEM f unctions are ava ilable f or working with data in IBM® SPSS® Modeler.You can enter t

134Chapter 8Convention DescriptionINT, INT1, INT2Any in t eger, such as 1 or –77.CHARA character co de, such as `A`.STRINGA strin g, such as "ref

135CLEM Language ReferenceConversion FunctionsConversion functions allo w you to construct n ew fields and convert the storage type of existingfiles. Fo

136Chapter 8FunctionResultDescriptioncount_greater_than(ITEM1,LIST)IntegerReturns the number of values from a list of fiel ds that aregreater than ITEM

137CLEM Language ReferenceFunctionResultDescriptionmax_n(LIST)NumberReturns the maximum value from a l i st of numeric fields ornull if all of the fiel

138Chapter 8Numeric FunctionsCLEM contain s a number of commo nly used numeric functions.FunctionResultDescription–NUMNumberUsed to negate NUM. Return

139CLEM Language ReferenceFunctionResultDescriptionmean_n(LIST)NumberReturns the mean value from a list of numeric fields or null ifall of the field val

5About IBM SPSS ModelerIBM SPSS Modeler Administration Console User Guide.Information on installing and using theconsole user interface for monitorin

140Chapter 8FunctionResultDescriptioncdf_normal(NUM, MEAN,STDDEV)RealReturns the probability that a value from the norm aldistribution with the specifi

141CLEM Language ReferenceFunctionResultDescriptioninteger_bitcount(INT)IntegerCounts t he number of 1 or 0 bits in thetwo’s-com plement representatio

142Chapter 8In CLEM, a string is any sequence of characte r s between match ing double quotation marks("string quotes"). Characters (CHAR) c

143CLEM Language ReferenceFunctionResultDescriptionismidstring(SUBSTRING, STRING)IntegerIf SUBSTRING is a substring of STRINGbut does not start on the

144Chapter 8FunctionResultDescriptionlocchar(CHAR, N, STRING)IntegerUsed to identify the location of charactersin symbolic fields. The function searche

145CLEM Language ReferenceFunctionResultDescriptionstripchar(CHAR,STRING)StringEnables you to remove specified charactersfrom a string or field. You can

146Chapter 8FunctionResultDescriptionunicode_char(NUM)CHARReturns the character with Unicode v al ueNUM.unicode_value(CHAR)NUMReturns the Unico de val

147CLEM Language ReferenceNote: Date and time functions cannot be called from scripts.FunctionResultDescription@TODAYStringIf you select Rollover days

148Chapter 8FunctionResultDescriptiondate_in_weeks(DATE)RealReturns the time in weeks from the baseline date tothe date or timestamp represented by DA

149CLEM Language ReferenceFunctionResultDescriptiondatetime_time(ITEM)TimeReturns the time value of the given ITEM.datetime_timestamp(YEAR,MONTH, DAY,

6Chapter 1Demos FolderThe data files and sample streams used with the app lication examples are installed in the Demosfolder under the pr oduct install

150Chapter 8FunctionResultDescriptiontime_mins_difference(TIME1,TIME2)RealReturns the time difference in minutes between thetimes or timestamps repres

151CLEM Language Reference Sequence functions Record indexing Averaging, summing , and comparing value s Monitoring change—differentiation @SINCE

152Chapter 8For this reason, @SINCE does not evaluate its condition for the current record. Use a sim ilarfunction, @SINCE0, if you want to evaluate t

153CLEM Language ReferenceFunctionResultDescription@MAX(FIELD, EXPR)NumberReturns the maximum value for FIELD over the lastEXPR records received so fa

154Chapter 8FunctionResultDescription@OFFSET(FIELD, EXPR, INT)AnyPerforms the same operat i on as the @OFFSET functionwith t he addition of a third ar

155CLEM Language ReferenceFunctionResultDescription@SUM(FIELD, EXPR)NumberReturns the sum of values for FIELD over the lastEXPR records receive d by t

156Chapter 8FunctionResultDescription@GLOBAL_MIN(FIELD)NumberReturns the mini mum value for FIELD over the wholedata set, as previously ge nerated by

157CLEM Language ReferenceSpecial FieldsSpecial functions are used to denote the specific fields under examination, o r to generate a listof fields as in

Chapter9999Using IBM SPSS Modeler with aRepositoryAbout the IBMSPSS Collaboration and Deployment Services RepositoryIBM® SPSS® Modeler can be used in

159Using IBM SPSS Modeler with a RepositoryFigure 9-1Objects in the IBM SPSS Collaboration and Deployment Services RepositoryExtensive Versioning and

Chapter2222New FeaturesNew and Changed Features in IBM SPSS Modeler 15From this release onwards , IBM® SP SS® Modeler has the following editions.IBM®

160Chapter 9For more informati on, see th e topic Connecting to the Repo sitory on p. 161.Storing and Deploying Repository ObjectsStreams created in I

161Using IBM SPSS Modeler with a RepositoryOther Deployment OptionsWhile IBM SPSS Collabora tion and Deployment Services offers the most extensive fea

162Chapter 9Ensure secure connection.Specifies whether a Secure Socke ts Layer ( SSL) connection shouldbe used. SSL isa commonly used protocol for secu

163Using IBM SPSS Modeler with a RepositoryFigure 9-4Browsing the IBM SPSS Collaboration and Deployment Services Repository contentsThe explo r er win

164Chapter 9Storing Objects in the RepositoryFigure 9-5Storing a modelYou can store streams, nodes, models, model palettes, projects, and output objec

165Using IBM SPSS Modeler with a RepositoryChoosing the Location for Storing ObjectsFigure 9-6Choosing the location for storing an objectSave in.Shows

166Chapter 9Figure 9-7Adding information about the objectAuthor.The username of the user creating the object in the repository. By default, this shows

167Using IBM SPSS Modeler with a RepositoryAssigning Topics to a Stored ObjectTopics are a hie r archical classification s ystem for the content s tore

168Chapter 9Figure 9-9Setting security options for an objectPrincipal.The repository username of the user or group who has access rights on this objec

169Using IBM SPSS Modeler with a RepositoryAdding a User to the Permissions ListFigure 9-10Adding a user to the permissions list for an objectSelect p

8Chapter 2Default settings for database connections.You can now specify default settings for SQL Serverand Or acle database connections, as wel l as t

170Chapter 9Read.By default, a user or gr oup that is not the object owner has only Read access rights to theobject. Select the appropriate check boxe

171Using IBM SPSS Modeler with a RepositoryE Specify connection settings to the re pository if necessary. For more information, see thetopic Co nnecti

172Chapter 9Storing Models and Model PalettesYou can store an individu al model as a .gm file in the repository, from whe r e it can be accessedby othe

173Using IBM SPSS Modeler with a RepositoryorFile > Projects > Retrieve Project...orFile > Outputs > Retrieve Output...E Alternatively, ri

174Chapter 9Look in.Shows the fo lder hierarchy for the curr ent f older. To navigate to a different folder, selectone from this list to navigate ther

175Using IBM SPSS Modeler with a RepositoryE Select the object version you want to work with.E Click C ontinue.Searching for Objects in the Repository

176Chapter 9When searching for objects by name, an as terisk (*) can be used as a wildcard character to matchany string of characters, and a question

177Using IBM SPSS Modeler with a RepositoryTopics.You can s earch on mo dels associated with specific topics from a list set by repositoryusers with th

178Chapter 9Figure 9-16Locked objectTo lock an objectE In the repository explorer win dow, right-click the required object.E Click Lock.To unlock an o

179Using IBM SPSS Modeler with a RepositoryFigure 9-17Select versions to deleteManaging Properties of Repository ObjectsYou can control various object

9New FeaturesSQL generation enhancements.The Aggregate nod e n ow supports SQL generation for date, time ,timesta mp, and string data types, in additi

180Chapter 9Displays the folder name, creation, and modification dates.Permissions tabSpecifies read and write permissions for the f older. All users an

181Using IBM SPSS Modeler with a RepositoryE In th e repository window, right-click the required object.E Click Object Properties.Figure 9-20Object pr

182Chapter 9Figure 9-21Version propertiesThe following pr operties can be specified or modified for specific versions of a stored object:Version.Unique i

183Using IBM SPSS Modeler with a RepositoryFigure 9-22Object access rightsUsers And Groups.Lists the repository users and groups that have at leas t R

184Chapter 9To define a new label and apply it to the objectE Type the label name in the New Label field.E Click the right-arrow button to move the new

185Using IBM SPSS Modeler with a RepositoryFigure 9-23Storing a stream in the repositoryE In the Repository: Store dialog b ox, choose the folder wher

186Chapter 9Figure 9-24Stream Deployment optionsDeployment type.Choose how you want to deploy the stream. All streams require a designatedscoring node

187Using IBM SPSS Modeler with a RepositoryScoring node.Select a graph, output or export node to identify the s tream branch to be used forscoring the

188Chapter 9Scoring and Modeling ParametersWhen deploying a stream to I B M SPSS Collaboration and D eployment Serv ices, you can choosewhich paramete

189Using IBM SPSS Modeler with a RepositoryFigure 9-26Stream with scoring branch highlightedIf the stream already had a scoring bra nch defined, the ne

Note: Before using this information and the product it supports, read the general informationunder Notices on p. 249.This edition applies to IBM SPSS

10Chapter 2New features in IBM SPSS Modeler PremiumIBM® SPSS® Modeler Premium is a separately-licensed product that provides a dditional featuresto th

190Chapter 9To designate a branch as the scoring branch (Tools menu)E Connect the model nugget to a terminal node ( a proces sing or output node downs

191Using IBM SPSS Modeler with a RepositorySingle Model in StreamFigure 9-28Scoring branch with single model in the streamIf a single linked model nug

192Chapter 9Figure 9-29Scoring branch with more than one model in the streamYou right-click the Analysis node and use its menu to set the scoring bran

193Using IBM SPSS Modeler with a RepositoryFigure 9-31Scoring branch with refresh link switched to first model nuggetIf you subsequently deselect both

194Chapter 9No Models in StreamIf there are no models in th e stream, or only models with no mode l links, the deployment type isset to Scoring Only.C

Chapter11110000Exporting to External ApplicationsAbout Exporting to External ApplicationsIBM® SPSS® Modeler provides a number of mechanisms to export

196Chapter 10E Specify connection settings to the re pository if necessary. For more information, see thetopic Connecting to the Repository in Chapter

197Exporting to External ApplicationsE In the Export (or Save) dialog box, s pecify a target directory and a unique n ame for the model.Note: You c an

198Chapter 10Figure 10-3Selecting the XML file for a model saved using PMMLUse variable labels if present in model.The PMML may specify both variable

199Exporting to External Applications Neural Net C5.0 Logist ic Regression Genli n SVM Bayes Net Apriori Carm a K-Means Koho nen TwoStep K

11New FeaturesThe Netezza Time Series node analyzes time series da t a and can predict futurebehavior from past events.The Netezza Generalized Li near

Chapter11111111Projects and ReportsIntroduction to ProjectsA project is a group of files r elated to a data mining task. Projects include data streams,

201Projects and ReportsCRISP-DM ViewBy s upporting the Cross-Industry Standard P r ocess f or D ata Mining (CRISP-DM), IBM® SPSS®Modeler projects prov

202Chapter 11Classes ViewThe Classes view in the project pane organizes your work in IBM® SPSS® Mod eler categoricallyby the types of objects created.

203Projects and ReportsE On the main menu, click:File > Project > New Project...Adding to a ProjectOnce you h ave created or opened a project, y

204Chapter 11Adding Nodes from the CanvasYou can add individual nodes from the stream canvas by using the Save dialog box.E Select a node on the canva

205Projects and ReportsTransferring a ProjectMake sure that the project you want to transfer is open in the project p ane.To transfer a project:E Righ

206Chapter 11Summary.You can enter a summary for your data mining project that will be displayed i n theproject report.Contents.Lists the type and num

207Projects and ReportsFigure 11-6Annotations tab in the project properties dialog boxE Enter keywords and text to describe the project.Folder Propert

208Chapter 11Figure 11-7Project folder with CRISP-DM annotationName.This area displays the name of the selected field.Tooltip text.Create custom ToolTi

209Projects and ReportsClosing a ProjectWhen you exit IBM® SPSS® Modeler or open a new project, the existing project file (.cpj)is closed.Some files ass

Chapter3333IBM SPSS Modeler OverviewGetting StartedAs a data mining application, IBM® SPSS® Modeler offers a strategic approach to find ing usefulrelat

210Chapter 11Figure 11-9Generated report windowTo generate a report:E Select the project folder in either CRISP-DM or Classes view.E Right-click the f

211Projects and ReportsFigure 11-10Selecting options for a reportThe options in the report dialog box p r ov ide several ways to generate the t ype of

212Chapter 11File type.Available file types are:HTML document.The r eport is saved as a single HTML file. If your report contains graphs,they ar e save

213Projects and ReportsThe total number of nodes in each stream is listed within the report. The numbers are shownunder the following headings, which

214Chapter 11Figure 11-11Report displayed in a web browser

Chapter11112222Customizing IBM SPSS ModelerCustomizing IBM SPSS Modeler OptionsThere ar e a number of operations you can perform to cu stomize IBM® SP

216Chapter 12Figure 12-1System Options dialog boxMaximum memory.Select to impose a limit in megabytes on SPSS Modeler’s memory usage. Onsome platforms

217Customizing IBM SPSS Modeleris the path used for all client-side operations and output files (if they are referenced withrelative paths).Set Server

218Chapter 12Figure 12-2User Options dialog box, Notifications tabShow stream execution feedback dialog. Sele ct to display a dialog box th at include

219Customizing IBM SPSS ModelerVisual NotificationsThe options in this gr oup are used to specify the behavior of the Outputs and Models tabs in thema

13IBM SPSS Modeler OverviewLaunching from the Command LineYou can use the command line of your oper ating system to lau nch IBM® SPSS® Modeleras follo

220Chapter 12 Select Always to always open a new output window. Select If generated by current stream to open a new window for output generated by t

221Customizing IBM SPSS ModelerStandard Fonts & Colors (effective on restart).Options in this control bo x are used to specify theSPSS Modeler scr

222Chapter 12Figure 12-4User Options dialog box, PMML tabExport PMML.Here you can configure variations of PMM L tha t work best with your targetapplica

223Customizing IBM SPSS ModelerCustomizing the Nodes PaletteStreams are built using nodes. The Nodes Palette at the bottom of the IBM® SPSS® Modelerwi

224Chapter 12Figure 12-6Palette Manager showing the tabs displayed on the Nodes PalettePalette Name.Each available palette tab, whether shown on the N

225Customizing IBM SPSS ModelerCreating a Palette TabFigure 12-7Palette tab creation on the Create/Edit Palette dialog boxTo c r eate a custom p alett

226Chapter 12Figure 12-8Palette Manager showing the tabs displayed on the Nodes PaletteTo selec t which tabs are to be shown on the Nodes Palette:E Fr

227Customizing IBM SPSS ModelerFigure 12-9Subpalettes available for the Modeling Palette tabTo selec t subp alettes for display on a palette tab:E Fro

228Chapter 12the palette tab. For example , if you created a palette tab that contains the nodes you use mostfrequently for creating your streams, you

229Customizing IBM SPSS ModelerTo change the nodes shown on a pa lette tab, selec t the palette tab and then, f rom the menu on theleft, sel ect to di

14Chapter 3To Connect to a ServerE On the Tools menu, click Server Login. The Server Login dialog box opens. Alterna tively,double-click the connectio

Chapter11113333Performance Considerations forStreams and NodesYou can design you r streams to maximiz e performance by arranging the nodes in the most

231Performance Considerations for Streams and NodesThe following operations cannot be performed in most databases. They should be placed in thestream

232Chapter 13Figure 13-1Caching at the Type node to store newly derived fieldsTo Enable a CacheE On the s tream canvas, right-clic k the node and clic

233Performance Considerations for Streams and NodesNote: The following databas es support temporary table s for the purpose of caching: DB2,Netezza, O

234Chapter 13Aggregate.When the Keys are contiguous option is not set, this node reads (but does not store) itsentire input data set before it produce

235Performance Considerations for Streams and Nodesoffset v alue is not a literal integer; for example, @OFFSET(Sales, Month). The offset value is the

AppendixAAAAAccessibility in IBM SPSS ModelerOverview of Accessibility in IBM SPSS ModelerThis release offers greatly enhanced accessibility for all u

237Accessibility in IBM SPSS ModelerControlling the Automatic Launching of New WindowsThe Notifications tab on the User Options dialog box is a ls o us

238Appendix AControlling the Automatic Launching of New WindowsThe Notifications tab on the User Options dialog box is a ls o used to control whethe r

239Accessibility in IBM SPSS ModelerShortcut Key FunctionCtrl+F6 Moves focus to the stream canvas.Ctrl+F7 Moves focus to the managers pane.Ctrl+F8 Mov

15IBM SPSS Modeler OverviewNote: You cannot edit a server connection that was added from IBM® SPSS® Collaborationand Deployment Services, since the na

240Appendix AShortcut Key FunctionCtrl+Alt+L When a model nugget is selected in the stream, opens an Insertdialog box to enable you to load a saved mo

241Accessibility in IBM SPSS ModelerShortcut Key FunctionCtrl+End Wit h focus on any control in the Expression Builder, this w i l lmove the insertion

242Appendix AShortcuts for CommentsWhen working with on-screen comments, y ou can use the following shortcuts.Shortcut Key FunctionAlt+C Toggles the s

243Accessibility in IBM SPSS ModelerCluster Viewer onlyThe Cluste r Viewer has a Cluste r s view that contains a c luster-by-feat ures grid.To choos e

244Appendix AESpacebar.Selects the Variable File node.ECtrl+Enter.Adds the Variable File node to the stream canvas. This key combinat ion also keepsse

245Accessibility in IBM SPSS Modeler Use F3 to destroy all connections for a selecte d node in the canvas. Once you have created a stream, use Ctrl+

246Appendix AAccessibility in the Interactive Tree WindowThe standard display of a d ecision tree model in the Interactive Tree window may cause probl

247Accessibility in IBM SPSS ModelerTyping the first letter to find element in tree list.When looking for an element in the categoriespane, extracted

AppendixBBBBUnicode SupportUnicode Support in IBM SPSS ModelerIBM® SPSS® Modeler is fully Un icode-enabled for b oth IBM® SPSS® Mode ler and IBM®SPSS®

AppendixCCCCNoticesThis information was developed for produc ts and services offered worldwide.IBM may not offer the products , services, or feature s

16Chapter 3Searching for Servers in IBM SPSS Collaboration and Deployment ServicesInstead of entering a server connection manually, you can select a s

250Appendix CSuch information may be a vailable, subject to appropriate terms and conditions, including insome cases, payment of a fee.The licensed pr

251Notices

Index508 compliance, 236abs function, 138accessibility, 2 36, 247example, 243–244features in IBM SPSS Modeler, 236tips in IBM SPSS Mod el er, 246addin

253Indexcommentskeyboard shortcuts, 242listing al l in a stream, 84on nodes an d st reams, 78comparison functions, 135concatenating strings, 135condit

254Indexdiv function, 138documentation, 4domain name (Windows)IBM SPSS Mod el er Server, 13DTD, 197enable nodes , 46encoding, 56, 248endstring functio

255Indexoptions, 215overview, 12 , 215running from command li ne, 13tips and shortcuts, 96IBM SPSS Modeler Advantage, 160, 184IBM SPSS Modeler Serverd

256IndexMEAN function, 150, 152@MEAN function, 150, 152mean_n function, 115, 138member function, 135memorymanaging, 215–216Merge nodeperformance, 233m

257Indexobjectsproperties, 208OFFSET function, 152@OFFSET function, 150, 152performance considerations, 234oneof function, 141openingmodels, 90nodes,

258Indexrefreshsource nodes, 57refreshing models, 190regression, 245rem function, 138renamingnodes, 86streams, 74replace function, 141replacing models

259Indexstream rewritingenabling, 60streams, 12adding comments, 78adding nodes, 43,46adding to projects, 202–203annotating, 78, 86backup files, 88build

17IBM SPSS Modeler OverviewE Edit options.cfg, located in the /config directory of your SPSS Modeler installation directory. Editthe temp_directory par

260Indexuppertolower function, 141user IDIBM SPSS Mod el er Server, 13user options, 217user-defined functions (UDFs), 120user-missing values, 99UTF-8 e

18Chapter 3This sequence of operations is known as a data stream becau se the data flows record b y recordfrom the source through each manipulation and

19IBM SPSS Modeler OverviewField Ops.Nodes perfor m operations on data fields, such as filtering, deriving new fields, anddetermining the measurement l

PrefaceIBM® SPSS® Modele r is the IBM Corp. enterpr is e-strength data mining workbench. SPSSModeler helps o rganizations to improve customer an d cit

20Chapter 3Figure 3-8Outputs tabThe Models tab is the most powerful of the manag er tabs. This tab contains all model nuggets,which contain the models

21IBM SPSS Modeler OverviewFigure 3-10CRISP-DM viewThe C lasses tab provides a way to organize your work in SPSS Modeler categorically—by thetypes of

22Chapter 3Cut & move to clipboard Copy to clipboardPaste selection Undo last actionRedo Search for nodesEdit stream properties Preview SQL genera

23IBM SPSS Modeler OverviewCustomizing the ToolbarYou can change various aspects of the toolb ar, such as: Whether it is display ed Whether the icon

24Chapter 3Figure 3-12Maximized stream canvasAs an alternative to closing the nodes palette, and the managers and project panes, you can use thestream

25IBM SPSS Modeler OverviewFigure 3-13Changing the icon sizeTo scale the entire stream (stream properties method)E From the ma in menu, chooseTools &g

26Chapter 3Using the Mouse in IBM SPSS ModelerThe most common uses of the mouse in IBM® SPSS® Modeler include the following:Single-click.Use either t

27IBM SPSS Modeler OverviewTable 3-2Supported shortcuts for old hot keysShortcutKeyFunctionCtrl+Alt+D Duplicate nodeCtrl+Alt+L Load nodeCtrl+Alt+R Ren

28Chapter 3revenue data or as complex as transforming web log data into a set o f fields and records withusable information. For more information, see

Chapter4444Understanding Data MiningData Mining OverviewThrough a variety of techniqu es, data mining identifies nuggets of information in bodies o f d

Contents1 About IBM SPSS Modeler 1IBM SPSS Modeler Products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

30Chapter 4Typically, you will use these facilities to identify a promising set of attributes in the data. Theseattribut es can then be fed to the mod

31Understanding Data Miningtogether. It may not even be onl ine. If it exists only on p aper, data entry will be r equired beforeyou can begin data mi

32Chapter 4A Strategy for Data MiningAs with most business endeavors, data mining is much more effective if done in a planned,systematic way. Even wit

33Understanding Data MiningFigure 4-1CRISP-DM process modelThe six phases inclu de:Business understanding.This is perhaps the most import ant phase o

34Chapter 4have been resolved adequately. Similarly, the evaluation phase can lead you to reevaluate youroriginal business u ndersta nding, and you ma

35Understanding Data MiningClassification nodesThe Auto Classifier node creates and compar es a number of different models forbinary outcomes (yes or no

36Chapter 4The PCA / Factor node provides powerful data-reduction techniques to reducethe complexity of you r data. Principal components analysis (PCA

37Understanding Data MiningThe Self-Learning Resp onse Model (SLRM) node enables you to build a model inwhich a single new case, or small number of ne

38Chapter 4preconditions. Apriori requires that input and output fields all be categorical butdelivers better performance because it is opti mized for

39Understanding Data MiningSegmentation nodesThe Auto Cluster node estimates and compares clustering models, which identifygroups of records that have

Changing the icon size for a stream . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24Using the Mouse in IBM S

40Chapter 4Data Mining ExamplesThe be st way to learn about data mining in practice is to start with an example. A number ofapplicat ion examples are

Chapter5555Building StreamsStream-Building OverviewData mining using IBM® SPSS® Modeler focuses on the process of running data through a serie sof nod

42Chapter 5Figure 5-1Completed stream on the stream canvasThis section cont ains more detailed information on w orking with nodes to create more compl

43Building StreamsA runnable node that processes stream data is known as a terminal node. A modeling oroutput node isa ter minal node if it is located

44Chapter 5Figure 5-2Stream created by double-clicking nodes from the palettesTo Connect Nodes Using the Middle Mouse ButtonOn the str eam canvas, you

45Building StreamsFigure 5-5Connected nodesWhen connecting nodes , there are severa l gu idelines to follow. You will receive an error mes sageif you

46Chapter 5Disabling Nodes in a StreamProcess nodes with a single input within streams can be disabled, with t he result that the node isignored durin

47Building StreamsFigure 5-8Connecting a new node between two connected nodesE With th e middle mous e button, click and drag the connection arrow int

48Chapter 5Figure 5-10Deleting the connection between nodes in a streamTo delete all connections to and from a node, do one of the following: Select

49Building StreamsFigure 5-11Pop-up menu options for nodes Click Edit to open the dialog box f or the s elected node. Click Connect to manually conn

7 Building CLEM Expressions 105About CLEM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

50Chapter 5 Click Save Node to save the node’s details in a file. You can load node details only intoanother node of the same type. Click Store Node

51Building StreamsFigure 5-12Caching at the Type node to store newly derived fieldsTo Enable a CacheE On the s tream canvas, right-clic k the node and

52Chapter 5Note: The following databas es support temporary table s for the purpose of caching: DB2,Netezza, O r acle, SQL Server, a nd Teradata. Othe

53Building StreamsFigure 5-13Data Preview from a model nuggetFrom the Gen erate menu, you can create several types of nodes.Locking NodesTo prevent ot

54Chapter 5Figure 5-14Streams tab in the managers pane with pop-up menu optionsFrom this tab, you can: Access streams. Save streams . Save streams

55Building StreamsLogging and status.Options controlling SQL logging and record sta tus. F or more information,see the topic Setting SQL logging an d

56Chapter 5Figure 5-15Setting general options for a streamDecimal symbol.Select either a com ma ( ,) or a period (.) as a decimal sep arator.Grouping

57Building StreamsMaximum number of rows to show in Data Preview.Specif y the number of row s to be shown whena preview of the data is requested for a

58Chapter 5Figure 5-17Setting date and time options for a streamImport date/time as.Select whether to use date/time storag e for date /time fields or w

59Building Streams2-digit dates start from.Specif y the cutoff yea r to add cen tury dig its f or yea r s denoted with onlytwo digits. For exam ple, s

Logical Functions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137Numeric Function

60Chapter 5Decimal places (standard, scientific, currency).For numbe r display formats, specifies the numberof deci mal places to be used when displayi

61Building StreamsFigure 5-19Setting stream optimization optionsNote: Whether SQ L pushback an d optimization ar e supported depends on the type of da

62Chapter 5reduce network traffic and speed stream operatio ns. Note that the Generate SQL check bo xmust be selected for SQL optimization to have any

63Building StreamsSetting SQL logging and record status options for streamsThese setting s include various options controlling the display of SQL stat

64Chapter 5Reformat SQL for improved readability.Specifies whether SQL displayed in the log should beformatted forreadability.Show status for records.S

65Building StreamsStream scroll rate.Specify the scrolling rate for the stream canvas to control how quickly thestream canva spane scrolls when a node

66Chapter 5Figure 5-22Messages tab in stream properties dialog boxIn addition to mess ages regarding stream operation s, err or m essages are reported

67Building StreamsFigure 5-23Stream running with error reportedIf SQL optimization and logging options are enabled in the User Options dialog box, the

68Chapter 5Figure 5-24Viewing execution times for nodes in the streamIn the table of node execution times, the columns are as follows. Click a column

69Building StreamsParameters can also be set for SuperNodes, in which case they are visible only to nodesencapsulatedwithin that SuperNode.To Set Stre

Deploying Streams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184Stream Depl

70Chapter 5Note that long name, storage, and type options can be set for param eters through the userinterface only. These options cannot be set using

71Building StreamsFigure 5-27Specifying available values for a parameterType.Displays the currently selected measurement leve l. You can change this v

72Chapter 5Analytical De cision Management or Predictive Applications 5.x. All streams r equire a designatedscoring br anch before they can be deploye

73Building StreamsSearching for Nodes in a StreamYou can search for nodes in a stream by sp ecifying a number of sea r ch criteria, such as node name,

74Chapter 5Node category.Check t his box and click a categ ory on the list to sear ch for a particular type ofnode. ProcessNode means a node from the

75Building StreamsFigure 5-30Opening section of stream descriptionThe stream description is display ed in the for m of an HTML d ocument consisting of

76Chapter 5Inputs.Lists the input fields together with their storage types (for example, string, integer,real and so on).Outputs.Lists the o utput fie

77Building StreamsExporting Stream DescriptionsYou can export the contents of the stream description to an HTML file.To e xport a stream description:E

78Chapter 5Some nodes have further displays givin g additional information about s tream execution. Theseare displayedby selecting the corresponding r

79Building StreamsFigure 5-34Stream with comments addedOthers can then view these comments on-screen, or you can print out an image of th e streamthat

Customizing the Nodes Palette . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223Customizing the P

80Chapter 5The appearance of the tex t box changes to indicate the current mode of the comment (or an notationshown as a comment), as the fo llowing t

81Building StreamsFigure 5-37Comment in edit modeWhen you click again, the border changes to solid lines to show that editing is complete.Figure 5-38C

82Chapter 5 Right- click the stream background and click New Comment on the pop-up men u. Click the New Comment button in the too lbar.E Enter the c

83Building StreamsE Edit the commen t text. You can use standard Windows shortcut ke ys when editing, for exampleCtrl+C to copytext. Other options dur

84Chapter 5If the comment was o r iginally a stream or SuperNode an no tation that had been converted to afreestandingcomment, the comment is deleted

85Building StreamsFigure 5-39Listing comments for a streamText.The text of the comment. Double-click the text to change the fiel d to an editab le text

86Chapter 5E Click the Annotations tab.E Select the Show annotation as comment check box.E Click OK.To convert a SuperNode annotation to a commentE Do

87Building StreamsFigure 5-40Annotations tab optionsName.Select Custom to adjust the autogenerated name or to create a uniqu e name for the nodeas dis

88Chapter 5Show annotation as comment.(For stream and SuperN ode annotat ions only) Check this boxto con vert theannotation to a freestanding comment

89Building StreamsSaving Multiple Stream ObjectsWhen you exit IBM® SPSS® Modeler wi th multiple unsaved objects, such as streams, projects,or model nu