Educational Data Mining Workbench User Manual V3.00
Contents
1. AddProccess Clip Sampling Labeling Logs of Studens in Section AE Row Sample Name Anon Student Id SessionId Time Time Zone Duration sec Student Response Type Student Response Subtype Tutor Response Type Tutor Section A txt 1 All Data Stu_043c2ec6c6390dd0ac5519190a57c88c _ PCT50877 _ 2005 10 15 02 08 56 0 UTC _ ATTEMPT RESULT ection A txt 2 All Data Stu _043c2ec6c6390dd0ac5519190a57c88c PCT50877 2005 10 15 02 09 00 0 JUTC 4 ATTEMPT SULT ection A txt 3 All Data Stu_043c2ec6c6390dd0ac5519190a57c88c P 2005 10 15 02 09 18 0 18 SULT Section Axt P pe Stu_043c2ec6c6390dd0ac5519190a57c88c P 2005 10 15 02 09 21 0 2 ATTEMPT RESULT Seana e Mib iu leese PCTSNR7 2005 10 15 0208 540 jure 3 ameer OO Su ection A txt 8 All Data Stu_043c2ec6c6390dd0ac5519190a57c88 P 2005 10 15 02 10 08 0 8 ATTEMPT RESULT econ Abt 3 Me Stu D43c2ec6c6390dd0ac5519190a57c88c_PCTS0877 _ 200s 10 t502 10 10 0 Tc 2 eero O FS ection A txt 10 All Data Stu_043c2ec6c6390dd0ac5519190a57 88c PCT50877 2005 10 15 02 10 29 0 RESULT ection A txt u All Data Stu_043c2ec6c6390dd0ac5519190a57 88c PCTS0877 2005 10 15 02 10 30 0 UTC RESULT Section Abt 2 Me Stu D43c2ec6c6390dd0ac5519190a57c88c_PCTs0877 00510150210310 alter P Section A txt 13 All Data e 50877 2005 10 15 02 10 55 0 UTC 24 ATTEMPT RESULT ection A t2. Stratified set the number of samples in the Samples textbox and in the Strata list select the column name and log grouping Fig 22 Sampling Method Stratified Set Samples Size Maximum Sample Size 16010 Anon Student Id Session Id Time Time Zone Duration sec Figure 22 Stratified Sampling Ateneo Laboratory for the Learning Sciences F206 AdMU Page 19 Educational Data Mining Workbench User Manual V3 00 3 3 Save Button Save Button allows the user to save the set properties into a sampling xmi file 3 4 Load Button Through the Load button the system allows the user to choose a template from a list and then inherit its properties zi Load Box Select File Name Sample RandomsSampling sampling xmi v Figure 23 Load prompt 3 5 Submit Button It closes the Sampling Form implements the sampling process and then displays the result in a new tab Ateneo Laboratory for the Learning Sciences F206 AdMU Page 20 4 2 43 4 4 4 5 Educational Data Mining Workbench User Manual V3 00 4 Add Process This allows the user to add multiple processes and run them in sequence Required Columns Logs of Studens in Section A E Row Sample Name Anon Student Id Session Id Time Time Zone Duration sec Student Response Type Student Response Subtype Tutor Response Type Tutor Response Subtype Level Unit Level Section Problem Name Step Name
3. Attempt At Step Outcome Selection Action Input Feedback Text Feedback Classification deti mml Figure 24 Add Feature Screen Add Feature Button Allovys user to add a function and edit its functions Add Clipping Allows user to set the desired clipping properties See Clipping and when done the form applies the selected properties in the clipping form Add Sampling Allows user to set desired sampling properties See Sampling and when done the form gets the sampling properties set in the sampling form Cancel button Cancels and closes the Add Process form Save Button The system shall save all the properties set in the Processes List which are checked in a process xmi file Ateneo Laboratory for the Learning Sciences F206 AdMU Page 21 Educational Data Mining Workbench User Manual V3 00 4 6 Load Button The system will load the user selected process xml file upon clicking the load button 47 Run Process Button The system runs all checked processes distil clipping and or sampling in the process list The system shall display information feedback in the Status Bar on what process it is currently taking and throws an error dialogue when the system encounters error Add Process Required Columns Row l My Row Sample Name Anon Student Id New Time Session Id l Size Clip Process Time E g d il Duration sec Student Response Type Student Response Subtype
4. File iis gt n Save File as Import Log Filet end Datashop File and or tat file le Save Template caxtandsy F Clip by Size a p extends Add Features Load Template Ve stance exen n s cade L dse l nden Clip by Time Figure 2 EDM System Map Load Features Zeep i Save Features Add Clipping Ateneo Laboratory for the Learning Sciences F206 AdMU Page Educational Data Mining Workbench User Manual V3 00 I System Overview ee eee File Functions Help wr M D A B Load Save Import Export Add Proccess Clip Sampling Labeling Row Count ei EDM Workbench git 20120214 Figure 3 The EDM Workbench upon system launch This section discusses the interface of the system from Top to Bottom including its features buttons and functions The discussion will only consist of a feature and function description and does not go into how these functions are carried out 1 Title Bar dj EON Workbench E Figure 4 System Title Bar The name of the system may change in later versions e g EDM VVorkbench version 4 is displayed here 2 Menu Bar File Function Help Figure 5 EDM Menu Bar Composed of 3 Menu options File Function and Help consisting of actions buttons Ateneo Laboratory for the Learning Sciences F206 AdMU Page 8 Educational Data Mining Wo
5. Sample zip Netyvork Files of type EDM Files zip Figure 36 EDM file load window 8 Export Current Tab By clicking the export button located either in the File menu Figure 5 1 or Toolbar Figure 6 the system will save the current active tab into a csv file or into another specified format Users must specify the directory in which the file will be saved Note Exporting a file will take time depending on the datasets size Save Save in sh Sample export vd F ete EN Sample Export a Recent Items My Documents e En Computer d File name Sample EDM Export Network Files of type EDM Files csv Figure 37 Export window Ateneo Laboratory for the Learning Sciences F206 AdMU Page 28
6. Educational Data Mining VVorkbench User Manual V3 00 EDM VVorkbench User manual Version 3 00 Prepared by Bon Teogene Balonzo Gamaliel Dela Cruz Francis Bautista Ateneo Laboratory for the Learning Sciences F206 AdMU Page 1 Educational Data Mining VVorkbench User Manual V3 00 Table of Contents Klod SE E 4 ergeet Ee EE 5 DO 5 bel Gje GODO ctne mecca E E E E 5 KK ege KE 5 No e egeedeeegeeeedeeeegedeeedeengdeeerereee Fi NS IV eee r N A A E E E E 8 le KT E 8 2 VMN By Ae saree Sets circle cis ste ci a te ace a cis a ice ss chee te te aie eee secrete Sets dee et ate ete te eee a es ae es ected 8 OO ba a nh ee ne nC ene ee ee ee ene en nC eee 10 BEEG i t E 1 Ae NEN NT NYT 11 How EE e lem EEEE 4 DO ctioesacaarescseuseacmaaseaseacensc 12 SE DP ance gee aa cin genic in scinsse cie ge ea tenes ss seein a ec eee 14 HE E E e e DT I 15 Custom Sort Button EE 16 Time e KE 16 Cancel Button aaa Ba dv dd d 16 Save Button EEN 17 ood e e ME 17 Senati do La en RE VER NENE neat erat VER eer eeee erat VER ENET E VER ENTER E tirrrerseereterer EEE 18 e TW r E R E se 18 Randon aPN O oreen E E E EEEE S 18 Stratified Sampling cccccssssssccccsessceccessessceccccesesscecscessessceescecsessneescecsssssaescecessssenacens 19 av DINI E 20 Load Burton 20 Din ON e sdaaatanstnestcocisnateastonetccataeatanstencteooi weaisestaeoheocixacieasteacmeectiaataesteaumeactince 20 4 Funetion Addition PTOG SS nc oret ona SVEN EE
7. Form as an XML file It will display a dialogue which will ask for a file name see Figure 19 and when submitted it will store the XML file in a folder which contains the XML clipping files Input File Name Clipping Sample Time Figure 18 Save Dialogue 2 5 Load Button Allows the user to select and choose a clipping xmi file from a list When selected the system loads the properties from the xml file to the system see Figure 20 amp 21 Clipping Box Select Column Name For Clip Grouping Select Clip Type Size Logs of Studens in Section A E Clip Size Custom Sort Complete clips only 2 een Select File Name Clipping Sample Time clipping xml D dette Figure 19 Initial load window Ateneo Laboratory for the Learning Sciences F206 AdMU Page 17 Educational Data Mining Workbench User Manual V3 00 Note From the list of clipping xml files the selected template is Clipping Sample Time clipping xml 2 6 Submit Button This closes the Clipping Form clips the dataset from the current tab and displays it with its properties set in a new tab Double click a row to view the logs within it EDMWorkbench File Functions Help Di fa SH e e aL Ke Load Save Import Export AddProccess Clip Sampling Labeling E Datashop__ Datashon cere 9 clip dipSize Logs of Studens in Section A E de 3202 ISection C txt jdip3 13202 Section D txt dp 3202 Se
8. S Er P RNA ESERSE EEEE EA 21 Ateneo Laboratory for the Learning Sciences F206 AdMU Page 2 Educational Data Mining VVorkbench User Manual V3 00 Add Feature Button ENEE 21 Te NDDIM LAEE EE S A A 21 EE 21 Go e e ERE PN RE 21 Ave DALON eee 21 TOA RE Ce 1 00ssecroveenederssisneteversssdeeutinrevonetmentenstionnsioostssstuauberebenetoostenstienntionssssstuenterateontienetead 22 R n Process Button nen eee ian aciesontan enietiontautectevow sutantowawestentowieneulestovian euiestoweneuleniowtacewenostates 22 E I E ace eee 24 non a T clk omeeeeerere terrence rerennrr etree tere eErr renters tere E ere tere tenn tere ereterrrenr tern err teres 24 D E PO 25 e 27 TAG FD BN E 27 po 28 Ateneo Laboratory for the Learning Sciences F206 AdMU Page 3 Educational Data Mining VVorkbench User Manual V3 00 Revision History J Contillo 20120221 User manual for version 2 V2 00 Gamaliel dela Cruz 20120526 Edit content V3 00 Francis Bautista 20120607 Formatting and editine V3 00 Ateneo Laboratory for the Learning Sciences F206 AdMU Page 4 Educational Data Mining Workbench User Manual V3 00 Introduction a Purpose The purpose of this document is to support the user in understanding the EDM workbench s features functions and their usage This document is intended for the project stakeholders and the development team b Project Scope The EDM Workbench is a tool that helps researchers with processing da
9. Tutor Response Type Tutor Response Subtype Level Unit Uncheck All Process Level Section Problem Name Check All Process Invert Checked Process Feedback Classification Help Level Figure 25 Process list Ateneo Laboratory for the Learning Sciences F206 AdMU Page 22 Educational Data Mining Workbench User Manual V3 00 daj EDMWorkbench File Functions Help io gr e Load Save Import Export fx c 5 AddProccess Clip Sampling Labeling S Section A txt Section A txt c x Anon Student Id tu_043c2ec6c6390dd0ac55 19 190a57c88c Stu_043c2ec6c6390dd0ac5519190a57c88c 1105 wi tu 043c2ec c6 390ddoac5519190a57c88c 105_w02 swf Stu_043c2ec6c6390dd0ac55 19190a57 B8c a w02 swf Stu 043c2ec6c6390dd0ac5519190a57 88c tu_043c2ec6c6390dd0ac5519190a57c88c J05 1 w03 swf w03 swf tu_043c2ec6c6390dd0ac5519190a57c88c 05_w04 swf tu_043c2ec6c6390dd0ac5519190a57c88c 105 vv04 syvf tu_043c2ec6c6390dd0ac5519190a57c88c 1105 w auf i tu_043c2ec6c6390dd0ac5519190a57c 88c 05_w05 swf z tu _043c2ec6c6390dd0ac5519190a57c88c 05_w05 swf tu _043c2ec6c6390dd0ac5519190a57c88c 105_w06 swf tu _043c2ec6c6390dd0ac5519190a57c88c JOS w auf tu 043c2ec6c6 390ddoac5519190a57c88c 05_w07 swf tu 043c2ec6c6 390ddoac5519190a57c88c JOS YV07 svvf 05_w07 swf 05_w08 swf Stu_043c2ec6c6390dd0ac55 19 190a57c88c tu_043c2ec6c6390dd0ac55 19190a57
10. c88c tu _043c2ec6c6390dd0ac5519190a57c88c 105_w08 swf tu_043c2ec6c6390dd0ac5519190a57c88c 05_w09 swf 120 19 21 Row Count S202 S CH er Tue Feb Tue Feb Tue Feb Tue Feb Tue Feb 21 21 21 1 21 OB 44 228 OB 44 28 OB 442278 O8 442 29 O8 44 29 KC Unique KCe96 KCSI14 KIC 1592 KC A3 KIC 1422 KCI1415 KOS KOS KIC 1410 EC 15447 KC 750 KOSOS KCOSS KCiso KOTA KIC 114435 cine OS OC Ch OS OO GHT Os OO Ch OS OO GHT Os OO SHT QE OO 2012 2012 2012 2012 2012 Figure 27 Clipping run feedback KC Catego TEHE i Process Process Process Process Process Process Figure 26 Clipping display he E et Default Default Default Default Default 21 ha SO done Hew Time atarted Hew Time done EK e Clip Proces a started Size Clip Process done Pr done 2005 10 15 7 Figure 28 Distil features Ateneo Laboratory for the Learning Sciences F206 AdMU Page 23 Educational Data Mining Workbench User Manual V3 00 5 Labelling Loads the Labelling Form and asks the user to specify Labelling parameters other additional information and the actual labels The user may use a Template label or not The Use Template checkbox is selected by default S Labeling Set Up Select column name School Note Click Add Label or press Enter to add label name Run StudentNo SeiNo ProbNo AbsProbN
11. ction E txt Row Count 5 e Mon Feb 20 11 35 56 GMT 08 00 2012 INFO Imported C VUsers Paul Documents DataShop IMon Feb 20 13 55 46 GMT 08 00 2012 INFO Clipped DataShop clipped EDM Workbench git 20120214 Figure 20 Submitting the clips 3 Sampling To start sampling the dataset click Sampling Button located either in the Function menu Figure 5 2 or Toolbar Figure 6 Sampling functionalities involve creating subsets from the dataset using automatic select and grouping options A user may take samples or a subset from the loaded dataset and save as a new dataset Sampling can be stratified or random 3 1Random Sampling In the random sampling method the system selects samples randomly To execute this in the Sampling interface select the Sampling Method dropdown menu and click on Normal and set the number of samples in the Samples textbox Ateneo Laboratory for the Learning Sciences F206 AdMU Page 18 Educational Data Mining Workbench User Manual V3 00 Sampling Sampling Method Set Samples Size Maximum Sample Size 16010 100 Figure 21 Sampling Method prompt Note The size inputted in the textbox should not exceed in the indicated maximum sample size 3 2 Stratified Sampling For stratified sampling users have to specify the features that define the strata from which the data will be selected Click on the Sampling Method dropdown menu and select
12. ect Clip Type Size Clip Size Complete dips only 2 Figure 15 EDM Workbench Clipping Box 2 1 1 Custom Sort Button This allows the user to set how the logs are formatted by sorting them Add Level Button adds another row of level sorting properties and Delete Level deletes the selected Row Clicking the Submit button will implement the selected formatting properties Ateneo Laboratory for the Learning Sciences F206 AdMU Page 15 Educational Data Mining Workbench User Manual V3 00 Custom Sort Column Order By Logs of Studens in Section A E Ascending Then by Ar 1 Student J o E W Descending Student Response Subtype utor Response Subtype Figure 16 EDM Custom Sort 2 2 Time as Clip Type By choosing Time as the Clip Type the user will specify a time period per clip e g 1 clip 5 minute interval Column name must be specified in order for clipping to work When done hit the submit button and the clips will be displayed on the new data grid Double click the data grid to view the logs in the selected clip Clipping Box Select Column Name For Clip Grouping 0 Select Clip Type Time Select Time Column Time Interval d Figure 17 Time as Clip Type Ateneo Laboratory for the Learning Sciences F206 AdMU Paqe 16 Educational Data Mining Workbench User Manual V3 00 2 3 Cancel Button This cancels clipping 2 4 Save Button The Save button saves the set properties in the Clipping
13. f logs Model A detector of meta cognitive and motivational behaviour Row A set of attributes in the dataset Usually refers to 1 log Interface Refers to the system graphical user interface Text Mining support TagHelper Machine Learning support Sec RapidMiner ondary data analyses l Sequential Analyzer Statistical Packages SAS SPSS R Log data sources Machine Leaming Data Export i Detector Import PSLC DataShop Sen EDM Workbench an Data Anvers Mult Liser Daplay and Detillation Collaboraive Funcbornaly Functionality Funchonality Defining melacag Sampling Techniques inter rater Collaborative cat D stili Reliabitty Analyst eg Cateqornes E Mr ehabikty Anatysts ng Custommed Log features for Log Collaborative def System Acton Viewer CLAN Action Viewer and al tabeling database Automatically set up machine karin manual pe redlabeling when Auio configuration of metacoj calegones ML software for noteri appropriate Cross veldabon Figure 1 EDM Workbench this is Learmng Systems CTAT buillt tutors Ah Cognitive Tutor l i Si if Collaborative Leaming System Other leaming systems eng educational games Ateneo Laboratory for the Learning Sciences F206 AdMU Page 6 Educational Data Mining Workbench User Manual V3 00 d Overall Use Cases Stratified i vextendsn Sample specified number of Kap Load Datashop
14. f rows therein the current active tab E Mon Feb 20 09 46 48 GMT 08 00 2012 INFO Imported C Usera Paul Documents DataShop Figure 14 Status Bar with timestamp and directory The Status bar displayed the information of a file imported together with the location C User Paul Documents Datashop and the current time Monday February 20 9 46 AM and 48 seconds 2 Clipping To clip the dataset click Clip Button HE located either in the Function menu or Toolbar Figure 6 When clicked the system will display a form with the column names the basis for grouping e g group data with the same Logs of Student in Section A E with the same Anon Student Id and with the same Time and so on Clips can be divided by Size or Time Ateneo Laboratory for the Learning Sciences F206 AdMU Page 14 Educational Data Mining Workbench User Manual V3 00 2 1 Size as Clip Type By choosing Size as the Clip Type the user will need to specify the clip size maximum number of logs per clip The system will then provide custom sorting options for the selected logs When the complete clips only checkbox is checked the system will display only clips where the number of logs is equal to the inputted clip size Clipping Box Select Column Name For Clip Grouping 7 Logs of Studens in Section A E Anon Student Id Session Id Student Response Type Student Response Subtype utor Response Type utor Response Subtype Sel
15. he final output from the active tab in the DataGrid into cvs file formats or other specified file formats 3 5 Clip Button groups or compiles logs from a given batch with user selected parameters EN Sampling 5 6 Sampling Button creates subsets from the dataset using automatic select and grouping options A oF Labeling 3 7 Labelling Button allows the user to label each clip Ateneo Laboratory for the Learning Sciences F206 AdMU Page 10 Educational Data Mining VVorkbench User Manual V3 00 fx al 3 8 Add Process allows the user to add multiple processes 4 DataGrid Row Count Figure 7 Data Grid The Data Grid displays the logs that are active and are to be processed The downwards arrow button hides the data grid Row Count 39458 Row Count controls the amount of rows shown in the active tab a Status Bar Row Count 48443 16 37 31 E II w iG I J ete input columns Thu Nov 17 16 37 40 GMI Q8 00 2011 INFO lled All Data export txt Thu Nov 17 16 38 02 GMI 08 00 2011 INFO Clipped All Data 1 export txt clipped Thu Nov 17 16 38 31 GMI 08 00 2011 WARNING Feature Incomplete input columna features FirstAttempt Thu Nov 17 16 38 32 GMI 08 00 2011 WARNING Feature Incomplete input columns features pKnowDirect Thu Nov 17 16 38 40 GMT 08 00 2011 INFO Distilled All Data export txt EDM Workbench git 20110916 Figure 8 Status Bar The Status Bar displays feedback infor
16. ltext My Documents Et Class Diagram Getting Started With the NetE A B ClosableTabbedPane Ti How to how Table layout A TS Ay eclipse jee indigo win32 KO Import Activity Diagram Computer KA EDM Export Activity Diagram 2 Multi column combobox in L Computer i mm nm d File name DataShop d File name Section A txt Network Files of type all Files Network Files of type All Files Cancel Case 1 Importing log file After locating and choosing the log file it will be displayed to the DataGrid Figure 7 Case 2 Importing batches of log files After locating and choosing the batch of log files another dialog box will appear asking for a label describing the log files imported e g Logs of Students in Section A to E Figure 10 Click Submit Button to display the logs in the Data Grid Ateneo Laboratory for the Learning Sciences F206 AdMU Page 12 Educational Data Mining Workbench User Manual V3 00 Label Column Label 1 Example Section B txt Section A txt Section D txt Section C txt Logs of Students in Section A Figure 10 Label Box The system will take some time to display the dataset depending on the log file size The system display should be similar to that of Figure u All actions buttons save for the Labelling button should be enabled at this point EDMWorkbench File Functions Help i Load Save Import Export lib N
17. mation such as status error messages time elapsed and others Ateneo Laboratory for the Learning Sciences F206 AdMU Page 11 Educational Data Mining Workbench User Manual V3 00 II How to use the System 1 Import Le Import log file by clicking Import Button 9 located either in File menu Figure 5 or Toolbar Figure 6 The system will then pop up a dialog box asking what type of logs you want to import CSV or Datashop file Figure 9 Click the Select Button after selecting the type of Log Log Type Chooser Select Log Type Figure 9 Log Type Another dialog box will appear asking for the location of the log file User may import a log file or batches of log files E Import Data Log File Se U Import Data Log File i i Look in EDMFILES X st epa Lookin DataShop v cia i Ei EDM Import Activity Diagram I si A EDM Design ii EDM Import and Export Class zy cj Section B RE Me di Getting Started With the NetBeans IDE Tutorial files EDM Questions Recent Items Section C Netbeans Tutorial E EDM Sampling Activity Diagri Section D A sample Bi edm cj Section E di Scatterplot KO EDM Design Distillation Act ki Valenzuela City Warehouse System Documentation Ti EDM_SRS_v1 00 Ba 7GB EFOGA_20110203212610615 is bluejsetup 304 EFOGA 20110203214047551 ss CheckBoxDemo Kl Export Activity Diagram MyDocuments Class Diagram Logimport Zi ful
18. o Date TimesStar ted Difficulty StepNo Use Template Duration CurrentTime Action Error Etape Expression Etat Cursor Selection School Equivalence Resolution Figure 29 Labelling window 5 1 Labelling without a Template User may not use a template by unchecking the Use Template checkbox The system will require the user for input in the Labels and Required columns only A Set Up Labelling parameters Labeling Set Up Select column name School Note Click Add Label or press Enter to add label name StudentNo Remove Label Use Template Figure 30 Labelling window w o template Ateneo Laboratory for the Learning Sciences F206 AdMU Educational Data Mining Workbench User Manual V3 00 B Labelling the dataset 3 4915 Logs of Student in Section AE Sample Name Anon Student Id Session Id Section A txt All Data Stu_043c2ec6c6390dd0ac5519190a5 7c88c PCT50877 2005 10 15 Section A Gei ai Data Stu_043c2ec6e6390dd0ac551 9190a57c88c PCT50877 12005 10 15 Confused D Not Confused e 2 1 Figure 31 Labelling wio template 5 2 Using a Template Using template allows the user to load or manually set a desired template by selecting the required columns typing in labels and constructing sentences A Set up Labelling parameters 1 Add Button In constructing sentences that describe a clip users can manually ty
19. pe in the features by enclosing them in brackets Jl Users can also select a feature from the dropdown list and click the Add button to add it to the sentence Labeling Set Up Select column name School Note Click Add Label or press Enter to add label name Confused Run 1 Not Confused Sujdesntiio SetNo Pron Add Label Remove Label V Use Template Duration A student from School with a student ID of StudentNo given a Expression problem ProbNoj and able to finish it in Duration secs Etat Cursor Selection Equivalence Resolution Figure 32 Parameter addition Ateneo Laboratory for the Learning Sciences F206 AdMU Page 25 Educational Data Mining Workbench User Manual V3 00 Note The system will automatically select the parameter in the textbox to the Select Column Name list 2 Save Template The system allows the user to save the set Labelling properties The system will then pop up a dialogue and ask for the desired name for the template and then save the file as a Labelling xml file Q Input File Name SK My Sample Labeling Template Figure 33 File Name input window 3 Load Template The user may select a template from the list of labelling templates displayed by the system The system will then load the properties from the selected template to the labelling form zi Load Box Figure 34 Label template loading B Labelling the da
20. rkbench User Manual V3 00 EDMWorkbench 1 Function Help ma Load Ctri L P Sa VE Ctri Import Ctrl T ED Export Ctri E J Exit trQ The File Menu is composed of 5 actions Load Save Import Export and Exit that handle the files and logs to be displayed and or saved in the Data Grid File Functions Help Clipping Ctrl Shift C Sampling Ctri Shift 5 E Labeling Ctri Shift L faa E ER 4 Add Process Ctrl Shift A The Function Menu consists of 4 log processing actions that will either be enabled or disabled depending on the state of the system 4 EDMWorkbench S File Function Help E About f e Load Save Import Export Dis The Help Menu contains the About action that displays the details of the system e g current version system description etc Ateneo Laboratory for the Learning Sciences F206 AdMU Page 9 Educational Data Mining VVorkbench User Manual V3 00 3 Tool Bar NE e l sh B on S Load Save Import Export Add Proccess Clip Sampling Labeling Figure 6 Toolbar TheTool bar is composed of action buttons that are also found in the menu barforease ofuse Es 3 1 Load Button loads logs vyhich are in a zip format By savel 3 2 Save Button saves the logs from the active tab in the DataGrid into EDM format J A GE 3 8 Add Features Button allows the user to add new Features for distilation of logs sp ae 3 4 Export Button exports t
21. ta from various sources for developing meta cognitive and behavioural models The concept diagram in Figure 1 illustrates the system functionalities and entities interacting with it The EDM VVorkbenchs functions allow users to Define and modify behaviour categories of interest Label previously collected educational log data with the categories of interest considerably faster than current methods s Collaborate with others in labelling data by providing ways to communicate and document labelling guidelines and standards Validate inter rater reliability between multiple labellers of the same educational log data corpus Analyze textual data in collaborative learning situations by integrating a text categorization tool Automatically distil additional information from log files for use in machine learning Export student behaviour data to tools which enable sophisticated secondary analysis c Definition of Terms Batch A group of log files The criteria for grouping are determined by the user Examples of the criteria for grouping include source and timing Clip A subset of logs from a given batch Column A single attribute within the dataset Dataset The data from the imported files EDM Educational Data Mining Log A record of a single action Ateneo Laboratory for the Learning Sciences F206 AdMU Page 5 Educational Data Mining Workbench User Manual V3 00 Log File A file that contains a collection o
22. taset Ateneo Laboratory for the Learning Sciences F206 AdMU Page 26 Educational Data Mining Workbench User Manual V3 00 1 4950 A student from Section A txt with an ID of Stu_O043ctec6c6390dd0acs519190a5Tc c was give n a problem 105 voOl svij and was able to finish it in 2005 10 15 02 08 56 0 seconds Logs of Students in Section AE Anon Student Id Time Problem Name Section A txt stu_043c2ec6c6390dd0ac9519190a5 c88c 2005 10 15 02 08 56 0 109 vv01 5 7 E WEEN 1 2 Figure 35 Dataset labelling window 6 Save Save Saves the dataset in the current tab by clicking the Save button located either in File menu Figure 5 1 or Toolbar Figure 6 The system will ask for the directory and then save it in zip format Note Saving files will take time depending on the size of the dataset and speed of the computer 7 Load EDM File Load Loads EDM files by clicking the load button located either in the File menu Figure 5 1 or Toolbar Figure 6 Error dialogues will be displayed if any error is found with the specified directory or file Note The action button will be enabled depending on the file loaded Ateneo Laboratory for the Learning Sciences F206 AdMU Page 27 Educational Data Mining Workbench User Manual V3 00 Load EDM File Look in F Saved EDM Files F ee E EDM Save Sample p r Ee Recent Items F My Documents L Computer di File name EDM Save
23. xt 14 All Data Stu_043c2ec6c6390dd0ac5519190a57c88c PCT50877 2005 10 15 02 10 56 0 UTC 1 ATTEMPT RESULT Secion Abt Ip Wi Fu 04Scdec co3 0ddOacSSi9i00e57cR6c PCTSOG77 2005 10 15 02 11 08 0 UTC pO mer RS ection A txt JE _ allData Stu_043c2ec6c6390dd0ac5519190a57c88c P 2005 10 15 02 11 05 0 2 RESULT Section A txt 17 All Data Stu 043c2ec6c6390ddoac551919057c88c PP 2005 10 15 02 11 06 0 UTC 1 ATTEMPT RESULT Section Abt 18 Me Stu O4ScecbcS360ddDacSS19100857 88 PCTSOG77 2005 10 18 02 18 120 E liter P Secion Abt ho JAdata ice tegen Fe 2005 10 150211 130 UTC Ji ATTENPT RESULT ection A txt 20 All Data Stu_043c2ec6c6390dd0ac5519190a57c88c PCTSO877 2005 10 15 02 11 21 0 UTC 8 ATTEMPT SULT ection A txt 21 All Data Stu_043c2ec6c6390dd0ac5519190a57c88c PCI50877 2005 10 15 02 11 22 0 UTC 1 ATTEMPT SULT 7 d m Mon Feb 20 11 35 56 GMT 08 00 2012 INFO Imported C Users Paul Documents DataShop EDM Workbench git 20120214 Figure 11 Data Set in EDM UI Ateneo Laboratory for the Learning Sciences F206 AdMU Page 13 Educational Data Mining Workbench User Manual V3 00 Figure 12 EDM VVorkbench Data Grid Tab The DataShop string beside the close button is the folder name that the user specified The label selected in the label selection dialogue Fig 10 can be seen under the tab name Row Count 39468 Figure 133 EDM Workbench Row Count Row Count shows that there are 39468 records number o
Download Pdf Manuals
Related Search