300 likes | 313 Views
SSW - Spatial Statistics on Web. Tuuli Pihlajamaa, Marja Tammilehto-Luode September 10th 2015 Nordic Forum for Geography and Statistics. Contents. Introduction Oskari p latform SSW in practise : Development by Scrum methodology Testing procedure User stories in the project
E N D
SSW - SpatialStatistics on Web Tuuli Pihlajamaa, Marja Tammilehto-LuodeSeptember 10th 2015 Nordic Forum for Geography and Statistics
Contents • Introduction • Oskari platform • SSW in practise: • DevelopmentbyScrummethodology • Testingprocedure • User stories in theproject • Examples of theanalysismethods to bedeveloped • Oskari Pilot in Statistics Finland • Concludingremarks Etunimi Sukunimi
Introduction • Objectives of theproject • To providespatialstatistics on theweb • To improvethenationalgeoportalfromthepoint of view of statistics • To gainexperienceabout an open sourcewebapplication • To promotecooperationbetweenthe NSI and theNMA • Focus on usability of grid-base data – use of Inspire data and concept • Eurostatgrant 2014-2015 • Inspiringcooperationwiththe National LandSurvey • SF projectleader, 4 SF projectmembers, 4 NLS projectmembers + scrum team • SupportedbybothDGs Etunimi Sukunimi
Oskari platform Open source applications • Created and maintained by the National Land Survey • Built on standard Open Source components (OpenLayers, Geotools, Geoserver…) • Promotes extension of functionalities in a coordinated manner – integration of applications • Open Source - MIT/EUPL dual licensing • Guidelines, source code and all the content on the Developer Web Site and in GitHub • Oskari network (more than 30 user organisations), management group, integrator (e.g. responsible for maintaining the Developer Web Site) • The National Geoportal is an implementation of using the Oskari platform Etunimi Sukunimi
DevelopmentbyScrummethodology • The Oskari software development in National LandSurveyof Finland (later NLS) is donebased on agile software developmentmethodologyScrum • In thisprojectonesprintlastedusuallytwoweeks • StatisticsFinland’s(later SF) part in theprojectwas to writeuserstoriesfor NLS and testthatthestorieswereimplementedright • NLS dividedtheuserstories into smallerstoriesto beused for development EtunimiSukunimi
Testingprocedure • NLS provided a demo environment, that SF coulduse to testthetools and functionsthatweredeveloped, beforelaunchingthem in Paikkatietoikkuna • SF gavefeedback and reportedifthetoolsneededimprovements • Mostlytestingwasdonebythemembers in theproject team, butbesidesthatweorganizedalsotwoworkshops, wheretheapplicationwastestedbyusers outside theprojectgroup Etunimi Sukunimi
User stories in theproject • 13 storieswere to befinished • Functionsincluded in thestories: • Key ratioscomputationbased on differentareaselections (freehand, areacode, sectors and buffers) • Handlingthepopulationgrid data in calculations, dealing data withprotected/no-data values • Differencecomputation, calculatingdifference in populationgrid data valuesbetweendifferentyears • Filtering databased on thevalues in thedata orbased on theresults of keyratioscomputation • Spatial join, joining data based on location • Heatmapanalysis EtunimiSukunimi
Previousanalysismethods to bedeveloped • Buffer • creating buffers around features (buffers for multiple points, buffers for lines and polygons) for buffer analysis • Key ratios computation • calculating median was added • handling the grid data with no-data values/protected values • filtering data by using results from key ratios computation Etunimi Sukunimi
Buffer- creating buffers around features (buffers for multiple points, buffers for lines and polygons) for buffer analysis Etunimi Sukunimi
Key ratioscomputation- handling the grid data with no-data values/protected values Etunimi Sukunimi
Key ratioscomputation– filtering data byusingresultsfromkey ratios computation Etunimi Sukunimi
New analysismethods to becreated • Buffers and sectors (Multiple Buffer) • creating buffers and sectors for analysis • Difference computation • calculatingdifference in populationgrid data valuesbetweendifferentyears • Spatial join • enriching data based on spatial location • using spatial join in key ratios computation Etunimi Sukunimi
Buffers and sectors (Multiple Buffer) - creating buffers and sectors for analysis Creatingbuffers and sectors Etunimi Sukunimi
Difference computation - calculatingdifference in populationgrid data valuesbetweendifferentyears Etunimi Sukunimi
Spatial join – enriching data based on spatial location • Example: Givingpostalcode for educationalinstitutions Etunimi Sukunimi
Spatial join - using spatial join in key ratios computation, case buffers and sectors Using populationgrid data for theaggregation Etunimi Sukunimi
Spatial join - using spatial join in key ratios computation, case buffers and sectors Key ratios of populationbysectors Etunimi Sukunimi
Heatmap(Kerneldensity) • ChooseWMS-layerfotheanalysis • Choosetheradius of kernels, pixels per celland weightproperty Etunimi Sukunimi
Heatmap(Kerneldensity) • Results: Etunimi Sukunimi
Oskari Pilot in Statistics Finland • Aim of thepilot is to haveexperience of thetechnicalimplementationof Oskari and it’susability in SF. ResultswillsupporttheGIS technologyreview. • To reachtheaim, wewill: • Buildinternalcatalogserviceof spatialstatistics data of StatisticsFinland • Documenttheprocess, howto build Oskari services, based on practicalexperience Etunimi Sukunimi
Approach to the Oskari Pilot Firststage: Technical understanding of Oskari platform • Building a testenvironment • How to implementOskari functionsin thePilotservice • Questionsregardingupdateand theadministrationof the intranet service • Definingthedemands of openinga publicservice(maybe in thefuture) Second stage: Building a pilotservice • Definingtheservice (data and functions) • Implementing data and metadata in theservice • ImplementingOskari functionsto thePilotservice Etunimi Sukunimi
Firststage: Technical understanding - Intranet application in function Etunimi Sukunimi
Second stage: Pilotservice to bebuild • Viewingservicein intranet • Open data from SF interfaceservicesand other data • Readyservicewillsupportstatisticsproduction and increasetheunderstanding of the dataavailable in SF Etunimi Sukunimi
Data in thepilotservice • INSPIRE-data (newest) • Municipality-basedstatisticalunits • Grid net for statistics 1 km x 1km • PopulationDistribution • Production- and Industrial Facilities • EducationalInstitutions • PAAVO – Open data bypostalcodearea • Municipalsub-areas • Prices of dwellings in housing companies – by postal code area Etunimi Sukunimi
Pilotservice– draftfromthe data list and metadata from National Geoportal Etunimi Sukunimi
Pilotservice– dataproducer’stools • Layeradministration • Layerrightsadministration • User administration Etunimi Sukunimi
Currentsituation of Oskari Pilot and results • Technical understandingis stillevolving • Results of thepiloting - bytheend of theyear 2015 • SSW projectends at theend of theyear • Resultsaremeant to supportthe GIS technologyreview, thatwillbedonebytheend of theyear Etunimi Sukunimi
Concludingremarks • Mutual interest – a concretecooperationproject • Learning bydoing – learningfromeachothers • Promotedfurthercooperation • Open source application on the web feasible - Further development is promising (graphs and tables) • Cooperation with other Oskari platform users • Final report due January 2016 (Interim report available) • http://www.paikkatietoikkuna.fi/web/en/map-window • http://www.oskari.org/ Etunimi Sukunimi
Questions? tuuli.pihlajamaa@stat.fi marja.tammilehto-luode@stat.fi