Warning:
This wiki has been archived and is now read-only.

Best Practices/TextClustering

From Share-PSI EC Project
Jump to: navigation, search

Clustering Text From Best Practices

Clustering the text from the best practice drafts using n=7 clusters gives the following result:


Top terms per cluster:

Cluster 0 words: already, websites, web, techniques, tool, https, formats, organisation, practice, only,

Cluster 0 titles: Identifying_what_you_already_publish, Discover_published_information_by_site_scraping,

Cluster 1 words: quality, publicly, sector, bodies, published, stakeholders, web, engines, publicly, sector,

Cluster 1 titles: Human_Readability_and_Machine_Processing, Cost_of_Publication, Stakeholders%E2%80%99_Interests_and_Rights, Feedback_to_Improve_Quality, Optimization_for_Search_Engines, Publication_with_Common_Metadata, Catalogs_and_Indexes, Open_Data_quality_assessment,

Cluster 2 words: engage, users, sources, dataset, likely, department, psi, measure, assess, provide,

Cluster 2 titles: Holistic_Metrics, User_engagement_and_collaboration_throughout_the_lifecycle, Organisational-internal_engagement, Encourage_crowdsourcing,

Cluster 3 words: strategy, support, agency, plan, level, =intended, high, =status=, =status=, draft,

Cluster 3 titles: Cross_Agency_Strategy, High_Level_Support,

Cluster 4 words: standard, title, https, form, work, bodies, sets, info, links, provide,

Cluster 4 titles: Publish_spatial_data_on_the_web, Monitoring_and_Benchmarking, Make_the_data_available_in_the_language_people_want_it, Open_Data_2.0_-_Changing_Perspectives,

Cluster 5 words: open, practice, best, dataset, projects, information/data, sector, needs, sharing, national,

Cluster 5 titles: Management_Of_A_Wide_Public_Actors_Network, Making_Research_Results_Open_For_The_Country, Publishing_Statistical_Data_In_Linked_Data_Format, Supervizor_-_An_Indispensable_Open_Government_Application_(Transparency_Of_Public_Spending), Civic_Use_Of_Open_Data, Open_Data_Publication_Plan, A_Federation_Tool_For_Opendata_Portals, Traffic_Light_System_For_Data_Sharing, Open_Data_To_Improve_Sharing_And_Publication_Of_Information_Between_Public_Administrations, Commercial_Considerations_in_Open_Data_Portal_Design, The_Central_Role_of_Location, Free_our_maps,

Cluster 6 words: business, open, model, companies, practice, process, services, best, needs, product,

Cluster 6 titles: Using_Business_Process_Paradigm_For_Open_Data_Lifecycle_Management, Infomediary_Sector_Characteristics, Open_Data_Business_Model_Patterns_and_Open_Data_Business_Value_Disciplines, An_ongoing_open_dialog_in_an_open_data_ecosystem,



Best Practices Cluster Plot


Best Practices Cluster Hierarchy