1 /90
Workshop 1 Big Data for Libraries Kia Siang Hock kia_siang_hock@nlb.gov.sg 2 /90
The Workshop Programme 14:00 Welcome 14:10 About National Library Board, Singapore 14:20 What is Big Data? 14:40 Big Data in Libraries 15:15 Break 15:45 Examples of Big Data Implementations: Recommendations, Text Analytics, Ngram Viewer, Named Entity Extraction, Image Matching 16:45 More Q&A 17:00 End of workshop 3 /90
About the National Library Board, Singapore 4 /90
About National Library Board, Singapore 1 National Library Libraries & 1 National Archives Archives 26 Public Libraries 5 /90
About National Library Board, Singapore Vision Readers for Life, Learning Communities, Knowledgeable Nation Mission We make knowledge come alive, spark imagination and create possibilities. 6 /90
About National Library Board, Singapore Public Library Services The Public Library seeks to be a social learning space that nurtures active readers and knowledge seekers, through the provision of relevant, timely and engaging library services and reading programmes, using physical and digital means. 7 /90
About National Library Board, Singapore National Library Only library in Singapore that collects comprehensively published and distributed Content in the country for preservation and long term access Enable easy access to country’s Shared Memory to build rootedness and national identity Forge International Collaborations and advise on library development 8 /90
About National Library Board, Singapore The National Archives of Singapore (NAS) is the official custodian of Singapore’s collective memory. Ranging from government files, private memoirs, historical maps and photographs to oral history interviews and audio- visual materials, the NAS is responsible for the collection, preservation and management of Singapore's public and private archival records. The Asian Film Archive is founded to preserve the rich film heritage of Singapore and Asian Cinema, to encourage scholarly research on film, and to promote a wider critical appreciation of this art form. 9 /90
A typical day in Singapore libraries… 79,000 people visit libraries 300 new 100,000 loans members join are made the library 27,000 people attend library programs and exhibitions 10 /90
About National Library Board, Singapore 1 National Library Libraries & 1 National Archives Archives 26 Public Libraries Membership More than 2m members Visits More than 27m visits Loans More than 35m loans More than 1m titles Collection More than 8.5m items Digital User Visits: > 11m Online Usage e-Retrievals: > 70m FY2013 figures 11 /90
What is Big Data? 12 /90
What is Big Data? “data of a very large size, typically to the extent that its manipulation and management present “The ability of society to harness significant logistical challenges.” information in novel ways to produce useful insights or goods and services of significant Oxford English Dictionary values” and “… things one can do at a large scale that cannot be done at a smaller one, “an all -encompassing term for any collection of to extract new insights or create new forms data sets so large and complex that it becomes of value.” difficult to process using on-hand data Viktor Mayer-Schonberger & management tools or traditional data processing applications.” Kenneth Cukier Wikipedia “The broad range of new and massive data “datasets whose size is beyond the ability of types that have appeared over the last typical database software tools to capture, decade or so.” store, manage, and analyze.” Tom Davenport McKinsey Source: http://www.forbes.com/sites/gilpress/2014/09/03/12-big-data-definitions-whats-yours/ 13 /90
What is Big Data? Top recurrent themes in the definitions of Big Data by 40 thought leaders Source: http://datascience.berkeley.edu/what-is-big-data/ 14 /90
The Four V’s of Big Data Source: http://www.ibmbigdatahub.com/infographic/four-vs-big-data 15 /90
The Fifth V: Values Big is relative. Five broad ways in which using Big Data can create value ❶ Unlock significant value by making information transparent and usable at much higher frequency ❷ Collects more accurate and detailed performance information ❸ Allows ever-narrower segmentation of customers ❹ Sophisticated analytics can substantially improve decision-making ❺ Improves the development of the next generation of products and services Source: Big data: The next frontier for innovation, competition, and productivity (McKinsey) http://www.mckinsey.com/insights/business_technology/big_data_the_next_frontier_for_innovation 16 /90
IDA Infocomm Technology Roadmap ‘Big Data’ is a key technology theme that will shape the ICT landscape Opportunities Challenges > Analysis of unstructured data > Understand and such as images and audio on top framing Big Data of text data to unearth insights problems from a bigger data pool > Maturity in some of > Insights from the data analytics the underlying analytics outcomes to augment decision algorithms making processes > Shortage of data > Analytics (retrospective to analytics talent predictive) to proactively identify opportunities or tackle problems Source: IDA’s Public Consultation on Infocomm Technology Roadmap 2012, 17 Aug 2012 http://www.ida.gov.sg/Technology/20060417212727.aspx 17 /90
IDA Infocomm Technology Roadmap ‘Big Data’ is a key technology theme that will shape the ICT landscape Radar Technology Stack < Hadoop MapReduce & distributed file system < NoSQL DBMS < Text Analytics < Visualisation-based discovery <03 < In-memory analytics Years < Audio analytics < Predictive analytics < Master data management < SaaS-based business analytics - Complex event processing 03-05 - Data-federation/visualisation - Video analytics Years - Mobile business analytics - Non-volatile memory Source: IDA’s Public Consultation on Infocomm Technology Roadmap 2012, 17 Aug 2012 http://www.ida.gov.sg/Technology/20060417212727.aspx 18 /90
Big Data for Libraries 19 /90
Disclaimers Not a comprehensive study of the use of big data in libraries. A practitioner's high level overview of use of big data in libraries. Do not cover big data issues including data management, privacy and ownership. 20 /90
Big Data Goals Leverages NLB’s unique Actionable Insights data assets Structured & Unstructured Data Better foresights for VCDs Browse future libraries planning Count Books Visits DVDs Productivity gain with Patrons Events better decisions Digitised Blogs Tweets newspapers Loans Digitised books Customer satisfaction E-Books Newspapers Demographics improvements with Locations E-databases better service offerings Facebook pages Better usage of NLB services and resources Unearthing the hidden treasures 21 /90
Big Data for Libraries Library Patron Planning Profiling Collection Business Optimisation Operations Digital Service Library Delivery 22 /90
Big Data for Libraries Library Planning using Geospatial Analytics Where are our users? What do they read? Are our libraries serving the residents in the vicinity? Where shall we target our outreach campaign? What is the impact on the usage of existing libraries when a new library opens? Can our libraries cope with the population growth? 23 /90
Big Data for Libraries Patron Profiling & Footfall Analysis to Optimise Use of Library Space Crowd Density Human Traffic Flow Audience Profiling Source: Video Analytics as a Service http://vaaas.kaisquare.com/ 24 /90
Big Data for Libraries Measuring & Analysing Energy Consumption using Smart Meters 25 /90
Big Data for Libraries Collection Optimisation – Collection Planning Source: http://www.ifla.org/files/assets/hq/news/documents/nlb-collection-management-e- newsletter-april-2013.pdf 26 /90
Big Data for Libraries Collection Optimisation – Collection Planning Initial Planned Final Collection collection Planned Available Budget budget Planned Forecast of Acquisition Collection usage Planning Model Planned Weeding Cost of books Planned Space Shelf space Projected Min/max Loans collection size Source: http://www.ifla.org/files/assets/hq/news/documents/nlb-collection-management-e- newsletter-april-2013.pdf 27 /90
Big Data for Libraries Collection Optimisation – Demand Forecast Source: http://www.ifla.org/files/assets/hq/news/documents/nlb-collection-management-e- newsletter-april-2013.pdf 28 /90
Big Data for Libraries Business Operations (Corporate KPIs, Finance, HR) Source: http://www.ifla.org/files/assets/hq/news/documents/nlb-collection-management-e- newsletter-april-2013.pdf 29 /90
Big Data for Libraries Library Analytics Toolkit The Library Analytics Toolkit is a dashboard that pulls library data together in a way that allows both librarians and library users to identify and respond to trends and changes in collections, usage, and other data Source: https://osc.hul.harvard.edu/liblab/projects/library-analytics-toolkit 30 /90
Big Data for Libraries Integrated & Operational Analytics Patrons can easily access to ~20% of items are popular items borrowed within 3 days of Libraries can reduce resources their return for shelving Just Return Bin Auto-sorter 31 /90
Recommend
More recommend
