Discovering new and exciting ways of improving data used for organisational decision making

Dr Philip Woodall presented the latest research of the Data Management group at the British Computer Society (BCS) Data Management Specialist Group event in London on 10 April. The event centred around decisions and the impact that data quality can have on decision making. A video of the presentation can be found here. The presentation was also featured in a DataIQ article available here.

Philip described the concept of “data repurposing”, which is when data is used for a completely different task to what it was intended to be used for (note that most data analytics activities are a repurposing of data), and the resulting challenges that can arise. He also described the data tagging research, which can help to label data as accurate or not automatically within an information system.

The DIAL DASHLog project team recently visited the project sponsor YH Global in Shenzhen, China. During the visit, the team presented a specification of a new type of system that can better manage the changing circumstances that are faced by warehousing and logistics organisations. This system operates in different ‘modes’ in order to best cater for each circumstance, and also defines how and when to transition between modes. Over the next year, the DASHLog project will develop and test this system further, and two case studies have been selected that YH Global is keen to optimise.

The DIAL VIPr project has also been engaging closely with its sponsor, Boeing, in order to present the latest findings from the Supply Chain Miner tool, which can automatically build a representation of existing supply chain networks using online, publicly-available data. The core technique being used is Natural Language Processing (NLP) and this is being combined with intelligent searching and filtering to produce the final network. Work is currently underway to exhaustively evaluate the performance of this new and novel approach.

Publications in progress 

(If you would like advanced access to any publications or research results, then please contact Dr Philip Woodall at phil.woodall@eng.cam.ac.uk)

• Woodall et al. Potential Problem Data Tagging: Augmenting information systems with the capability to deal with inaccuracies.
• Wichmann et al. Who’s behind the curtain? Towards automatically generating supply chain maps from natural language text.
• Fischer, Zaki, Woodall. Factors Influencing the Acceptance of Self-Service Data Preparation and Analytics Tools.