There is a large quantity of information communicated through speech audio streams (phone calls, broadcasts, commentary etc.) across many industries. In many instances this information is of potentially high value but is discarded for lack of efficient and accurate information retrieval strategies. Speech recognition technologies partially solve this problem but the output texts are still high volume, noisy and difficult to interpret and summarize.

With the rise of service customisation, data analytics is becoming increasingly important so as to understand the needs of a company's customers. In order for the potential of data analytics to be fully realised, analytics tools must be developed for use by the widest user-base possible. Under the CeADAR Intelligent Analytic Interfaces: Ease of Interaction theme, smart analytics tools are being developed to aid non-analytics specialist users in exploring datasets and performing analytics tasks. The first task selected for focus under this theme is customer segmentation, an especially common data analytics task and one that organisations tend to perform when they first start to use data analytics. Presenting segmentation results to the user in a way that is interpretable is the key to the success of SmartSeg. To understand customer segmentations analytics practitioners compare the distributions of features within a segment to the distributions of the same features in the underlying full population. While major existing analytics tools have enabled widespread adoption of data analytics, they remain mostly targeted at expert analytics users. At CeADAR we are developing smart analytics tools for non-analytics experts - the so called next billion users.

Providing analytics-driven insight does not guarantee that people will take note of it and change their decision-making behaviours. For example, did a customer identified as a churn risk respond to the intervention recommended; or did loan officers in a bank actually base their lending decisions on the output of risk models? Measuring the success of any analytics-driven project, therefore, requires looking beyond the insights that are produced to ways in which these insights are communicated and delivered to ensure that behavioural change takes place. This research will benefit any company that wishes to make sure that the application of analytics results produces the required or expected outcome. For example, did a customer identified by a supermarket as a user of baby products take advantage of a promotion on nappies that was on offer that week? If not, can a message be tailored for that customer to encourage them to take advantage of such a promotion the next time? Using tailored messages, our research is working on encouraging the greatest level of change in a customer/user’s behaviour .

As the voice and sometimes face of a company, Contact Centre Agents must deal with customer inquiries efficiently and professionally. Pressures on agents include throughput-based service level goals as well as the need to ensure a top-quality service and sales experiences to customers at every point. Maintaining this quality and efficiency requires the provision of key customer history data to agents during calls without requiring agents to engage in time consuming searches across interfaces. Unfortunately the provision of such customer and case histories to the agent is far from a trivial task. Customer history summarisation is made difficult not only by enterprise-wide information integration challenges, but also by the computationally demanding task of determining the most relevant information that can be provided to the agent in bite-sized chunks.

Business process modelling is a technique for representing existing or future processes within an organisation, particularly with a view to documenting and improving them. Many business processes manipulate sophisticated data sources to generate a data set that will be subject to subsequent data analytics tasks, and their study falls within the general field of data management. User-friendly editors are available which allow business users to define business processes as a structured collection of tasks towards a specific goal. What they do not provide, however, is the ability to verify and optimise those processes or the data that they produce.

Database queries are often business-critical. But there are few methods for ensuring their correctness. Businesses need new methods and tools that increase confidence in the correctness of both standard and ad-hoc queries.

Searching online holidays websites can return hotels or villas that are very similar in location, function, and quality. Similarity is important but introducing diversity in information retrieval can significantly improve the accuracy, relevance, efficiency and effectiveness of these searches, particularly when queries are vague. Search result diversity is important as well as query similarity. When queries are vague, results should be similar to the query while at the same time being different from one another. Result diversity can provide the user with a representative sample of relevant items, rather than an incorrectly biased set of mutually similar results.

This ARCH report provides an overview of the Connected Health market in the UK – with a particular focus on England as the largest market within the UK region. The report gives an overview of the healthcare financing and payment structures; the potential sales channels and routes to market for Connected Health providers. The report provides information and suggested strategies for selling into statutory commissioners such as NHS England and Clinical Commissioning Groups (CCGs). It gives an overview of the Connected Health market landscape – including competition, size and value, reimbursement landscape and overall SWOT analysis. This report also looks at one condition, Dementia, in more detail. This provides an exemplar of how Connected Health technologies are being used in the care pathway of one of the most costly conditions for the UK healthcare system.

The accumulation of charge in the dielectric of a MEMS (Micro Electro Mechanical Systems) device results in a parasitic source of electrostatic force which, if uncontrolled, can lead to device malfunction and permanent stiction that negatively impacts on reliability. This dielectric charge accounts for a number of reliability issues in electrostatically actuated MEMS and hence is a major barrier to the commercialisation of many such devices; as in the case of RF switches. Researchers from UCD and UPC have developed this novel control method that improves the accuracy and resolution of all devices that use electrostatic actuation.

A system for search and discovery of information in a real time network, comprising: means for gathering data indicative of a message posted in an real time network, the data comprising information identifying a uniform resource locator, URL and textual information associated with the URL