ULfAD – Unsupervised Learning for Anomaly Detection
Project Description MARKET NEED ULfAD explores unsupervised learning techniques for anomaly detection. The idea is to find patterns of interest such as outliers or exceptions that deviate from normal data behaviour. Nowadays the early detection of these unexpected and rare events is an important part of a business because it helps to reduce the downtime of the processes, reduce money lost preventing equipment damage and allowing make corrective decisions more quickly. TECHNOLOGY SOLUTION ULfAD is a web-based tool that offers three state-of-the-art methods for anomaly detection based on unsupervised learning schemes in a user-friendly manner: Three cutting-edge methods: Self-Organising Maps (SOM), Autoencoders
XPlainIT – Explainable AI for Deep Models
Project Description MARKET NEED XPlainIT is about opening the “black box” decision making of machine learning algorithms so that decisions are transparent and understandable. The users of this capability to explain decisions are data-scientists, end-users, company personnel, regulatory authorities, or indeed any stakeholder who has a valid remit to ask questions about the decision making of such systems. The focus of XplainIT is on Deep Learning Models for structured data. TECHNOLOGY SOLUTION XPlainIT is a Web-based tool which offers a series of functionalities for explaining the decision-making process of Deep Learning models applied to structured data in a user-friendly manner,
SimXpert – Business Process Mining with Simulation
Project Description MARKET NEED A common problem faced by companies with complex work processes is how to ensure that the best use and configuration of resources is achieved. In a process-driven environment, each object (e.g. customer, insurance claim, patient) is a participant in a process, competing against other cases for resources resources. Business need to be examine what-if scenarios on replicas of their real-life business scenarios – combining the domains of simulation with process mining – to enable low-risk low-cost process optimisation testing. TECHNOLOGY SOLUTION SimXpert tool generates visualisations of live business processes from logs of actual system events. Visualisations display bottlenecks and process paths. Simulation
ASAP – Applied Time Series Analysis and Prediction
Project Description MARKET NEED Time Series are used in many business areas, such as finance, health, environment and energy. Daily sales, weekly orders, monthly overhead or yearly income are all examples of Time Series information. Formally, a Time Series (TS) is a sequence of values for a variable with associated timestamps. The series represents the variation of this variable over a period of time. Analysing time series information is complex. State-of-the-art tools require expert knowledge to produce reliable, efficient and useful reports. The ASAP demonstrator removes this obstacle by providing an easy to use, intuitive and interactive environment which allows
TagTell: Automatic tag detection & extraction in large scale text corpora
Project Description MARKET NEED Many companies face an increase in data volume of unstructured text documents from customer feedback (surveys, customer support, chatlines, correspondence) or from company-specific documentation. There is a need to quickly understand the content of such data and its main topics. User-tagging of large volume of text data entries is expensive, time-consuming and user-dependent. A solution is to automatically assign tags to such datasets. Tags can be extracted directly from the text data or can represent text content at a semantic level. Tags can have multiple uses: Re-structuring data collections to support classification of data; Information retrieval based on new and more
NPSVu: Enhanced NPS Sentiment Analysis
Project Description MARKET NEED BI decisions often need to consider customer feedback in terms of ratings, comments and opinions in the context of their location and temporal trends. A variety of companies would benefit from a visualisation tool that combines Net Promotor Scores, the opinions expressed in customer feedback, and spatio-temporal data. TECHNOLOGY SOLUTION NPSVu: an easy-to-use web tool enables visual exploration of NPS scores and customer feedback from surveys. Aspect Category Detection used on customer feedback to identify common features discussed by customers. Chart representation of feedback trends for chosen location and time of year. NPSVu enables the visualisation
DataGen: A flexible tool to generate synthetic data
Project Description MARKET NEED A common problem faced by companies working in analytics is the difficulty in creating “random” datasets that resemble actual company data. Companies need purpose-built data for a variety of reasons, including the demonstration and testing of systems and analytical models. Companies are often reluctant to use their own company or client data due to privacy issues, but they need data with the same characteristics as their real data. TECHNOLOGY SOLUTION DataGen allows users to generate synthetic data. DataGen offers two modes of operation: Manual generation: The user specifies the features, rules and inter-feature relationships. Automatic generation: The
CrowdTrack: Spatio-temporal estimates of pedestrian concentrations
MARKET NEED Locations with greater footfall attract more customers and are likely to be more profitable, especially when new locations fill gaps in the existing network, and provide access to key demographics. By applying machine learning techniques to open spatial and temporal datasets, CrowdTrack provides spatiotemporal estimates of pedestrian concentrations. TECHNOLOGY SOLUTION PEDESTRIANS | USER AND CENSUS DATA CrowdTrack allows users to gain insight into: when and where urban crowds tend to form spatial and temporal variations in user datasets, such as differences in store performance or timing of peak sales according to location desirable locations for new stores
DocoPool: Discovering hidden knowledge in text
MARKET NEEDMany businesses store pools of information-rich text in their systems – such as reports, customer reviews, user comments and general word processing documents.Insights can be uncovered by analysing these text sources to discover underlying hidden topics or trends, such as unexpected clusters of words across documents.In reality, however, it can be difficult to manually examine all this of information to discover such hidden trends or topics.TECHNOLOGY SOLUTIONFigure 1. Visualising document groups based on word clusters or topicsDocoPool – a web tool that allows users to explore the content of text documents for hidden knowledge:identifies and visualises word groupings
SmartSearch: Search and retrieval of information using semantic reasoning
MARKET NEED In many companies, the search and retrieval of information matching certain criteria is a key part of the business process. If we take, for example, those companies that work with many different services or products, the task of searching through product or service queries can be a key part of their data analysis or customer care processes. Typically employees will have to query internal knowledge bases or consult with external sources of information to gather sufficient knowledge to solve domain specific tasks. At present, a typical keyword-based search will not retrieve the wealth or depth of information
SmartAd: Measuring advertising campaigns efficacy
MARKET NEED One of the major questions in the advertising space is how to determine which advertising initiatives have been successful and which have not. This age-old question in advertising gave rise to the famous John Wanamaker quote: “Half the money I spend on advertising is wasted; the trouble is, I don’t know which half.” Today, advertising runs through various channels such as television, radio, print, web and social media. Each of these has different characteristics in terms of form, audience and data about customers interaction with the advertisements. We know that channels and campaigns interact: A customer that has
SmartSeg: A customer segmentation tool designed to be accessible to a broad user base
Market Need 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
SmartContact: An analytics-driven solution to optimise the services of Customer Contact Centre Agents
Project Description 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. Technology Solution SmartContact is an analytics-driven solution that automatically provides Customer Contact Centre Agents with a concise review of customer history and other relevant information points
Summit: Summarising processes using structured and unstructured data
Process summarisation techniques generate overviews of the current state of a business process to enable evaluation of progress to date and to aid future planning. An overview of any process essentially reduces to answering the following key questions: What is the actual status of each process? What are the recent changes? Is there a problem looming up? While current issue trackers already provide an impressive amount of structured information on the status of issues, there is a substantial amount of free text provided by users and administrators in these systems that, up to now, has not been widely exploited
OntoCore: Semi-autonomously created ontologies based on Linked Open Data
Unstructured and semi-structured data within an enterprise can be many folds larger than structured data. But this data is often difficult to leverage to new business opportunities due to a lack of domain models for the data. Figure 1. Untamed Assets: Unstructured and Semi-Structured Data Identifying concepts held in existing data and the relationship between these concepts is a crucial step in opening up new product potential from existing data. The OntoCore Project helps to achieve this goal. TECHNOLOGY SOLUTION OntoCore – a tool-kit for the semantic enrichment of unstructured and semi-structured data with relevant concepts and relations extracted from publicly
ModelVis: Regression Model visualisation
MARKET NEED Data analysis is an increasingly important task for industry; one they must typically perform before making a business decision. Linear and logistic regression are commonly used as predictive data analytic models. The key characteristics of a regression model are: the regression weights, also known as regression coefficients, regression estimates or parameter estimates, the sign of each weight, a statistical significance associated with each weight. These characteristics are associated with the model’s features and interaction features (i.e. interaction terms or combined features for the purposes of regression). ModelVis is a regression model visualisation system that provides an intuitive understanding of how these three characteristics