Programme

Data in Public: Engaging the Public for Better Innovation
Dr Mhairi Aitken, Ethics Fellow, The Alan Turing Institute

While data-intensive innovation aims to bring benefits across diverse fields including healthcare, housing and urban planning, these benefits are often not fully realised because projects are designed for the public, not with or by the public.

Drawing on more than fifteen years’ experience of public engagement with science and technology, Dr Mhairi Aitken will explore the importance of public involvement to inform responsible and beneficial data-intensive innovation.  Through examples of previous public engagement, she will demonstrate that involving the public is key to increasing the value, benefits, acceptability and overall success of projects. The talk will map out paths forward to expand the role of the public in shaping the future of data-intensive innovation, moving from innovating for the public, to innovating with the public.

How Might We … turbocharge the impact of the social sector by leveraging data, technology and human connection? 
Hannah Underwood, How Might We

As a region we already have most of the ingredients needed to tackle our wicked social and economic problems. But we need to create new recipes. 

Join Hannah on a journey exploring interconnected, cross-sector ideas emerging from that wonderful place where data, technology and human connection overlap. Simple innovative ideas we can evolve and deliver together to tackle social injustice and drive inclusive growth for the North East and beyond.

AI Bluffers and Experts (and how to live happily together)
David Baillie, Founder, Seascape Systems

Intelligent and Sustainable Heritage Management
Roxana Montazerian, CEO and Founder, ROXID

Roxana's presentation will cover how ROXID revolutionises heritage management through innovative data management techniques. By leveraging cutting-edge technology and advanced methodologies, ROXID has transformed the way we preserve and maintain our cultural heritage. She will delve into the specifics of these innovations and demonstrate their impact on the field.

Furthermore, Roxana will explore how the structured approach of 'Project Scoping' has allowed us to clearly define our objectives and identify key areas for improvement. The 'Arrow Program,' backed by the National Innovation Centre for Data, has provided essential resources and expertise, enabling us to tackle complex data science problems effectively. This collaborative effort has not only enhanced our problem-solving capabilities but also paved the way for future advancements in heritage data management.

Debt Data Discovery
Nick Lamb, Service Design Lead, Gateshead Council

Join to see how Gateshead Council are using data science to support operational delivery and strategic decision making. Using probabilistic record linkage (entity resolution) to create an overview of an individual’s debt.

Nick will outline how the National Innovation Centre for Data has supported the council to apply new approaches and techniques in how it uses it data, and how this experience has shaped the council’s data transformation.

Innovating the Home Buying and Selling Process
Simon Priestley, Partner, Home Sale Pack

Designed to revolutionise the home buying and selling process by providing as much information as possible prior to the sale of the property, Home Sale Pack is an upfront information provider,  providing all the information needed to sell a property at the beginning, including title documents, property forms, supporting documents, searches, EPC and a property health check.  

In collaboration with the National Innovation Centre for Data, Home Sale Pack started exploring the usage of Generative Artificial Intelligence (GenAI) models to build prototypes of question-answering systems capable of extracting, verifying and generating novel responses based on user queries related to the aforementioned legal documents. By leveraging GenAI-based solutions, Home Sale Pack is expected to considerably speed up the conveyancing process.

Home Sale Pack has made an impressive start since launching on February 5, it already has more than 200 registered estate agents nationally and is actively preparing packs for sellers.  Home Sale Pack vision is to be the market leaders in providing upfront information which is backed be innovative technology that speeds up the home buying and selling process.

Teacher Coach - an AI-supported coaching solution for beginning teachers 
Liz Maxwell,  Director of Partnerships, Teacher Success Platform

Since the launch of Chat-GPT in November 2022, Artificial Intelligence (AI) has quickly become a, if not the, hot topic in education and other fields. Much of the commentary, however, is focused on the potential threats posed by AI, with less attention focused on its positive, transformative possibilities. 

We share the aims and development of Teacher Coach, an AI supported coaching solution for trainees and beginning teachers. This technology has exciting possibilities, especially for educators in rural and remote communities where access to professional support can be limited. 

Using AI to supercharge complex data extraction
Dr Paul Hands, Chief Technical Officer, Parallel-51

There is a lot of value held in legal contracts between financial firms, and a lot of legal contracts between these firms can be decades old, and still held in the paper they are written on. A huge hurdle in the legal world is digitising these documents as quickly and efficiently as possible. It takes many hours of time (and hence has a large price tag) to convert from a paper based system to a more digital one.

Parallel 51 has embraced the cutting edge capacity of AI to help streamline and optimise this process, using the newest methods and techniques to let machine learning models do the heavy lifting. In collaboration with NICD we had a goal to improve our model to extract from multiple page length documents data that was prevalent to legal companies.

Exploring newest techniques such as retrieval augmented generation we were able to refine and improve our system, which we will talk about more in depth today. 

Shop by Shape - Dress with Confidence

Janine Craig, Shop by Shape

Shop by Shape's mission is to revolutionise how women shop by providing personalised style recommendations based on body shape. Nowadays, women are faced with challenges in finding clothes that fit and flatter their body shapes.

This talk will explore the role of NICD in providing technical support and guidance from the concept to the prototype of Shop to Shape, including how our partnership with NICD accelerated our prototype development.

We will conclude by outlining our vision for the future of personalised shopping and the broader fashion industry, offering our thoughts on the importance of innovation.

Data science to complement highly skilled accrued technical experience

Simon Welsh and Jennifer Longyear, Akzo Nobel

Analytical chemistry is a foundational tool that International Paint uses to understand various aspects of coating formulations.  Specialist chemists in our Analytical Department run material samples in the lab, carry out highly-skilled analyses of technical results and provide a customised interpretive report tailored to each job request.  Our Analytical team draw on decades of experience and can link current problems to their personal back-catalogues, resulting in both hugely valuable insights AND insight bottlenecks.

We have been working with NICD to explore how data science approaches applied to our raw NMR proton spectroscopy archives might complement the experiential analyses of our team.  Interpreting proton spectra is a highly specialised skillset, as the spectra of chemical mixtures are characterised by patterns which are modulated by sample chemical structures in complex ways not yet describable with explicit rules.  Compounding the challenge, our raw datasets are effectively unlabelled.  

In this talk we will share how upskilling with NICD has helped us move towards leveraging our rich, but human-centric data archives.  

Trustworthy Machine Learning for Data-driven Decision Making 
Dr Matt Edwards, Senior Data Scientist. National Innovation Centre for Data

Deploying a machine learning model into production often requires trust from stakeholders. This trust can come from developing a trustworthy model that aligns with your goals, can achieve and continue to achieve your goals and can explain its own behaviour. In other words, trustworthy models should be aligned, competent, reliable and open. 

In this masterclass, we will explore these four key dimensions of trustworthiness in the context of data-driven decision making and discuss practical strategies to help utilise them. Topics will include accuracy, fairness, privacy, robustness and explainability. 

Searching for Meaning with LangChain: Intro to Semantic Search 
Dr Fergus McClean, Data Scientist. National Innovation Centre for Data

Recently, semantic search has become very popular as organisations seek to use LLMs to answer questions about their data through Retrieval Augmented Generation (RAG). In this interactive workshop, we will explore what semantic search is and how you can use it to find the information you’re looking for in your own text data.  

Join us to discover some practical tips and tricks and get the opportunity to experiment with building your own miniature search engine! 

AI in the Wild
Dr Peter Michalák, Senior Data Scientist, National Innovation Centre for Data

Discover the latest trends in applied AI and learn how it can benefit your organisation. This masterclass offers an overview of AI use cases with real-world applications, including LLM for information extraction, object detection, classification, natural language processing, and more. Additionally, you will learn about NICD’s paid and fully funded service offerings.

Prompt Injection - The No.1 AI Security Risk
Doro Hinrichs, Associate Developer, and Kira Clark, Test Engineer, Scott Logic

With the tremendous opportunities that AI offers come significant security risks that you need to be aware of and be able to manage. In this interactive workshop, you will learn about the challenges of Prompt Injection Attacks and try out hacking an AI chatbot yourself.

From Noise to Nuance: Exploring Image Generative AI Models 
Dr Antonia Kontaratou, Data Scientist, National Innovation Centre for Data

If the world of image generation seems blurry, join our interactive workshop to uncover how AI models work their magic! Explore the inner workings of diffusion models as we journey through the process of transforming random noise into diverse images. Through a blend of theory and hands-on practice, you will gain insight into this process where no text prompts are used!  

Large Language Models (LLMs) for Natural Language Processing (NLP) Tasks
Louise Braithwaite MSc, Data Scientist, National Innovation Centre for Data

Join us for our Large Language Models (LLMs) workshop focused on Natural Language Processing (NLP) tasks. Through a blend of theory and hands-on practice, participants will delve into the core concepts of prompt engineering, gaining practical experience in applying these techniques to real-world scenarios using the LangChain framework and Python.  

If you would like to learn more about LLMs or sharpen your NLP skills, this session is for you! 

MLOps: Deploying Models into Production 
Myles Mitchell, Data Scientist, Jumping Rivers

Join us for an insightful workshop on MLOps, where we'll explore the crucial steps of deploying machine learning models into production. This workshop will cover practical techniques and best practices for ensuring your models are not only functional but also robust and scalable. We'll start with a brief introduction to MLOps and its importance in the machine learning lifecycle.  

By the end of this workshop, you'll have a solid understanding of the end-to-end deployment process, enabling you to take your models from development to production efficiently. This workshop is ideal for data scientists, ML engineers, and anyone interested in learning about production-ready model deployment. 

Applications of Machine Learning in Wildlife Conservation
Dr Georgia Atkinson, Data Scientist, National Innovation Centre for Data

Discover the transformative role of machine learning in wildlife conservation in this masterclass designed for enthusiasts and professionals alike. We will explore how advanced technologies are being applied to monitor and protect wildlife, with a focus on audio analysis. 

Experience a live demonstration of using low-cost hardware to monitor wildlife and discover practical ways to implement these tools in your own backyard!