Maynooth University Love Data Week 2021

Monday, February 8, 2021 - 14:00 to Friday, February 12, 2021 - 13:30
Online

Programme

All events are free of charge, however, booking is essential

Monday 8th February

Time: 14:00-15:00

Event: Irish Sea Level Change in the Atlantic Context

Speaker: Maeve Upton, Maynooth University

The island of Ireland, located at the edge of Europe, is in a vital position to study the changing Atlantic Ocean. It is crucial to understand the evolution of sea level through the centuries, as the past is the key to the future. In addition, 40% of the Irish population is located 5km from the coast thus, it is vital to understand what impact a change in sea level may have. As a result, the Marine Institute and the European Regional Development fund have invested €2 million for five years into its’ research.

The A4 project (Aigéin, Aeráid, agus athrú Atlantaigh—Oceans, Climate, and Atlantic Change) aims to examine ocean and climate changes in the Atlantic. The project targets three aspects of the Atlantic; its’ changing ocean dynamics, sea level changes and Irish decadal climate predictions. 

In this talk, Maeve will discuss sea level change along the US Atlantic coast using the statistical approach of Gaussian Process models. These models are built in a Bayesian framework which uses prior information to capture the evolution of sea level change. The proxy data is collected as cores of salt-marsh sediment and is dated using biological and geochemical sea level indicators such as foraminifera. By combining statistical models and raw proxy data, Maeve examines historic sea level which is pivotal for our understanding of current sea level. The results from Maeve’s models have shown that sea level rise has increased dramatically since the 20th century along the US Atlantic coast. The current sea level along North America’s east coast is the highest it has been in at least the last 15 centuries. 

Click to register for the event

Tuesday 9th February

Time: 14:00-15:00

Event: Estimating Modern Contraceptive Use - Moving from Models to People

Speaker: Hannah Comiskey, Maynooth University

FP2020 is a global initiative funded by the Bill and Melinda Gates Foundation. It aims to bring on an additional 120 million additional users of modern contraception in 69 of the world poorest countries by the year 2020. This year, this goal has been extended to 2030 in line with the United Nations Sustainable Development goals. 

The initiative uses a data driven approach to monitor the progression of its goals. One of the key indicators used to monitor this progress is estimating a country's modern contraceptive use. Presently, the Family Planning Estimation Model (FPEM) calculates modern contraceptive prevalence rates (mCPR) using a Bayesian hierarchical model and Demographic Health Survey (DHS) data. The DHS are a highly reliable data source but expensive to collect. As such, the dataset used to predict modern contraceptive use is very scare with some countries having no data whatsoever. 

Hannah’s talk will look at the work she has been doing to use an alternative grassroots data source called Service Statistics for producing estimates of modern contraceptive use (EMUs). EMUs are derived from service statistics. They are biased estimates of mCPR. Hannah will highlight the existing methodology associated with producing EMUs from Service Statistics and the modelling she has been doing to quantify the uncertainty associated with these estimates. Lastly, Hannah will show an R package she has been working on called “ss2emu” which takes raw service statistics and produces EMUs using the existing methodologies as well as incorporating some of her own modelling strategies.

Click here to book a place

Time; 15:30-16:00

Event: Estimating Abundance in Animal Communities

Speaker: Niamh Mimnagh, Maynooth University

Obtaining accurate estimates of animal population size (abundance) is increasingly important in areas such as conservation and wildlife management. The univariate N-Mixture model (Royle, 2004) was developed with the aim of estimating animal abundance from simple count data, which is prone to imperfect detection. In this presentation, Niamh will examine this existing methodology, and how the univariate N-Mixture Model can be extended in order to model populations of multiple species simultaneously, and to determine the relationships between those species. Niamh will finish by looking at the modelling she has been doing to quantify uncertainty associated with these estimates.

Click here to book a place

Wednesday 10th February

Time: 13:00-13:50

Event: Tell us what you're doing! : using library data to inform scholarly communications

Speaker: Leo Appleton

Audience: Library staff interested in writing for academic publication and those already published who wish to develop their peer-reviewed publication output.

It is commonly thought that the role of the librarian in research and scholarship is to support it - to be the gatekeeper and organiser of the information, data, knowledge and resources that are accessed by others to use in their research. Of course, this is not the only role that librarians play in the world of research, scholarship and publication. They are also major contributors, with library and information science being a large academic discipline, into which many sub-disciplines fall including that of library and information services management, and library practitioners have been contributing to the body of knowledge about librarianship and library management for many years. Librarians are surrounded by data about their practice, whether this be quantitative or statistical data or qualitative data from case studies, reflections or evaluations. This session will look at how practicing librarians, from all sectors, can use their day-to-day library practice and activity to inform their scholarship and how to turn what they do into a publication. From the professional press to peer reviewed publication, there are many outlets for library practitioners to publish. This presentation will unpick this 'library publishing' environment and will provide advice and tips for 'getting started' and also from a 'next steps' perspective for those who may already have experience in academic publishing.

Click here to book a place

Time: 14:30-15:15

Event: Open Science & Research Data Management: An Overview

Speaker: Ciarán Quinn, Research Support Librarian, MU Library

This short talk will attempt to set the context for open science & open data as part of a move to make the primary outputs of publicly funded research accessible and introduce Research Data Management. The RDM component will examine why data needs to be managed and how to develop Data Management Plans.

Click here to book a place

Thursday 11th February

Time: 11:00-11:40

Event: Ireland’s Open Data Initiative – making data accessible and available for reuse

Speaker: Rhoda Kerins, Head of Open Data Unit, Department of Public Expenditure and Reform

Click here to book a place

Time: 14:00-14:45

Event: What does open research mean? An introductory guide 

Speaker: Peter Mooney, Maynooth University

In the past twelve months we have witnessed an unprecedented coordinated effort from the scientific community in tackling COVID-19 pandemic across the world. This global effort has been enhanced through the use of open datasets, sharing of results, public access to scientific papers, etc. But what does open research mean for young scientists, early career researchers, etc. as they embark on their own scientific careers and endeavours. In this short talk we consider what open research means in terms of the use of open data, use of open access for publishing, the use of open source for software code and other technologies. Why does opening up science and research have such potential impact. The talk will be of interest to a wide range of students, academics and professionals who are interested in learning about opening up research - not just to other scientists but to policy makers, researchers, citizen scientists and the public. 

Click here to book a place

Time: 15:30-16:15

Event: EndNote X9 

Speaker: Ciarán Quinn, Research Support Librarian, MU Library

MU has recently acquired access to EndNote X9 which is the Desktop version of Endnote Online which MU users may be already familiar with. This session will guide attendees through the process of downloading Endnote X9 on to their devices, synchronising it with their EndNote Online accounts and detailing the various enhancements they can expect by using this new software. These include unlimited storage of PDFs, annotation of PDFs, finding full text for bibliographic records and much more.

Click here to book a place

Poster presentation

Event: Visualizing Importance and Interaction effects using the R-package vivid

Speaker: Alan Inglis, Maynooth University

Variable importance, interaction measures and partial dependence plots are important summaries in the interpretation of statistical and machine learning models. We describe new visualisation techniques for exploring these model summaries. We construct heatmap and graph-based displays showing variable importance and interaction jointly, which are carefully designed to highlight important aspects of the fit. We describe a new matrix-type layout showing all single and bivariate partial dependence plots, and an alternative layout based on graph Eulerians focusing on key subsets. Our new visualisations are model-agnostic and are applicable to regression and classification supervised learning settings. They enhance interpretation even in situations where the number of variables is large and the interaction structure complex. Our R package vivid (variable importance and variable interaction displays) provides an implementation and provides more easily customisable and plotting options than displayed here.

Click here to view the poster presentation