SeNSS is offering the opportunity to become a SeNSS Digital Scholar.

As one of the first scholars, you can learn new skills in data mining, data analytics, digital ethnography and AI, through a series of specialist practical workshops in computational digital social science research methods. Full details of each session can be seen below.

There will be 15 places available on each workshop so places are limited.

If you are interested in attending the workshops, please apply by completing the Online Application Form by 26 April 2019


WORKSHOP 1: Web Scraping for Social Scientists - A practical introduction

WHEN: 18 June 2019

WHERE: University of Essex

SESSION DETAILS:

The internet offers a vast array of potential data that could open up valuable opportunities for social science research. However many of these spaces do not offer easily accessible versions of their data, making the manual extraction of the information a slow and laborious task. Using web-scraping, researchers can build their own custom tools to automate these processes, transforming unstructured online phenomena into well-structured analysable datasets. This lab-based session will introduce students to the Python programming language and provide step by step instruction on building a custom web scraper. Students will learn about the underlying HTML structure of websites, how to retrieve HTML code, and how to extract specific components for storage in a structured data frame that can be exported for further analysis. Students will also learn how to automate all of these processes in an ethical and respectful manner, dramatically increasing both the variety and scale of data available for social science research.

LEARNING OUTCOMES:

By attending this session Scholars will be able to:

  • Initiate a Python programming environment
  • Utilise Python at a beginner's level for a range of basic activities.
  • Retrieve and extract data from the web
  • Automate the extraction process
  • Export a structured version of the data for further analysis.

SESSION DELIVERED BY:

JAMES ALLEN-ROBERTSON

Dr. James Allen-Robertson is a Digital Sociologist at the University of Essex. His research focuses on utilizing data science techniques for innovative social science research, and in studying the relationship between digital technologies and society. Methodologically he focuses on using data extraction techniques such as web scraping and social media APIs to develop new kinds of social science data, utilizing natural language processing for large scale analysis of text, and using machine learning for the exploration of media discourses. He is currently using these skills for projects on mapping the dark web, social media discourses on environmental activism and exploring the discourses of workers under algorithmic management.

WORKSHOP 2: Telling a story through maps – a practical introduction to Geographical Information Systems

WHEN: 23 July 2019

WHERE: University of Essex

SESSION DETAILS:

The use of Geographical Information Systems (GIS) in the social sciences is increasing. A GIS allows us to visualise, question, analyse and interpret data to understand relationships, patterns and trends with both qualitative and quantitative data. This lab-based session will introduce students to ArcGIS online, a software tool widely used by research and commercial communities. Students will use a number of datasets to explore social science issues including crime, deprivation and community cohesion. As well as mapping secondary data, students will also learn how to add their own data to a map and how to collect georeferenced data in the field and how to present their data as an interactive storymap - an emerging technique in digital ethnography.

LEARNING OUTCOMES:

By attending this session Scholars will be able to:

  • navigate the ArcGIS Online website
  • map administrative data
  • add their own data
  • create a data capture app and
  • use a storymap to present their maps

SESSION DELIVERED BY:

RUTH WEIR

Ruth Weir is a Research Fellow in the department of Government at the University of Essex. After graduating in Geography from University College London (UCL) in 2001 Ruth worked as a researcher in the Community Safety Unit at Suffolk County Council. It was whilst working here that she developed an interest in crime mapping, which resulted in her returning to UCL to complete an MSc in Geographical Information Science. Ruth has since worked as a GIS specialist in the Research, Development and Statistics Directorate at the Home Office before returning to local government to work in research and intelligence roles. Ruth has presented at many national and international conferences on GIS and published a Home Office Research paper on the use of GIS by crime analysts. Her PhD research used advanced GIS techniques to identify potential predictors of unreported domestic abuse and to conduct a community asset mapping exercise.

WORKSHOP 3: Using mobile devices as data collection tools for social scientists.

WHEN: September 2019

WHERE: Goldsmiths

SESSION DETAILS: Smart phones, tablets and smart watches are becoming increasingly popular and hold great potential for data collection outside the laboratory, and over extended periods of times. In this session we will introduce some of the methodologies, chances and challenges that underlie experience sampling and continuous self-report measures.

LEARNING OUTCOMES

By attending this session Scholars will be able to:

  • understand the current state of the art of data collection from mobile devices, including smart watches and mobile phones
  • identify chances and challenges for experimental design and data analysis
  • identify and apply suitable statistical procedures for the analysis of continuous self-report and experience sampling data, including agreement analyses and granger causality.

SESSION DELIVERED BY:

Guido Orgs

I studied Psychology and Performing Dance in Germany, and hold a PhD in cognitive neuroscience. My research deals with how people form preferences and how pleasure and liking can be measured in people’s brains and bodies. I apply experimental methods to understand how branding and advertising work (Neuromarketing), but also study people’sexperience of the live performing arts, including dance, theatre and music. I am a Senior Lecturer in Psychology at Goldsmiths, University of London and direct the MSc in Psychology of the Arts, Neuroaesthetics and Creativity, the first postgraduate programme in the world for the scientific study of how people appreciate and make art .

WORKSHOP 4: Predictive analytics, topic modeling and text analytics using machine learning in social science – a practical introduction

WHEN: 11 October 2019

WHERE: University of Reading

SESSION DETAILS:

There is huge potential for social science researchers to leverage big data to answer research questions with more insights from different sources and formats.Compared with traditional statistical methods, machine learning methods, one of the streams of AI, offer more robust ways to discover patterns to make accurate predictions. This session will introduce basic machine learning methods including classification and topic modelling with case studies of using healthcare open data and financial disclosures data. The student will learn about feature selection, dimension reduction, model training and evaluation under python and metlab environment. The student will leave the session with an understanding of key machine learning models (support vector machine, naïve bayes, random forest and artificial neural network), latent Dirichlet allocation (LDA) and text analytics methods. The students will also learn about case studies of machine learning based prediction model in a real healthcare scenario and understand core issues and opportunities about machine learning in practice.

LEARNING OUTCOMES:

By attending this session Scholars will be able to:

  • Understand machine learning based classification models
  • Understand text analytics and topic modelling methods
  • Uitlise python-based library for data preprocessing and build prediction model
  • Utilise metlab based library to build topic modelling model
  • Understand the use cases of machine learning in practice.

SESSION DELIVERED BY:

WEIZI (VICKY) LI

Dr. Weizi Li, Associate Professor of Informatics and Digital Health, Deputy Director in Informatics Research Centre, Henley Business School. Her research focuses on digital health, integrated system for clinical pathways, artificial intelligence and machine learning applications in healthcare. She is a Fellow of Charted Institute of IT (FBCS). She worked as the system and policy analyst in Betsi Cadwaladr University Health Board, for the development of integrated research governance system in the largest NHS organization in North Wales. Her research output of integrated health data integration platform and intelligent artificial agent system have been successfully implemented in more than 400 major hospitals in China and received O2RB (University of Oxford, Oxford Brookes, University of Reading and Open University) Excellence in Impact Award supported by ESRC (April 2018). She is now working with NHS and companies on using machine learning in healthcare resource management and population engagement supported by ESRC Artificial Intelligence PhD studentship funding.

KEIICHI NAKATARA

Keiichi Nakata is Professor of Social Informatics at the Informatics Research Centre, Henley Business School at the University of Reading, UK. His main research interests lie at the interface between technology and people, in the areas of computer-supported collaborative work, cognitive systems engineering, and information systems. Recently he has been engaged in research into acceptance of pervasive systems, social media, and participatory systems. Prior to the current appointment, he was Dean of School of Information Technology at International University in Germany. His past appointments include Associate Professor at the Institute of Environmental Studies at the University of Tokyo, and Research Scientist at German National Research Centre for Information Technology. He obtained his Ph.D. in Artificial Intelligence from the University of Edinburgh, UK, and M.Eng. and B.Eng. in Nuclear Engineering from the University of Tokyo.

ANUPAM NANDA

Anupam Nanda is currently Professor of Urban Economics and Real Estate, Research Division Lead/Director of Research (for Real Estate and Planning), Academic Director of the Centre for intelligent Places and Director of the MSc Real Estate Finance programme at the Henley Business School, University of Reading, UK.efore becoming a full professor, Anupam held positions of Associate Professor in Real Estate Economics (Oct. 2013-Jul. 2017) and Lecturer (Jan. 2010-Sep. 2013) at the University of Reading. Before joining Henley Business School, he worked with the Market Intelligence group of Deloitte & Touche in Mumbai (Apr. 2008-Nov. 2009), where his focus areas covered real estate and private equity sectors. He was at the National Association of Home Builders (NAHB) in Washington DC (Apr. 2006-Apr. 2008), as Senior Research Economist, where his responsibilities included developing and implementing housing market research studies and was a member of the team forecasting state and metro area housing markets in US. Anupam has also taught undergraduate Economics and Public Finance at the University of Connecticut.

WORKHOP 5: An introduction for social scientists to Bayesian data analysis and digital tools to help.

WHEN: 06 November 2019

WHERE: Goldsmiths

SESSION DETAILS: A feature of the “reproducibility crisis” in science and social science has been dissatisfaction with traditional null hypothesis significance testing (NHST), arguing this contributes to poor reproducibility. Bayesian methods are presented as an alternative framework which can help researchers achieve greater reproducibility, although their uptake across difference social science disciplines is patchy.

This workshop will introduce these methods to social scientists and will expose participants to digital tools (i.e., dedicated software) that allows such analyses to be carried out in a relatively straightforward manner.

LEARNING OUTCOMES:

  • Understand the principles of Bayesian approaches to data analysis and in what ways these analyses may (and may not) offer an improvement over traditional NHST methods
  • Use the free software JASP (which can read in data in SPSS format) to carry out a range of Bayesian counterparts to familiar statistical tests (t-tests, ANOVAs, regressions etc)
  • Utilise a Matlab based toolbox for carrying out Bayesian model comparisons on a variety of datasets

SESSION DELIVERED BY:

Alan Pickering

Alan Pickering is a Professor of Psychology and Dean of the Graduate School at Goldsmiths. He is also Deputy Director of the SeNSS ESRC Doctoral Training Partnership. After studying Natural Sciences at Cambridge as an undergraduate, Alan did his PhD in the cognitive neuropsychology of memory at the University of Manchester. During a subsequent post-doc at Kings in London, Alan switched to research on human personality and psychopathology, but always with an emphasis on neural, formal and statistical models of behaviour. After 11 years as lecturer and senior lecturer in psychology at St George’s Medical School, Alan joined Goldsmiths in 2001. His current research is primarily focused on understanding learning in rewarding contexts and the role that dopaminergic neurotransmission may play in reinforcing such learning. He also works on investigating the reinforcement sensitivity theory of personality, in which between-individuals variations in the functioning of the brain's reward and punishment pathways are thought to contribute to variations in extraversion and anxiety respectively.

CLOSING CONFERENCE

WHEN: 09 December 2019

WHERE: TBC

CONFERENCE DETAILS: More information to follow