Data Science Modelling

SOST70033 Data Science Modelling 20 Postgraduate Taught Level 7 Ioana Macoveciuc Unit coordinator 1 FHEQ level (Framework for Higher Education Qualifications) ' Masters/Integrated Masters P4 ' European Credit Transfer & Accumulation System Rating : 10.0 This unit aims to prepare students to handle high-dimensional and complex datasets in the social sciences (e.g. criminology, politics, sociology, psychology etc.). The course unit is designed to help students develop technical competence and robust foundations in and of the underlying principles of various supervised and unsupervised classification and forecasting methods to be able to competently interpret analyses output. The unit will make use of real data from across the social sciences and will further develop practical skills in R and RStudio software. Ethical considerations will also be integrated throughout the course unit to further cement integrity-based use of ‘big’ data.  This unit aims to prepare students to handle high-dimensional and complex datasets in the social sciences (e.g. criminology, politics, sociology, psychology etc.). The course unit is designed to help students develop technical competence and robust foundations in and of the underlying principles of various supervised and unsupervised classification and forecasting methods to be able to competently interpret analyses output. The unit will make use of real data from across the social sciences and will further develop practical skills in R and RStudio software. Ethical considerations will also be integrated throughout the course unit to further cement integrity-based use of ‘big’ data.  The unit aims to: Introduce students to some of the most popular models used in data science to draw valuable insights from high-dimensional, complex, and large datasets, with improved prediction accuracy.  · Summarise the necessary procedures for handling high dimensional and complex datasets;  · Outline the principles underlying the methods and models addressing different classification and forecasting problems;  · Discuss the ethical implications of machine learning and ‘big data’.  · Ability to select the appropriate analytical tools given specific applications;  · Interpret statistical output from a variety of methods and models;  · Assess statistical techniques in the context of research question and data used.  · Devise justified plans to handle large and complex datasets using R and RStudio;  · Select and estimate supervised and unsupervised learning models in the context of empirical applications.  · Utilise statistical programming software (R and RStudio) to interrogate high-dimensional data and draw valuable insights using various machine learning techniques and models;  · Formulate, organise, express, and communicate data-driven opinions effectively.  The course unit will consist of asynchronous videos, interactive activities, and guided online readings. In addition, there will be frequent synchronous sessions such as seminars, tutorials, and practicals. The course will include two summative assignments. Throughout the course unit, students will solve real world data science problems (practical exercises) and write up the main findings which will be assessed. In preparation for their second assignment, students will prepare a methodological critique for which they will receive formative feedback. Finally, the students will write a larger analytical report based on what they learned throughout the course unit. Students will also be provided with additional materials such as assigned readings, video demonstrations, and various other activities to supplement their learning.  0 Other 100% Methodological Critique (500 words, non-assessed) Practical Exercises (x4 problem-based exercises in R, 40%) Analysis Report (2000 words, 60%)  Online feedback within 15 working days.  James, G., Witten, D., Hastie, T. and Tibshirani, R. (2017) An Introduction to Statistical Learning with Applications in R. New York: Springer.  Hadley Wickham and Garrett Grolemund. (2017). R for Data Science: Import, Tidy, Transform, Visualize, and Model Data. O’Reilly Media. For Information and advice on Link2Lists reading list software, see: <a href="http://www.library.manchester.ac.uk/academicsupport/informationandadviceonlink2listsreadinglistsoftware/" rel="noreferrer noopener" target="_blank">http://www.library.manchester.ac.uk/academicsupport/informationandadviceonlink2listsreadinglistsoftware/</a>   0 0 173 There will be no on-campus contact hours and learning will take place fully online. Contact time – 27 hours