MA Computational Linguistics

This module will focus on how language structure relates to meaning and communication. We’ll look at various aspects of language structure – sounds, words, sentences and short texts, and relate them to function and meaning.

For example, we’ll discuss how speakers deploy different grammatical resources to give varied representations of events and their participants. We’ll also explore how different structural choices help speakers express their knowledge, and certainty or uncertainty of events, or how they impose obligations or grant permission to their conversational partners, what means they use to express attitudes and take a stance, and how they figure out what others mean even when they mean something different from what they say.

The module aims to make clear the wider aims of linguistic research, as well as enable you to apply theoretical notions to specific data sets and develop your own skills of linguistic analysis, especially as it relates to human interaction and communication. Language structure will be seen as contextualised and contingent, subject to variation at any one time and change across periods of time.

This module aims to provide you with the programming skills you will need to carry out for programming tasks you will encounter in the other modules in this programme. You'll learn about general programming techniques such as variables, functions and control flow. You'll explore how to work with different types of data structures such as arrays and dictionaries.

You will develop data processing pipelines, which allow you to convert raw data into data that you can analyse. You will apply mathematical and statistical procedures to data. You'll learn how to plot graphs of various types. You'll also familiarise yourself with an industry-standard data science programming environment which you can use throughout the programme.

The following is an indicative list of topics:

1. Data structures
2. Reading and writing data on the filesystem
3. Retrieving data from the web
4. Retrieving data from databases using query languages
5. Cleaning and restructuring data, part 1
6. Cleaning and restructuring data, part 2
7. Data plotting
8. Version control systems
9. Unit tests
10. Data processing pipelines

Corpora - large collections of language in electronic form - have become one of the most powerful tools in the linguistics toolbox. They provide researchers with large bodies of naturally occurring linguistic data on which to test hypotheses and verify generalisations. We can use corpora to explore how language works, how it changes over time and how it varies between speakers and situations.

Corpora can be used to study relatively small elements of language, e.g., morphemes, or particular words, but can also aid our analyses of larger texts and discourses. Corpora give us another way to query linguistic data – rather than using introspection to understand language and our intuitions, we can use technological tools and objective statistical measures to identify significant patterns, trends and insights which might otherwise go unnoticed.

In this module we take a multi-disciplinary approach to the rapidly evolving field of corpus linguistics. We explore both how corpora are built, and how they are used. First, we cover the basics of corpus construction. How do we decide what texts to include, how do we capture the information we have about the texts we have selected, and how do we make our corpus usable for further investigation.

We then move on to how corpora can and have been used. We introduce the basic statistical concepts and analysis techniques used in corpus linguistics. We then explore some of the applications of corpus linguistic research, e.g. to the study of language structure, meaning, variation and change, and discourse.

This module gives you an introduction to machine learning and statistical learning. Topics include supervised learning (regression and classification as parametric and non-parametric methods), and unsupervised learning, but discusses also semi-supervised learning, reinforcement learning, and elements of the learning theory.

You'll explore recent applications of machine learning and deep learning by illustrating practically, using modern software libraries, how machine learning algorithms and methods are applied effectively, and how models’ performances can be optimised using real-world datasets. Ethical aspects related to machine learning are also discussed.

Computers process massive amounts of information every day in the form of human language. Although they may not understand it in the way that people do, they can learn how to do things like answer questions about it, summarise it, translate it into other languages, and so on. This module gives you a systematic introduction to the ideas that form the foundation of current language technologies and research into future language technologies.

For your final project, you'll be asked to design and implement an independent piece of work. This could be a project which investigates a linguistically relevant research question in some area of interest (grammar, vocabulary, semantics, pragmatics, discourse analysis, interactional data, machine translation, second language acquisition) in a theoretically informed way, and using some computational resource or natural language processing technique (for instance, a self-compiled or existing corpus, or sentiment analysis, or automatic text recognition, etc.).

You could also develop and evaluate computational techniques and tools that can be used to perform NLP-relevant tasks.  Students will be expected to situate their work within the wider field of (computational) linguistics, providing a critical discussion of theoretical problems relevant to their work; and/or similar analyses; and/or similar applications, technologies, etc. You'll be expected to present and evaluate the results of their work in a coherent project report, which adopts relevant conventions and professional standards of presentation.

The module introduces you to research topics related to Data Science, for example:

• Modern applications of Machine Learning
• Data Mining, Computational Statistics
• Computational Social Sciences
• Cognitive computing
• Business analytics and its applications

In particular, you will have an opportunity to learn from Data Science and Computing industry professionals and academics whose mono-disciplinary or interdisciplinary research in areas of Computer Science, Sociology, Bioinformatics, Biomedical Statistics and other disciplines is based or involves data analysis. The module will guide your work in exploring a research theme of interest, as a preliminary phase preparing you for the Final Project. 

The module will also expose you to learning techniques of conducting research and writing scientific reports. Moreover, you will be taught approaches for devising and optimising your strategies adapted to your skills in seeking and securing an employment, whether in research or the industry sector. The module will also expose you to a discussion of the ethical and legal issues related to the area of Data Science and its applications. 

In this module, you'll learn the key statistical concepts and techniques for data analysis and statistical data mining. We'll cover some of the most instrumental methods of data analysis and statistical data mining including:

• Descriptive statistics
• Elements of probability theory including Bayes theorem
• Elements of statistical inference (statistical tests and confidence intervals)

You'll explore methods for regression and classification such as Linear Regression and Logistic Regression, Bayesian methods such as Naïve Bayes and Discriminant Analysis, K-Nearest Neighbours, and also Decision Trees which are mostly applied in ensemble based methods.

You'll learn how to apply these statistics and data mining methods to analyse various datasets, and how to interpret the results.

In this module, you'll learn all about Big Data which is a key element of contemporary applications of Data Science. You'll gain practical skills related to working with Big Data computing resources as well as the conceptual context of how Big Data relates to methods and technologies in statistics and computing.

We'll cover ten topics on various theoretical, methodological, and technical aspects of big data analysis. A substantial amount of time will be spent on understanding big data technologies like Apache Hadoop and computing engines like Hadoop's MapReduce and Apache's Spark.

You'll start with a brief introduction to the basics of big data and the importance of distributed computing. We'll focus on the importance of bringing computing to the distributed data sets and how Hadoop Distributed File System (HDFS) is so crucial for doing that.

We'll then move on to distributed computing framework and intuitively understand the programming methodologies like MapReduce. We consider cluster analysis as our first example for big data analysis. We use MapReduce methodology and implement k-mean clustering algorithm by performing the required computation on the cluster of computers. You'll learn the basics (that are sufficient for the application developers) about the scheduling aspects of Hadoop by knowing the details of various scheduling algorithms related to Hadoop's YARN.

Later, we change gears towards Apache Spark where we focus our attention on Spark's Resilient Distributed Datasets (RDDs) and the concepts of Spark's transactions and actions. We go through the basics of functional programming in Spark. Towards the end, we focus on implementing machine learning algorithms on Hadoop cluster using Spark framework. Finally, you'll study Recommender Systems, which are so useful in filtering the required information from the rich data sets that are usually produced by the digitalisation of businesses in the real world.

Visualisation is essential for understanding and communicating information, and for making informed decisions based on data. This module takes the view that data visualisation is a core interdisciplinary component of data science. Effective visualisation requires a combination of computational skills, statistical knowledge, an understanding of human visual perception, and a rigorous and creative approach to working with data.

In this module, you'll explore both the practical skills necessary for manipulating and visualising data, as well as the theoretical knowledge essential for making judgements about how to most effectively discover and communicate insights from data.

You'll become familiar with core visualisation tools and techniques within the Python data science ecosystem. You'll establish a critical approach to data visualisation, starting with interrogating how data is collected, continuing throughout each stage of the visualisation process culminating in the communication of information and possible real-world implications. You'll gain practice answering questions with data and build confidence in presenting your findings, ensuring they are trustworthy and valid. You will gain knowledge of the human visual system, which is essential for designing effective visual communication.

Through this module, you'll learn a wide range of Artificial Intelligence techniques. The module complements the Machine Learning module by focusing on Artificial Intelligence-related topics it does not cover, such as symbolic representations, modelling, task learning and game playing. Through this module, you'll widen and deepen your knowledge of Artificial Intelligence techniques, which should provide you with an additional set of methods with which you can address Data Science problems. 

This module is focused on Artificial Intelligence techniques. You'll understand the historical development of Artificial Intelligence including search, vision and planning. You will be familiar with the foundations of the agent-based approach to software design, decision making and problem-solving including under uncertainty. You'll have an opportunity to apply Artificial Intelligence techniques. 

Neural Networks are widely used techniques for modelling and classifying data. They are used in industry for data analysis applications such as image classification, speech analysis and regression tasks.

This module will provide you with specialised theoretical and practical knowledge of a range of Neural Network architectures that are appropriate for data-oriented tasks. This module is complementary to the Machine Learning and Artificial Intelligence modules in the programme, focusing on the area of neural computation.  

You'll be introduced to the theory and practice of neural computation. You'll learn the principles of neuro-computing with Neural Networks widely used for addressing real-world problems such as regression, pattern recognition and time-series prediction. 

This module aims to cover a wide range of applied data analysis techniques using R. With its strong focus on the practical application of the techniques, you'll explore the following topics:

• Data sources
• Data gathering
• Interpretation of statistical results
• Visualisation
• Principal component analysis
• Practical implementation using R
• Decision-making based on data analysis.

This module will introduce you to common investigative methods for user-centred design projects. You'll explore research methodologies at each stage of the project lifecycle: data gathering, analysis, and evaluation. You'll be introduced to a theoretical and critical understanding of qualitative and quantitative research. You'll learn how to identify and employ the most appropriate methodologies and techniques for a given project. The module will include practical examples and scenarios that illustrate the principles and techniques.

In this module, you'll address the science behind interaction – understanding human cognitive/motoric/perception capabilities and the science of the human mind. You'll interrogate concepts from human-computer interaction, cognitive science and interaction design practice. Students will study modelling techniques, including HCI frameworks, computational models and connectionist/neural approaches. You'll explore the principles that shape human behaviour when engaging with interactive products and systems and how to apply those principles to the design of future technological solutions. 

This module will introduce you to the analysis of discourse and identity in spoken interaction. The course will allow you to develop in-depth, critical understanding of approaches, concepts and debates in spoken discourse analysis. The second aim of the module is to provide you with the opportunity to apply your newly acquired methodological insight to the study of discourse and identity in many different conversational and institutional settings.

A range of methodological frameworks and analytic concepts will be explored, including ethnographic approaches to language analysis, interactional sociolinguistics, conversational analysis, membership categorisation analysis, performativity and narrative analysis. Seminar discussions will seek to establish what each of these approaches have to offer to the analysis of discourse practices and identity constructions of speakers in naturally occurring talk. For example, we’ll consider the question if analysts should or avoid bringing inferable assumptions about the relevance of macro identity categories such as gender and social class to their data.

You’ll also be encouraged to carry out your own project by collecting, transcribing and analysing a sample of spoken language of your choice. You’ll then get the opportunity to present and discuss your work in seminars.

The aim of this module is to introduce you to a range of linguistic approaches of written discourse analysis. The module explores a range of frameworks and leads students to a discussion of how the analysis of texts can illuminate wider social issues, for example issues of power and ideology and issues of representation and identity. The seminars endeavour to give you the space to apply the techniques explored in the reading to a wide selection of texts (texts from the contemporary media, advertisements, textbooks, political and administrative texts, texts in translation, etc.)

You’ll acquire knowledge of different levels of linguistic analysis and learn to examine written discourse at the micro-level, and to link the micro to the macro. The use of a variety of texts is intended to lead students to debates about language use and social issues in different areas of human activity: mainly media, but also translation, education, etc. You’ll be encouraged to engage with the research literature and apply the theoretical concepts and linguistic approaches you become familiar with to independent analyses of self-selected data.

This module combines a sociolinguistic with a discourse analytic approach in order to explore the socio-cultural contextualisation of language and meaning from two angles: language use and language representation. This dual focus will be evident throughout the course; topics such as language and gender, language and ethnicity or language and the media will be examined in relation to the socio-cultural (and situational) contexts in which speakers use language as well as in relation to different representations of specific socio-cultural groups in the media and other (written) texts.

For example, we’ll investigate both how women speak and how women are spoken about (e.g. sexist language). Other topics that will be addressed in this module include the political correctness debate; attitudes to non-standard English; multicultural London English and the linguistic construction of identity.

The module will introduce you to a wide range of empirical research and methodologies. Drawing on analytical tools and frameworks from semiotics, pragmatics, (critical) discourse analysis, conversation analysis, feminist linguistics, ethnography and variationist sociolinguistics, we’ll investigate how language is used and meanings are created, interpreted and contested in a range of different texts, discourses and socio-cultural environments.

This course introduces you to key themes, studies and methods that have shaped the interdisciplinary field of intercultural communication. We’ll address questions of cultural difference and diversity from a range of theoretical and methodological approaches. We’ll study cross-cultural patterns of (communicative) practices, attitudes, values and cognition. We’ll ask how speakers construct culture and cultural identities in interaction, both in everyday private settings (e.g. couples talk) and in work/institutional settings (including business and education).

We’ll also explore how cultural norms and stereotypes are reflected in language and/or discourse and how they affect our thinking. We’ll focus our discussion on various levels of language, including speech events, speech acts, interactional styles, politeness phenomena and written discourse.

Throughout the course, we’ll consider the term ‘culture’ critically, comparing popular definitions of ‘culture’ as homogenous and static with postmodern models that highlight the heterogeneity and fluidity of ‘culture’. You’ll become familiar with a range of methodological approaches to the study of language and culture, including the ethnography of communication, discourse analysis, interactional sociolinguistics, intercultural pragmatics and politeness theory. The course will not only ask you to study cultural and communicative norms and practices in a range of different English-speaking countries and settings, but it will also draw on research from a wide variety of languages. Students will be asked to draw on their own linguistic and cultural backgrounds and personal experiences in their critical engagement with this interdisciplinary field of study.

The overall aim of this module is to explore the diversity of the English language in relation to linguistic and social issues involved in language contact and multilingualism. You’ll have the opportunity to study the spread of English and the rise to its current status as a global language, discuss the establishment of (English) language standards and (standard) varieties world-wide, the emergence of English as a Lingua Franca, translanguaging and other language contact phenomena.

The focus will be on the challenges and opportunities open to multilingual societies and to consider the impact of multilingual settings on individuals. An understanding of Global Englishes and aspects of multilingualism provides you with the necessary conceptual and theoretical tools to understand English practices in a multilingual world and to conduct your own research within an area of interest.

Artificial Intelligence, and particularly machine learning, has advanced tremendously in recent years and it is now being employed in systems that people use on a daily basis (e.g. Google search or Facebook) and often makes potentially life-changing decisions in areas such as policing or healthcare. These new applications necessitate a fundamental change in how we view AI. It is no longer simply a technical exercise, it has important impacts on people. It is therefore vital to take a human-centred approach to AI and apply the tools of User Experience Engineering to designing AI systems. In this module, you will learn about the human context of modern AI and how to design Human-Centred User Experiences using AI technology. These will be based on traditional AI methods such as neural networks and also more recent Human-Centred innovations such as Interactive Machine Learning, Explainable AI and Mixed-Initiative Systems.