NATURAL LANGUAGE PROCESSING - 2022/3

Module Overview

This module will demonstrate some of the latest advances in Natural Language Processing and get student up to speed with current research. It will provide the necessary skills to enable students build models for solving a range of problems, such as document classification, translation and conversation agents. The students will learn how to build NLP pipelines for preparing training data and choosing appropriate algorithms and techniques to build such models. Although traditional linguistic methods will be mentioned, special emphasis will be put on the state-of-the-art Deep Learning algorithms and Transfer Learning methods for building efficient Machine Learning based NLP solutions.

Module provider

Computer Science and Electronic Eng

Module Leader

KANOJIA Diptesh (CS & EE)

Number of Credits: 15

ECTS Credits: 7.5

Framework: FHEQ Level 7

Module cap (Maximum number of students): N/A

Module Availability

Semester 2

Prerequisites / Co-requisites

• Foundation of Machine Learning • Python familiarity or other programming language, ideally used in data science.

Module content

Indicative content includes:

•  Introduction to NLP


•  Traditional linguistic processes


•  Data pre-processing and text analytics


•  Neural Networks for NLP
  
  
  
  • RNN
  
  
  • LSTM/GRU 
  
  
  • CNN
  
  
  
  


• Coreference, question answering, and machine translation


• Syntactic and semantic processing


• Topic modelling


• Attention mechanisms and sequence embedding models


•  Hardware requirements in training scalable models for language data
  
  
  
  • Minibatching
  
  
  • GPU implementation issues
  
  
  
  


• Deploying and scaling NLP models as web micro-services


•  Monitoring and updating models

Assessment pattern

Alternative Assessment

Individual coursework covering the same learning objectives

Assessment Strategy

The assessment strategy is designed to provide students with the opportunity to demonstrate:

• ability to critically evaluate, experiment, and apply, appropriate techniques to datasets exemplifying specific characteristics in order to derive suitable and defensible results.

• ability understand the needs and build appropriate solutions for a range of NLP problems

Thus, the summative assessment for this module consists of:

• An individual coursework involving a report comprising the comparative evaluation of NLP methods when applied and adapted, with documentation of key insights derived. The coursework will be marked based on the approach and breadth of experimentation, rather than on the performance obtained. This will evaluate LOs 2, 4 & 5, and have a deadline in or near to week 8.

• A group coursework in which the students will bring together their experiments from the first coursework and will build a proof of concept that will need to demonstrate both technical understanding and good practice – submission deadline: beginning of week 11. This addresses LOs 3, 5 & 6.

• Students will be guided to work on weekly tasks through lab exercises, the solutions to which will provide for feedback on understanding and practice. Labs and feedback will then support the coursework. Individual feedback on the coursework will be given as soon as possible before the exam in order to feed forward to the exam.

Formative assessment and feedback

Students will be guided to work on weekly tasks through lab exercises, the solutions to which will provide for feedback on understanding and practice. Labs and feedback will then support the coursework. Individual feedback on the coursework will be given as soon as possible before the exam in order to feed forward to the exam.

Module aims

• • Provide an overview of the technologies and algorithms that support the development of NLP solutions.
• • Familiarise students with NLP applications and what approaches can be adopted to experiment and build such applications.
• • Demonstrate how NLP processing pipelines can be formed to perform necessary transformations for preparing data for training.
• • Bring students up to a sufficient level of development skill to be able to develop NLP models that solve specific business needs.

Learning outcomes

Methods of Teaching / Learning

The learning and teaching strategy is designed to:

• Develop a critical understanding of best practice in developing state of the art NLP solutions through directed learning and facilitated self-directed learning. The skills learned in this module will be transferable to other data science modules in the programme and the wider data science profession.

The learning and teaching methods include:

• Twenty-two hours of lectures with class discussion

• Twenty hours of lab classes

• Use of an online forum for facilitated discussion

Indicated Lecture Hours (which may also include seminars, tutorials, workshops and other contact time) are approximate and may include in-class tests where one or more of these are an assessment on the module. In-class tests are scheduled/organised separately to taught content and will be published on to student personal timetables, where they apply to taken modules, as soon as they are finalised by central administration. This will usually be after the initial publication of the teaching timetable for the relevant semester.

Reading list

https://readinglists.surrey.ac.uk
Upon accessing the reading list, please search for the module using the module code: COMM061

Other information

None

Programmes this module appears in

Please note that the information detailed within this record is accurate at the time of publishing and may be subject to change. This record contains information for the most up to date version of the programme / module for the 2022/3 academic year.