Dissertation

Background

The purpose of the project dissertation is to provide a complete record of the work carried out by you during the course of the project.

Learning outcomes

Students should be able to:

Description of the task

Your task is to produce a dissertation for your project.

The dissertation is a record of what has transpired over the course of your project. It should detail (at an appropriate level) what was the purpose of your project, what was achieved, what software was designed (if applicable), what hypothesis was being tested (if applicable), experiments performed, data gathered, etc.

The dissertation must be written by yourself using your own words (see the University guidance on academic integrity for additional information). See the section on Academic integrity below for more details.

The typical structure of the dissertation is outlined below.

Structure of dissertation

A suggested structure for the dissertation is as follows:

The dissertation must be self contained, and include a complete record of the work carried out. A target size of 7,000 words is recommended, with a maximum of 10,000 words. Appendices will not be included in the maximum, but assessors will not normally expect to read appendices in detail, so they are intended to supply supporting and illustrative material.

The content of the dissertation is at the discretion of the student, and will depend on the nature of the project, but for a typical project involving the development of a piece of software, the following elements of the dissertation would be expected:

Note: The exact content of these sections shouldn't be consider "fixed", nor do they necessarily need to be in this order. This is just a suggestion of aspects of the project that you want to address in some manner in your dissertation.

  1. ABSTRACT: a 350 word summary of the project as a whole.

  2. INTRODUCTION: This will give a brief overview of the project, what problem it addressed, the solution produced, and the effectiveness of the solution.

  3. BACKGROUND: Reading and research done to acquire the necessary information and skills to carry out the project. A clear statement of the project requirements.

  4. ETHICAL USE OF DATA (including human data and/or human participants): Comment about what data was needed for the project, and how that data was obtained.
    If human data was used for the project, note that this data should be information that is freely available in the public domain, or anonymised records and data sets that exist in the public domain. Otherwise, human data could only be used if ethical approval was obtained for gathering that data (prior to the acquisition of the data, and for the explicit use in this project).

    Note that data obtained through the use of the Twitter API (paying attention to the Terms and Conditions of use) is ok and does not need university ethical approval.

    If human data was used that does not fall under the "freely available" categories above, you must state (and demonstrate) that you (or, more likely, your supervisor) followed the university procedure to obtain ethical approval.

    The only exception to this rule requiring ethical approval is that the use of human data acquired through the type of questionnaires allowed to students and/or others to evaluate the software you developed, provided you followed the guidelines in requesting this data. See the Ethical Use of Human Data webpage as a reminder.

  5. DESIGN: Documentation of the design; while the organisation should be similar to the design presentation, full detail of the design is required. All original design documentation should be supplied (possibly as an appendix).

  6. REALISATION: How the design was implemented. Changes made to the design in the course of implementation. Testing of the implementation. Typically, (short) code listings, screen shots, and test runs will appear as appendices.

  7. EVALUATION: A critical appreciation of the strengths and weakness of the project as carried out. This may include, where appropriate, customer feedback.

  8. LEARNING POINTS: At least one page summary of key learning points in the project. Learning points could include (but are not necessarily limited to) the skills and knowledge that has been acquired or improved by the project, actions that you consider to have been crucial to the success of the project, and things that you would do differently in the future in order to avoid pitfalls, drawbacks, or failures in a similar piece of work.

  9. PROFESSIONAL ISSUES: Appropriate discussion of how your project related to the British Computer Society (BCS) Code of Conduct. You could also address any issues around the use of human-supplied/human-derived data and/or human participants in this section if you wish (if not done elsewhere in the dissertation).

  10. CONCLUSIONS: Summary, main findings, directions for further work.

  11. BIBLIOGRAPHY: A properly cited list of books, articles, online resources, and other materials consulted during the project and/or referred to in the dissertation.

  12. APPENDICES: Appendices are meant to contain detailed material, required for completeness, but which are too detailed to include in the main body of the text. Typically these include the full code listing (IMPORTANT: The code must be included in the electronic submission!), details of test data, screen shots of sample runs, a user guide to installation (e.g. "I used Python 3.6, together with the sci-kit learn and pandas modules") and usage of the software (as appropriate), full design diagrams, and similar material. One appendix should be a project log, which will list important dates in the projects: including completion of major stages, release of versions of the software, review meetings, and other quality assurance activities.

A LaTeX dissertation template is available, with a PDF file to show what it looks like. Note that the section headers and such provided are only examples.

A Microsoft "Word" dissertation template is also available. This template is reproduced here by permission of Laureate Online Learning, the University of Liverpool's eLearning partner.

For your guidance a copy of the feedback form that will be used to assess your dissertation are available from here: [feedback form]

Submission of work

Submission of the dissertation, any accompanying appendices, and all source code files must be done in electronic form (PDF format only for the dissertation and any separate appendices) via the Coursework Submission System.

Note that zipped versions of the documents and/or source code are acceptable for the electronic submission. Do not use any form of compression other than ZIP!!

Note that during submission of your work, you are also making an online Declaration on Academic Integrity.

The deadline for the electronic submission of the dissertation and source code files is 22 September 2017 (5:00pm).

Students who took resit exams are allowed to submit their work by 6 October 2017 (5:00pm). Please note that these deadlines are strict. Further extensions will only be granted in exceptional circumstances.

The dissertation submission deadline for students who have deferred completion of their project is Monday, 11 December 2017 (5:00pm).

NOTE: In the current version of the system, a student can submit files but not overwrite files on the system. You also cannot remove previously uploaded files. Therefore, it is strongly recommended that each student create and submit a text file with a table of contents (toc) of their submission. In particular, if your name is John Smith, please submit also a text file with the name:

John-Smith-toc.txt

that contains the list of files that need to be assessed. This is to avoid assessment of files that are submitted by mistake.

Academic integrity

All student should be aware that they are responsible for what they write. One of the pillars of progress in research is that authors can benefit from each other's earlier work. Arguments made in a dissertation should be supported by facts. One way of doing this is to refer to the existing body of work. For example one can argue that X is true because Y and Z demonstrated it was true in a number of articles published in reputable journals (and then give references to the publications in question). If readers want to disagree with you they also have to take issue with X and Y!

However, it is important for students to make clear, when writing their dissertation, what their original contribution is and what is not. If a student is unable to make a point more clearly than a source that they have found (a book, a paper, or a document on the web), they should use quotations: put the quoted sentence(s) in between quotes " and ", and make clear in the running text where the reference is taken from. Then, cite that source in your bibliography and/or list of references. There are many standards to do citation, students are free to use any style, but should make sure that they make citations in a consistent way.

It does not make sense to quote more than 3 or 4 sentences at one occasion. If readers really have to literally read another source, students should tell them in their introduction, and say that they assume that the reader has read that source before starting reading the students dissertation.

Apart from using somebody else's text, students may also come across figures, pictures and diagrams, which they think illustrate their point better than they could do otherwise. Again, if this is the case (and students should first check that they are not acting against any copyright law), students should state that the figure/picture/diagram is taken from a particular source, and give the full details of that source in their bibliography.

For projects that utilize data (e.g. to formulate or test hypotheses), data should not be fabricated to conceal a paucity of legitimate data, nor should legitimate data be altered, enhanced or exagerrated to mislead the reader.

For more information, see the University's Code of Practice on Assessment. See also Appendix L of the Code of Practice on Assessment for definitions of plagiarism and collusion, and the penalties for those actions.

Students are expected to have read this Code of Practice on Assessment, and will complete an online Declaration on Plagiarism, Collusion and Fabrication of Data form when they electronically submit their dissertation.

Assessment

The assessment will determine to what extent learning outcomes stated above have been achieved. The assessment will be conducted by two markers. Each marker will produce separate feedback and grades (according to the COMP702 project marking descriptors below) that will be reported back to you as well as a combined mark (with a maximum of 100 points).

This mark will count for 60% of the overall final grade for the project. Failure in this task can be compensated by higher marks on the other assessments of the project.

For your guidance, a copy of the feedback form that will be used to assess your dissertation is available.

COMP702 project marking descriptors

Table 1: Project marking descriptors
GradeClassificationPercentage Qualitative Description
A*Good Distinction80+Factually almost faultless; perceptive and focused treatment of all issues. Clearly directed; logical; comprehensive coverage of topic; strong evidence of reading/research outside the material presented in the programme; substantial elements of originality and independent thought; very well written. critical and scholarly presentation.
ADistinction70-79Logical; enlightening; originality of thought or approach; good coverage of topic; clear, in-depth understanding of material; good focus; good evidence of outside reading/research; very well written and directed.
BGood Pass60-69Logical; thorough; factually sound (no serious errors); good understanding of material; evidence of outside reading/research; exercise of critical judgement; some originality of thought or approach; well written and directed.
CPass50-59Worthy effort, but undistinguished outcome. Essentially correct, but possibly missing important points or inadequate treatment. Largely derived from material delivered in the programme, but with some evidence of outside reading/research; some evidence of critical judgement; some weaknesses in expression/presentation.
DCompensatable Fail40-49Incomplete coverage of topic; evidence of poor understanding of material; Poor presentation; lack of coherent argument. Very basic approach to a narrow or misguided selection of material. Lacking in background and/or flawed in structure.
FFail< 40Serious omissions; significant errors/misconceptions; poorly directed at targets; evidence of inadequate effort. Shallow and poorly presented work showing failure in understanding.

Late submissions

The University's standard policy on lateness penalties will be applied with respect to the latest electronic submission of the dissertation. See Section 6 of the Code of Practice on Assessment for further details.

Other Penalties

  1. If the electronic submission of the dissertation document (or accompanying appendices) is not a PDF file, then 5 marks (out of 100 available for the assessment) will be subtracted for each non-PDF file submission.
  2. For each 500 words (or part thereof) over 10,000 words for the dissertation (not including appendices), 5 marks will be deducted from the final total.
  3. If the project involved software development, and the source code files are missing from the final submission, then a penalty of 10 marks will be deducted from the final total.
  4. NOTE: The use of a compression format other than ZIP poses a serious risk that your work may not be marked. If we can't decompress it, then we can't read it!

However, penalties will not reduce the mark below the pass mark for the assessment. Work assessed below the pass mark will not be further penalized for electronic submission in an incorrect format, missing program source code files, or exceeding the word limit for the dissertation.