Discuss the current state of the art in data mining social media data.
a) the particular challenges presented by social media data;
b) how it can be harvested;
c) any issues specific to data cleaning/pre-processing for social media data;
d) any algorithms that have been adapted to mine such data;
e) any successful applications of data mining to such data;
f) how KDD should evolve to meet the challenges of social media data;
g) ethical issues in mining social media data.
Note that you do not need to cover all of those topics and you may indeed find other
related topics of interest. You could write an interesting essay using just one of those topics
or you may wish to cover a few if you find they are interlinked. Your essay should extract
information from a number of refereed publications (e.g. conference proceedings, journal
articles, etc). You should abstract information from the published articles and critically
review it, instead of merely repeating what is said on the sources you use. If you are going
to copy verbatim, which should only be done when strictly necessary (e.g. when quoting a
specific definition) then use quotations to enclose the copied text and a clear references to
say where the text comes from. Cite all relevant references, including on-line references.
Failure to do this could result in a penalty due to plagiarism and, in extreme cases,
disciplinary procedures. It should be written in a clear and professional manner, using good English.
Structure your document into sections, and do not forget to include conclusions.
Your essay is meant to be a very concise account of the current literature and should not
exceed 2000 words, including the reference section. Please include a word count with your
submission. It is about providing a succinct account of the current literature focused
specifically on Data Mining approaches and their application to social media data.
Write your essay in 12-point standard font such as Times or Helvetica, using a standard
page layout in terms of margins, etc.
– Include your word count.
– Present your references in a professional style suitable for thesis publications. You can
choose either of these two styles:
Author-year citations: References within the text may be cited by using the author’s
(or authors’) last name(s) and the year of publication (and a page number only for
direct quotations). These references refer to works cited in an alphabetically
arranged bibliography.
Assessment criteria
Marks will be awarded for:
– good literature survey including sufficient references to published peer reviewed work (i.e.
conference/journal articles, books, …) and good use of the WWW/library to locate resources
(30 marks);
– demonstration of capacity to read and abstract information from sources (20 marks);
– concentrating on relevant issues and giving a satisfactory answer to the question with
good coverage/quality of answers but in a concise manner (30 marks)
– good writing (good English, good document structure) and presentation style (10 marks)
– complete and well formatted reference section (10 marks)
Numbered bibliography: When a numbered bibliography is used, each entry in the
bibliography is numbered consecutively. These numbers are inserted in the text to
cite the references. The numbers in the text are enclosed in parentheses or brackets
or printed as superscripts; they will not appear in consecutive order in the text.
Last Completed Projects
| topic title | academic level | Writer | delivered |
|---|