Module Code: COMP 2121 Module Title: Data Mining UNIVERSITY OF LEEDS School of Computing August RESIT 2020/2021 Assessment information: There is 1 page to this assessment specification. There will be 2 weeks to complete this assessment. The number in brackets [ ] indicates the marks available for each part. You are reminded of the need for clear presentation in your answers. The total number of marks for this assessment is 30. You are allowed to use annotated materials You can use Word or another word processor to write your answers into a document, then when you have finished, Save as PDF and submit the PDF file via Minerva. Task: Use SketchEngine to find English-language words and multi-word expressions which are specialised terms in the field of Data Mining. First, use SketchEngine and WebBootCat to collect a corpus data-set of English-language texts on the topic of Data Mining. The corpus should be 30,000 to 100,000 words in size. Use SketchEngine to identify words and Multi- Word Expressions which are mo
re frequent in your corpus than in general English. Finally, assess a sample of the results, using your knowledge of Data Mining to judge whether the words and Multi-Word Expressions are really Data Mining technical terminology, or else are relatively frequent in your corpus for other reasons. Write a report of up to 4 pages, explaining your methods and results. The report must cover: 1. Statement of objectives, and data mining problem definition [0-2 marks] 2. Data collection methods, to include experiments you tried before arriving at the final version; what seed terms you used and how you chose these; other parameter settings in WebBootCat. [0-10 marks] 3. Corpus analysis methods, to include SkE tools and resources you used, results you extracted from your corpus, with example screenshots [0-10 marks] 4. Evaluate 8 of the specialised terms predicted by SkE; use your knowledge of the field of Data Mining to assess their validity [0-8 marks] [Total: up to 30 marks] Page 1 of 1 End Module Code: COMP 2121