代做COMPSCI 753: Algorithms for Massive Data Semester 2, 2024帮做Python语言
- 首页 >> C/C++编程COMPSCI 753: Algorithms for Massive Data
Semester 2, 2024
Research Project Rubrics
1 Topic Assignment
In order to prepare for your research projects, we provide the research topics together with its research paper under 4 main categories: locality-sensitive hashing, streaming algorithms, graph mining, and recommendation systems.
• You will work in a group of up to four students for the research project. Please form. a team before you select a research topic to work on.
• To register your research topic and research paper, please send an email as a group to the teaching team. Topic registration deadline is 5 pm, Friday 26 July. The email should include the UPIs and names of all group members.
• In the registration email, please provide a list of top 5 research topics along with their respective priorities. We will consider your priority and the current allocations to assign you to a suitable research topic.
• We use the first-come-first-serve principle to allocate your team to the topic. You can find if a topic has been allocated in “People” → “Project Groups”.
2 Presentation (15%)
Each group of three (or four) presenters will have 12 minutes presentation (accounts for 12% final marks) and 3 minutes Q&A (accounts for 3% final marks). We will publish the presentation schedule on the assignment page. The presentation schedule is ordered randomly. Please let us know as soon as possible if there is any clash on your presentation schedule since it is always tough to change the schedule when we are close by. We expect the presentation to focus on the seed paper by describing the applications, problems, technical solutions, and examples to illustrate the algorithms, as well as future issues.
2.1 Components
• 12 points: 12-minute presentation on the research topic.
• 3 points: 3-minute Q&A
2.2 Marking Rubrics
• Presentation (12 points)
1. Clarity (1 point): Clear voice, precise and unambiguous statements.
2. Visual (1 point): Clear font and images, accurate reference, visual aid.
3. Organization (1 point): Good motivation, logical sequence, statement of ob-jective, clear context.
4. Applications (2 points): Real-world applications, motivation to establish the importance of the problem.
5. Challenges (2 points): Computational bottleneck, hardness, how simple solu-tions fail to solve the problem efficiently.
6. Solutions (2 points): Algorithmic ideas, in-depth technical discussion of the paper, main contributions of the paper, how to solve the problem
7. Examples (2 points): Use of examples to illustrate concepts/algorithms effec-tively.
8. Future issues (1 point): Discussion of future issues, challenges, and their im-plications.
• Response to Questions (3 points): Understand questions from the audience and answer the question in a logical way.
3 Research Report (15%)
When you get the research topic together with the seed paper, you will have to branch out on your own from this paper. You are required to find at least 4 more papers based on the seed paper to complete the report. The four papers could discuss techniques for the same research problem or related techniques applied in different application domains. The report will be submitted in two stages so that we can keep track of your progress and provide feedback if necessary. The marking will be based on the final report.
• Stage 1: Report abstract and structure. Due by 11:59pm, Sunday 8 September. Note that you can change them while writing your report.
• Stage 2: Final report. Due by 11:59pm, Sunday 23 October.
For Stage 1, if you want to seek advice from the teaching team, please submit a PDF file that includes the abstract, the section titles, and a brief summary of what you would like to discuss in each section. This is, however, not mandatory.
For Stage 2, the report is expected to present a concise survey of the research topic, including a detailed discussion about the advantages and disadvantages of surveyed algorithms used in the research field. Based on the discussion, we expect to see your “new” ideas and justification for improving previous solutions.
3.1 Components
• 12 points: A literature review of at least 5 research articles on the research topic.
• 3 points: Your new ideas and justification.
3.2 Format
• 6-8 pages of IEEE proceeding format including references.
• Review at least 5 full, consistent electronic references from peer-reviewed sources.
3.3 Marking Rubrics
• Problem identification and motivation (1 point).
• Understanding of the literature in-depth or in-breadth, clearly describe how the re-viewed method works (4 points).
• Discussion about the advantages and disadvantages of the 5 papers reviewed (4 points).
• Use of effective examples to demonstrate the techniques (1 point).
• Future issues and implications of reviewed methods (1 point)
• Presentation & Organisation: well-structured report with appropriate organization of reviewed methods (1 point).
• Your new ideas or justification to improve the performance or to convince the efficiency of reviewed solutions (3 points).
3.4 How to prepare a report?
You might need to consider the following for a good report:
• Motivate the research topic by its importance to academic research or industrial do-mains.
• Clarify its computational/statistical challenges in big data.
• State the objectives or hypothesis of the topic.
• Understand in-depth or in-breath literature review, including a completeness of review state-of-the-art solutions published in recent years.
• Organize the reviews into an appropriate themes and present advantages and disad-vantages of each solution.
• Use some examples to illustrate the concepts and results.
• Show comparison of research papers against the current state of knowledge.
• Discuss research direction and recent challenges of your research topic.
• Present your new ideas or justification to improve the reviewed solutions.
Your report might have the structure as follows. Note that the main components will be the introduction and related work for the literature review and the body for your new ideas.
• Your Name, UPI
• Report Title
• Abstract
• Introduction (introduce background and research problems)
• Literature review (discuss and compare the reviewed papers including the seed paper)
• Your new ideas and justifications
• Future Work (if appropriate)
• Conclusions
• References
You might want to see https://cs.stanford.edu/people/widom/paper-writing.html for extremely useful tips in producing a report of high quality.
3.5 How to find good research articles?
Try to find best survey papers and high-quality research papers on top-tier conferences and journals. Below are some ranking systems used in Computer Science.
• http://gii-grin-scie-rating.scie.es/ratingSearch.jsf
• http://www.guide2research.com/topconf/
• http://www.conferenceranks.com/