CAN404 SOCIAL NETWORK ANALYSIS 2022/23 SEMESTER 2
Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit
CAN404
2022/23 SEMESTER 2
Individual Project (Final)
SOCIAL NETWORK ANALYSIS
PROGRAMMES
M.Sc. Social Computing
M.Sc. Applied Informatics
M.Sc. Business Analytics
INSTRUCTIONS TO CANDIDATES
• This project accounts for 70% of the final marks.
• Full mark is 100.
1. Learning Outcomes
On successful completion of the project, students should be able to:
1. demonstrate a critical and broad understanding of Social Network Analysis (SNA) with a series of well-defined concepts, models, algorithms, and applications.
2. explain the strengths and weaknesses of different social network models and algorithms.
3. adapt or combine some of the key elements of existing SNA models and mechanisms to design SNA solution to a real-world application problem.
4. devise a social network computer program for the real-world application.
5. implement and/or develop key algorithms of SNA.
The assessment is designed according to the learning outcomes stated above.
2. Report Submission Deadlines
This is an individual project. The project accounts for 70% of the final marks, conforming to the following requirements:
• Ensure that the topic and dataset for your final submitted report are those that have been approved as per Section 6 below otherwise 30% will be deducted from the final marks.
• Submit a soft copy of your report with cover page (in PDF format only) to Learning Mall before 11:59PM China/Beijing time on May 15, 2022 (Monday). Use the format CAN404-StudentID.pdf to name your file. For example, CAN404-2018181.pdf.
• Also submit all code (the R code that you used for the analysis and the generation of the plots), charts and graphs used in your report in their original size to Learning Mall before 11:59PM China/Beijing time on May 15, 2022 (Monday). Zipped those charts and graphs then rename the zipped file according to the format specified above. For example, CAN404- 2018181.zip.
• A 5-to- 10-minute project interview may be conducted to check on the originality of your work. The interview schedule will be released at a later date. Each of you need to prepare a 6- 10 presentation slides for the interview. The file name of your PowerPoint presentation needs to follow the same naming convention, CAN404- StudentID.ppt.
• SNA Problem
You are required to identify an SNA problem and propose a solution to gain a better understanding of real-world SNA applications. In order to achieve the stated learning outcomes, you must demonstrate your understanding and your ability to apply social network models and algorithms using the R programming language. Your chosen topic should have supporting data available. Below are some recommendations to help you identify an appropriate SNA problem:
1. Your selected dataset should have at least 100 nodes. Should your selected dataset be significantly less than 100 nodes, you are required to conduct more detailed analysis on the nodes’ attributes.
2. You may identify a problem at your workplace and propose a solution using SNA methods on data that you are able to obtain. Please ensure that your data are suitable for education and research purposes. If not, you may modify them as appropriate.
3. You may find one or several open datasets, and compare different SNA methods for analysing them.
4. Many journal articles demonstrate state-of-the-art techniques for the implementation of SNA solutions to real-world problems. You may choose a published journal paper, summarize the details, reproduce the results, and conduct a detailed critique.
3. General Guidelines
Your project must address the following requirements to gain higher credits:
1. Explain clearly why you chose the topic and how it is related to SNA.
2. State clearly at least two research questions you plan to answer with your project. 3. Apply at least 4 of the following methods that you have learned:
• Compositional and Structural Analysis, e.g. attributes and centralities
• Community Detection, e.g. MDS, CPM, Spectral Clustering, k-means
• Link Analysis, e.g. PageRank or HITS
• Proximity Measures, e.g. SNN
• Graph Cluster Analysis, e.g. MST, HCS, etc.
Besides the above methods, you are also encouraged to use other approaches or variations of the above methods that were not covered in your lectures.
4. Conduct comprehensive literature review, i.e. identifying the strengths and weaknesses of your methods and other SNA approaches in the context of your problem.
5. Identify the novelty of your project (if any).
6. List any important assumptions and/or limitations.
7. Analyse and discuss the results in the context of the research questions identified in Step 2 above. Ensure that you have properly answered your research questions.
8. Compare the results of your project with similar SNA research. What are the limitations and how would you improve in future work?
9. How is your project able to contribute to the related fields in theoretical issues and application domains?
4. Software Tools
You are expected to use the R programming language for your report, which has been covered in our online programming tutorials and workshops. Should you opt to use the other software applications, please address the reasons and motivation for your decision.
• IDEs (such as RStudio) for R language, and related packages such as iGraph, and SNA (required)
• Gephi (optional)
• Python (optional)
5. First Thing First – Proposing A Topic
The first thing you should do is to submit a project topic and dataset for approval (click here)before starting any work. Propose the project topic and dataset as early as possible (within one week of the project questions being released). The topic and dataset selection are on a “first-come, first-served” basis. You are not allowed to choose the same topic and dataset as other students on this module.
6. Report Format
In general, your report must be in English and should include the following contents:
1. XJTLU Cover Page
2. Introduction and Project Aim
• Overview of the problem of the application domain
• Introduction and problem definition
• Key approaches and models to address the problem
3. Literature Review and Proposed Methods
• Your approaches of related methods and tools
4. Implementation and Application Demonstration
5. Analysis of Results and Discussion
6. Conclusion
• Summary of results and future work
7. References
8. Appendices (optional)
The report should be formatted according to the IEEE conference template, which can be downloaded from Learning Mall. Use Times New Roman font-size 10 (as per template) for the main body of your report. The length of the report should be 6- 12 pages (excluding the Cover Page and Appendices). Appendices are optional and should not be more than 6 pages. Your programing code and your screenshots captured from the software application(s) maybe included in the Appendices, whilst the most essential codes and graphs should be included in the report body. This part is essential to assess your expertise of using software tool(s).
If you make use of any work from other sources (such as a dataset or alternative approach), the original work MUST be cited.
7. Policies on Academic Integrity and Late Submission
Please refer to your Student Handbook for policies on academic integrity and late submission.
2023-07-26