EDUCATION
DATA3406
Final Revision Part I
TUTOR: Aaron
关注我们获取更多资讯
ABOUT US
我们把教育质量永远摆在核心
高质量的课程是我们永恒的追求
感谢来自同学的耐心与包容
如果同学对我们的课程不满意
请及时向我们的小助手反映
我们一定会在第一时间进行修正

关于我
关注我们获取更多资讯
ABOUT ME
Aaron
• Bachelor of Adv Computing 大四
• Data Science and Computer Science major
• 2021 S2 DATA3406 HD (96)
• 2021 S1 DATA3888 HD (91)
• 创业公司实习经历
• 华为等大厂offer
• 目前在做Deep learning + Computer Vision相
关的research
DATA3406
全学期
Topics梳理
尽心做好教育，为学子保驾护航！
Week 1 Intro
❖ Topics
❖ HILDA
❖ What is it? -- Data analysis with consideration for people
❖ Why does it matter? -- It's the people who have the power to collect, interpret, make us
of the data. Problems with people leads to misinformation, breach of privacy, and wrong
interpretation. The negative effects propagate in the real world as biased results are
applied.
❖ Learning Models: BLOOM, SOLO
尽心做好教育，为学子保驾护航！
Week 2-3 Ethics
❖ Topics
❖ Ethical Frameworks: NHMRC, Big Data, AI, ACM
❖ FATE*: Fair, Accountable, Transparent, Ethical, Explainable
❖ Mearning of the terms?
❖ Why they are important?
❖ Ethics in Data Collection (Week 3)
❖ What are the potential problems with raw data? -- Collected data might be biased and do not represent the full population due
to coverage, sampling and response errors
❖ What are the qualities of a questionaire should possess? (wk3. p30)
❖ How do we protect the privacy of data providers? What are GDPR principles for personal data? (wk3. p65)

Week 3 Uncertainty
❖ Topics
❖ How certain you can be about the uncertainty of a measure?
❖ Stating the accuracy of raw data, then propagating uncertainty
❖ Communicate uncertainty
❖ Uncertainty related to measurement

Week 4 People--DA Team
❖ Topics
❖ How to ensure successful teamwork?
❖ The BIG-5: Leadership, mutual performance monitoring, backup behaviour, adaptability,
team orientation, shared mental models and mutual trust
❖ What are the potential difficulties for teamwork?
❖ Cognitive biases: availability bias, over-confidence, stereotyping, confirmation bias,
social desirability bias
❖ What are the problems cognitive biases help to solve -- (Wk6. p.55)

Week 5 Literate Programming
❖ Topics
❖ Literate programming:
❖ What is it? (Wk5, p20)
❖ How to do it?
❖ Tell a story, provide motivation
❖ Document the process, not just the results
❖ ...
❖ Why do it?

Week 6-7 Data Engineering
❖ Data Engineering:
❖ Why is it important? (Wk 6. p.15)
❖ What are EDA and CDA?
❖ How to do data engineering? -- Data Engineering Methods:
❖ Data Wrangling (Wk6. p. 45)
❖ Immutability of raw data
❖ Centrality measures
❖ Deal with missing data and outliers
❖ Initial Visualizations
❖ Data tidying

Week 8-9 Visual Analytics/Report
❖ What are the cognitive factors to consider for designing visual information?
❖ Mental models, Attention, Memory Limits, Executive Function, Cognitive Biases
❖ What is Hick's Law?
❖ Colour and Colour blindness: Why are colours useful? How to deal with colour blindness?
❖ Visualization in more formal term:
❖ What is visualization? Why visualization?
❖ What is a pre-attentive? -- Colour, Form, Movement, Spatial Positioning
❖ What are Gesalt Laws? (Wk9 p.41-p.48)
❖ What is Shneiderman’s Information Seeking mantra? (Wk9 p.51)
❖ What is Tuftes' 6 principles? (Wk11 p.15)
❖ How does John Stasko evaluatie visualizations? (Wk11, p10.)
❖ How to maximize data-ink ratio?

Week 10-11 People - End-user
❖ What are the key similarities and key differences between analysts and end-users? What are the
implications? (Wk10 p.20-22)
❖ What are key challenges for end-user communications? -- Analysts are not users. (Wk10 p.23)
❖ How to deal with them? Communicate uncertaintiy, think-alound
❖ Think aloud: What is it? Why use it? How to do it?
❖ Cognitive Walkthrough: What is it? Why use it? How to do it?
❖ What are some strategies to help end-user communication? (Wk11 p.125)

对于内容的复习
1. 跟随以上问题，浏览所有lecture slides， 记录问题答案
2. 有一些lecture slides中没有答案的问题，根据知识点进行思考，尝试回答
3. 在这个过程中查漏补缺，覆盖所有知识点
4. 除了lecture slides以外，也回看OLM表格，按照表格内容对照到知识点，帮助查漏补缺
5. 完成一份复习笔记，当作考试中的cheat sheet

微信扫码添加小助手
学科专享群各类福利资料、答疑、课程优惠
小助手微信号：xxxxx