AIDI 1010 –
Introduction to Emerging Technologies
WEEK3(B)
Jahanzeb Abbas (JB)
Week Objectives
(A) Research Emerging Technologies
Stats, Why, How, Where
(B) Review Group Assignment 1 (20%)
Details
Rubric
(C) Review Group Assignment 2 (20%)
Schedule
Details
Rubric
(D) Review Group Assignment 3 (30%)
Details
Rubric
2
(A) Research Emerging Tech (Stats)
Technology is evolving and updating at an exponential rate*
“By 2025, 38.6 billion smart devices will be collecting, analyzing, and sharing data.
The web hosting services market is to reach $77.8 billion in 2025.
70% of all tech spending is expected to go for cloud solutions.
There are 1.35 million tech startups.
Global AI market is expected to reach $89.8 billion.
There are 4,383 million internet users.
Solar energy adoption has grown by around 50%”
A.I/Machine Learning Stats*
“AI augmentation will produce $2.9 trillion in business value
AI in retail is projected to hit $4.3 billion by 2024
The wearable AI market will reach $180 billion in 2025
The global virtual assistant market will be worth $11.3 billion in 2024
Artificial intelligence will generate 2.3 million jobs in the US”
3
(A) Research Emerging Tech (Why)
Technology is evolving fast; therefore, we need a lot more folks who can
research properly and thoroughly
Information overload requires a focused approach
Establish direction on a topic/industry that you are passionate about
Prepare you to have an “educated” opinion rather than just an opinion
Practice researching the norm through an educated approach
Cultivate a balanced, yet critical, view of your topic/industry
4
(A) Research Emerging Tech (How)
Initially, avoid tutorials; understand what you are trying to solve
Approach
Formulate a problem statement first (next slide has details)
Based on this problem statement – you prepare a direction hat helps you research and resolve the problem
Brainstorm a couple of solutions or approaches that could resolve your problem
Formulate a hypothesis, and prove/confirm through research; find (peer reviewed) research papers on the same
topic (or videos) showing how others have attempted to resolve similar problems (successfully)
Share your ideas (through presentations) with your peers and embrace constructive criticism;
Always give references for sources of information/ideas
Bonus:
Read 20-30 research papers every week (more, if you are pursuing PhD) to stay up to date within the field
5
(A) Problem Statements (Components)
A detailed problem statement is usually made up of the following
components:
The Issue/Context
Actual problem, situation
How does it happen, why does it happen
The Relevance
Purpose, vision, and scope of the problem
The Objective/Solution/Proposed Method
Ideal solution(s) of the problem and objective of the solution
Can be a claim or proven hypothesis via research
Can be multiple solutions
6
(A) Problem Statements (Examples)
Example1
After exercising, the human body is depleted of hydration and electrolytes
from sweating (Context). The person then needs to decide how best to
rehydrate the body and replenish the lost electrolytes (Issue). Not all
beverages are equally beneficial after exercise. The coach must decide what
drink to offer volleyball players during a game or practice (Relevance). In this
experiment, we will investigate the electrolyte content found in standard
sports drink compared to what is found in plain coconut water (Objective).
Example2
Remote workers across the company should be able to communicate with More Examples:
https://
one another seamlessly and effortlessly, without getting bogged down in examples.yourdictionary.com/proble
unnecessary or irrelevant messages (Vision). Right now, messages tend to get m-statement-examples.html
lost or cluttered through multiple email strings. This results not only in lost
productivity, but also multiple problems in miscommunication (Issue
statement). I propose that all employees sign up for Slack and use it for the
majority of in-company communication. Conversations can be easily searched
and organized by channel. Only more formal communication, like official
announcements, should be sent over email. (Proposed method)
7
(A) Research Emerging Tech (Where)
Links to find Research Papers*
Link1; ArXiv Sanity; what’s ArXiv Sanity?
Reference to https://arxiv.org/
arXiv is a free distribution service and an open-access archive for 1,892,558 scholarly articles in the fields of physics,
mathematics, computer science, quantitative biology, quantitative finance, statistics, electrical engineering and systems
science, and economics. Materials on this site are not peer-reviewed by arXiv.
Link2; Reddit/Machine-Learning
Links on how to read Research Papers*
Link3; how-to
Link4; how-to
Link5; how-to
Link6; how-to
Link7; how-to
8
Evaluation Details
• Group Assignments 70%
• 20% – Assignment 1 (“Group’s Problem Statement”; initial view)
• Details to be shared WEEK 3; Due on WEEK 5
• 20% – Assignment 2 (“Group’s Prototype Presentation”; sharing project status so far”)
• Details to be shared WEEK3; Due weekly (per group) starting from WEEK9
• 30% – Assignment 3 (“Group Project”, implementation details)
• Details to be shared WEEK3; Due on WEEK 14
• Test 30%
• 15% – Mid-Term Exam (Multiple Choice, True/False, Multi-Answers)
• Evaluation held on WEEK 7 (all online, multiple-choice, true/false); WEEK1-WEEK7
• 15% – End-Term Exam (Multiple Choice, True/False, Multi-Answers)
• Evaluation held on WEEK 13 (all online, multiple-choice, true/false); WEEK8-WEEK13
Groups (Members)
Groups
# Group-Members Group-Industry
1 -- --
2 -- --
3 -- --
4 -- --
5 -- --
6 -- --
7 -- --
8 -- --
9 -- --
10
(B) Group Assignment 1 (Details)
20% of your grade
Group Assignment Description:
Choose any preferred industry of your choice
Create or acquire a real-life scenario (for a company or industry or situation)
Define a detailed problem statement and a hypothesis of solution for your selected problem
Review and use a minimum of three peer-reviewed research papers that support your current
hypothesis/solution idea (this can change in the future) and compare the approaches they took in
order to achieve their goal and how you can do the same with a similar ML solution
Based on your problem statement - formulate a solution that requires machine learning
predictions (it can be any ML module or approach you desire at this initiation phase, which you
have learned in this program at Georgian)
Show references used for ideas
Submit your work in .ipynb or .py file on Blackboard
Google Colab has an option to add ‘text’ in a ‘text cell’, please do not submit a word document
Group Assignment 1 is due WEEK5; End of Day (or 24hrs later)
11
(B) Group Assignment 1 (Rubric)
Group Assignment 1 - Rubric: • Above and beyond; 5 marks; optional, exceeding total marks
• Python code; Commenting in the code; 1.5 marks
• Python code; ML algorithm approach; testing/training and results? 1.5 marks
Total marks: 35 (+5 optional marks) • Appreciating effort, 2 bonus marks
(Definition) Problem statement; 10 marks, mandatory
How detailed is your problem statement? Does it cover all components of a “regular” problem statement? Does it have a focused scope?
Break down your problem statement into different chunks and explained thoroughly.
(Research) Machine Learning Prediction & Outcomes; 10 marks, mandatory
Did your research papers’ predictions match with expectation? Which algorithm did they use? Why? What were the results? How did the
research papers inspire you to craft your hypothesis?
(Hypothesis/Proposal) Machine Learning Solution; 10 marks, mandatory
How different or similar is your approach as compared to your research papers? What algorithm are you proposing to use? Why? What
would be the results?
(Citation) References; 5 marks, mandatory
Are all the links provided?
12
(C) Group Assignment 2 (Schedule)
Group Presentation Schedule
A schedule of your presentations will
be posted on Blackboard, please
check “Course Information” for
more details.
13
(C) Group Assignment 2 (Details)
20% of your grade
Group’s Prototype Presentaton - Description/Tasks
Solution Update
Update the approach/solution from Group Assignment 1 and make sure to use an AutoML module that will be assigned by professor
Present your problem statement again, findings/details (thus far) as per defined schedule
This is a presentation assignment, with an allotted time of 15mins. The main objective is to prepare a virtual presentation
(recorded video, which you can capture through ZOOM or TEAMS) and share your AutoML prototype’s update. Your prototype
must be based on the assigned AutoML module from the professor, including ways on how it can be improved (as a team); How
are you planning to visualize the results upon completion? You must also share what’s coming up in your final project; i.e: if there
are any enhancements or corrections being worked on between group assignment 2 and group project
Learn from Group Assignment 1 feedback
You must re-visit your problem statement for improvements and finalize proposed solution based on ML predictions
You must describe how your solution will add value to the situation
Present your AutoML prototype solution and share with peers through a presentation
Your prototype’s data must be from reputable sources (like Kaggle or others). Your prototype must be supported with code (using
Python), which must also be presented to your peers. Your prototype doesn’t have to work or can even work partially, but you have
to present it to your peers. This is to help iteratively build you a working AutoML pipeline for your final project. You have to make
this presentation-video available in a folder location that will be in “Course Information”.
Group Assignment 2 is due starting from WEEK09
After your group’s presentation - one member of the group must upload a copy of the presentation
video (4 slides minimum) in a location that will be indicated in “Course Information” on Blackboard due
by EOD (as per schedule) of your presentation-date
14
(C) Group Assignment 2 (Rubric)
Group-Share Assignment 2 (Presentation) - Rubric:
Total marks: 50
Problem statement; 10 marks
What is your problem statement? How did you improve your problem statement based on the new AutoML module being
assigned; does it cover all components of a “regular” problem statement now?
Machine Learning Prediction & Outcomes; 15 marks
What did the research tell you? What ended up being your goal/solution?
Does your prediction match with expectation using your designated AutoML, thus far? What were research results? Do you have
results of your own yet? What are you trying to do to meet your end goal/solution?
Solution/Python Code; 10 marks
Just a prototype; doesn’t have to be fancy or commented, it just has to be presentable to the class
Need a ML code example; testing/training with preliminary results?
Overall Presentation; 10 marks
As group, are you presenting in an effective way that is promoting discussion; any good visuals? Did each
member present in an engaging way? Must be a done in order to attain full marks
Documentation & References; 5 marks
What were some links or videos or articles online that support your idea or inspired you with the solution?
15
(D) Group Assignment 3 (Details)
30% of your grade
Group-Share, Project Description
Your main task is to finish prototype solution that you previously shared with the class
Your solution must be visually presentable (using any Python module or technique)
Your solution must be similar to three other peer-reviewed research papers
Your solution must be in code (using Python, and as a Google Colab Notebook)
Your Google Colab Notebook must have the following header, explaining in simple text:
Description of the problem statement & solution
Demonstrate the issues you faced and how you resolved them
Describe what approaches you have taken since prototype and which hyperparameters you adjusted to achieve your AutoML-based solution
How would this solution add value to your real-world situation?
Your solution must have a trailer, explaining in simple text:
What you learned; how else would you improve this code; final remarks from each member
Group Assignment 3 Project is due: WEEK14
One member of the group must upload a copy of the deliverables (including references) in .ipynb format on BlackBoard
This is not a presentation
16
(D) Group Assignment 3 (Rubric)
Group Assignment 3 - Rubric: • Above and beyond; 5 marks; optional, exceeding total marks
• Highlight the features in header and comment in code (GitHub, or integration
Total marks: 60 (+5 optional marks) •
of another module?)
Additional features, checks/balances that are pointed out by the group
Header Documentation; 5 marks, mandatory
Explain your project, the strategy, the intended audience, the challenges, and the instructions to construct the pipeline (if any)
Problem statement review; 10 marks, mandatory
How detailed is the problem statement; does it cover all components of a “regular” problem statement; do you have an outlined
solution? You can repeat the same problem statement from Assignment 1/2 if no marks were deducted previously
Python Code; 15 marks, mandatory
Code has to be working and must meet standards; it must be structured (distributed into sections) and commented/documented
Your code must properly outline details like: what data are you using, are you cleaning your data (if yes, how), where you are
fitting the data into the model, how are you creating test/training sets, where are you formulating results?
Machine Learning Prediction & Outcomes; 10 marks, mandatory
Did your prediction match with expectation using your designated AutoML? What were the results? This can be a combination of
text or commenting in the code. How does your work match with the research papers that you have selected? How did this fair
against the prototype you initially created in Assignment 2?
Visualization & Documentation; 10 marks, mandatory
How did you visualize the results? Are you comparing Assignment 2 attempts, research attempts, and your project attempts?
Can an executive review this and make sense of the results? How user-friendly is the output? What were some links or videos or
articles online that support your idea; the research papers selected for ideas and comparison
Trailer Documentation; 10 marks, mandatory
What did you learn, how else would you improve the code or visualization? Final remarks from each group member
17
Disclaimer
Due to nature of the course, various materials have been compiled from different open source
resources with some moderation.
The course designer (slides creator), sincerely acknowledges their hard work and contribution,
credit will be given wherever necessary
18
References
1. https://web.mit.edu/course/21/21.guide/prob-sta.htm#:~:text=Problem%20statements%20often%20have%20three,document%20the%20writer%20is%20preparing
2. https://hostingtribunal.com/blog/how-fast-is-technology-growing/#:~:text=Supposedly%2C%20every%2018%20or%20so,is%20known%20as%20Moore's%20law.&
text=The%20number%20of%20transistors%20that,over%2010%20billion%20in%202017
3. https://www.elsevier.com/connect/infographic-how-to-read-a-scientific-paper
4. https://leftronic.com/blog/how-fast-is-technology-growing-statistics/
5. https://www.indeed.com/career-advice/career-development/what-is-a-problem-statement
6. https://examples.yourdictionary.com/problem-statement-examples.html
7. https://www.library.georgetown.edu/tutorials/research-guides/15-steps
8. https://www.youtube.com/channel/UC1H1NWNTG2Xi3pt85ykVSHA
9. https://www.youtube.com/channel/UCWN3xxRkmTPmbKwht9FuE5A
10. https://www.youtube.com/channel/UCPk8m_r6fkUSYmvgCBwq-sw
11. https://examples.yourdictionary.com/problem-statement-examples.html
19
Thank You Very Much
Any Questions?
20
