MM1041 Introduction to Artificial Intelligence and Data Analytics in Business
The Complete Exam Bible · Semester 1 2026/27MM1041 Overview
- The Hong Kong Polytechnic University
- Semester 1, 2026/27
- Level 1 subject
- Information Technology
- Closed-book final examination
What this course is asking you to do
This is a first-year course from the Department of Management and Marketing that introduces artificial intelligence and data analytics, abbreviated AIDA, as tools for business decisions.
- Assessed by A closed-book final examination, an individual AI case study, a group project, and a participation block of attendance, an online test and Python exercises.
- The skill being tested Applying an AIDA concept to a real business: name it, place it in the case, state the gain, and name a risk with a safeguard.
- Hardest step Turning analysis into recommendations that clearly follow from it, which the top band of the group project rubric asks for.
- Where to confirm Dates, submission settings and the live window for the online test are controlled by the course Canvas site.
How MM1041 is assessed
| Component | Weight | Format |
|---|---|---|
| Class Attendance (Week 1 to Week 13) | 5% | Participation component |
| Online AIDA Test (Week 6) | 5% | Participation component; 20 minutes, 20 multiple-choice questions |
| Python Exercises (Week 13) | 5% | Participation component |
| AI Case Study (Week 8) | 15% | Individual assignment on the application of AI in business |
| Group Project (Week 12) | 30% | Group assignment: 15-minute video presentation and slides |
| Final Examination | 40% | Closed-book exam graded against the five learning outcomes |
The course outline for Semester 1, 2026/27 lists these six components and they total 100%. The first-week slides summarise the same structure as participation 15%, the individual assignment 15% and the group assignment 30%, alongside the 40% exam. Two details should be checked on Canvas. The outline schedules the online AIDA test in Week 6, while the test page on Canvas showed an availability window ending on 31 August, and that page also says you should answer at least thirteen of the twenty questions correctly. The first-week slides mention submitting through Turnitin on Blackboard and describe the group assignment as a report, while the outline asks for a video presentation through OneDrive and slides through Canvas. This guide follows the outline. Late submission is not accepted without permission given in advance.
Assessment structure
Segment widths reproduce the published percentage weights and total 100%.
Current MM1041 dates
| Date | Item | Control |
|---|---|---|
| 8 October 2026 | Online AIDA Test (Week 6) | Week 6 date in the tentative schedule of the course outline; the Canvas test page showed a different availability window, so check it. |
| 22 October 2026 | AI Case Study (Week 8) | Week 8 date in the tentative schedule; confirm the submission deadline on Canvas. |
| 19 November 2026 | Group Project (Week 12) | Week 12 date in the tentative schedule, when the video and slides are due. |
| 26 November 2026 | Python Exercises (Week 13) | Week 13 date in the tentative schedule, the second Python session. |
Dates are as published in the tentative schedule in the course outline for Semester 1, 2026/27. Confirm exact deadlines and submission settings in the live LMS.
What MM1041 covers
Eleven chapters follow the course from data, analytics and big data through machine learning, computer vision and generative AI to the risks of AI, human and machine collaboration and Python.
Business Data and the DIKW Pyramid
what business is and why AIDA matters · data, information, knowledge and wisdom · descriptive to prescriptive analyticsWhere the course begins
The first lecture starts from business itself, described as people working together in an organised way to supply the goods and services others need, usually in return for payment.
- Business as organised effort to meet needs
- Why AIDA changes business decisions
- The DIKW pyramid from data to wisdom
- Data as variables that can be computed
- Descriptive, diagnostic, predictive and prescriptive questions
- Naming the level before answering
Data Analytics: Process, Characteristics and History
collection, storage, retrieval and analysis · data driven, interdisciplinary, decision oriented · the cholera map and the censusA definition you can write from memory
The e-learning lesson treats data analytics and data analysis as the same thing and defines it through four actions: inspecting data, cleaning it, transforming it and modelling it.
- Inspecting, cleaning, transforming and modelling data
- Collection, storage, retrieval and analysis
- Data driven input, interdisciplinary method, decision output
- The cholera map and the outlier
- From a seven-year census to six weeks
- Better models, better computing, more data
Big Data and the Four Vs
what makes data big · volume, velocity, variety and veracity · structured, semi-structured and unstructured dataWhat counts as big
The big data lesson defines big data as a field that treats ways to analyse, systematically extract information from, or otherwise deal with data sets too large or complex for traditional data-processing software.
- Data sets too large or complex for traditional software
- Everyday sources of big data
- Volume and the scale of data
- Velocity of generation and of processing
- Variety across tables, text, XML and graphs
- Veracity and the trust problem
Big Data Systems, MapReduce and Applications
the big data driven approach · NoSQL, Hadoop and Spark · map, shuffle and reduce · AlphaGo, flu trends and eye scansThree eras of analytics
The big data lesson describes how the question behind business analytics moved forward in time. In the 1990s spreadsheets answered what happened. In the 2000s business intelligence dashboards answered what is happening.
- From what happened to what will happen
- Raw data, preprocessing, analytics, interpretation
- Why web firms built NoSQL stores
- Hadoop and Spark on clusters of machines
- Map, shuffle and reduce
- AlphaGo, flu trends and eye-scan diagnosis
AI, Machine Learning and Deep Learning
weak and strong AI · why firms adopt AI · learning from experience, task and performance measure · tacit knowledgeThree nested fields
Artificial intelligence, in the course's definition, is the study and building of computer systems that can do jobs which would ordinarily need a human mind, such as understanding language, reasoning, recognising speech, making decisions, navigating the visual world and manipulating objects.
- Artificial intelligence as tasks needing human intelligence
- Machine learning without explicit programming
- Deep learning and neural networks
- Weak AI against strong AI
- Cost, revenue and new business lines
- Mitchell's experience, task and performance
Three Types of Machine Learning
trained with targets, without targets and with mistakes · classification and regression · clustering · overfittingOne distinction, three labels
Both the first lecture and the e-learning lesson divide machine learning into three types, using the same labels.
- Trained with targets, without targets, with mistakes
- Classification with categorical labels
- Regression with continuous labels
- Clustering and dimension reduction
- Agent, environment, reward and penalty
- Training data against testing data
Computer Vision and Pattern Recognition
what computers see in an image · easy tasks and hard tasks · computer vision against pattern recognition · applicationsWhy vision matters
The e-learning lecture begins with people: about ninety per cent of the information sent to the brain is visual, and ninety-three per cent of human communication is said to happen through visual forms.
- How much human information is visual
- Definitions and a short history
- Pixels as intensity numbers
- Easy tasks for machines, hard tasks for people
- The comparison table
- Applications from maps to rescue drones
Generative AI and Prompt Engineering
what generative AI creates · workplace uses · prompting and iteration · declared use in the written assessmentsWhat generative AI is
In the course module, generative AI is AI that produces new material, whether text, pictures, program code or sound, by drawing on patterns it has learned from data that already exists.
- Generative AI creates new content from learned patterns
- Chatbots, prompts and natural-language replies
- Workplace uses from summaries to code
- Prompting and iteration
- Evaluating output critically
- Declared use in the written assessments
Risks, Bias and the Limits of Machine Learning
hidden bias, unverifiable output, missing data, unintended consequences and errors · safeguards · deepfakesFive risks named in the first lecture
The lecture lists five risks and limits. Hidden biases: a system that learns which job applicants to interview from past recruiters' decisions can learn to perpetuate their racial, gender or ethnic biases.
- Hidden bias from historical decisions
- Unverifiable deep learning output
- Problems without data
- Unintended consequences of machine advice
- Everyday AI errors
- Five safeguards
Collaborative Intelligence and AI in Hong Kong
what machines and people each do well · how each assists the other · flexibility, speed, scale, decisions and personalisationThe comparison behind the idea
Collaborative intelligence is the first lecture's answer to the question of whether machines will replace people. Machines contribute automation, speed, the ability to scale and quantitative power.
- Machine strengths and human strengths
- Four ways machines assist people
- Three ways people assist machines
- Flexibility, speed, scale, decision making, personalisation
- AI and Hong Kong as a technology hub
- Choosing what to give to AI
Python Programming Fundamentals
values, variables and operators · conditions and loops · lists and functions · notebooks in Colab and KaggleWhy Python carries weight here
Fundamental Python programming skills are one of the course's five learning outcomes.
- Why Python appears in the learning outcomes
- Values, types and variables
- Division, floor division and remainder
- If, elif and else
- For loops over lists
- Functions with parameters and return values
Its five learning outcomes ask you to understand the foundational concepts, gain basic skills with AIDA technologies and applications, give examples of how adopting AIDA could strengthen Hong Kong as an international innovation and technology hub, show awareness of the ethical issues and impact of AIDA on business and society, and acquire fundamental Python programming skills.
Teaching combines weekly lectures with a self-study e-learning module in the first five weeks, and the closed-book final examination is graded against the same five outcomes.
Business Data and the DIKW Pyramid
The opening chapter defines business as the organised effort to provide goods and services that meet needs, and explains why AIDA changes how that effort is directed.
It walks one dataset up the pyramid from data through information and knowledge to wisdom, then sets out the ladder of analytics questions from what happened, through why and what could happen, to how to make an outcome happen.
Data Analytics: Process, Characteristics and History
Analytics is defined by four actions on data, inspecting, cleaning, transforming and modelling, aimed at useful information and support decisions.
The chapter places it at the end of a pipeline of collection, storage and retrieval, gives its three characteristics, and tells the history through a cholera map that located a contaminated pump and a census that punched cards cut from years to weeks.
Big Data and the Four Vs
Big data is data too large or complex for traditional processing software.
The chapter compares volume, velocity, variety and veracity with the examples taught for each, shows variety inside a single email, and explains why veracity is the characteristic that makes the others dangerous.
Big Data Systems, MapReduce and Applications
This chapter follows the big data driven approach from raw data through preprocessing and analytics to interpretation, explains why large web firms built NoSQL stores and analytic platforms such as Hadoop and Spark, and works through the map, shuffle and reduce steps in words.
It closes with applications including AlphaGo, flu prediction from search terms and diagnosis from eye scans.
AI, Machine Learning and Deep Learning
Artificial intelligence is the development of systems that perform tasks normally needing human intelligence.
The chapter separates weak from strong AI, gives the three business reasons for adopting AI, and uses Mitchell's definition of learning from experience with respect to a task and a performance measure as a method for deciding whether a product really learns.
Three Types of Machine Learning
Supervised learning is trained with targets, unsupervised learning without targets, and reinforcement learning with mistakes.
The chapter works through classification and regression, clustering and dimension reduction, the agent and environment loop, and the difference between training and testing data that exposes overfitting.
Computer Vision and Pattern Recognition
To a computer an image is a grid of intensity numbers.
The chapter explains what that means for easy and hard visual tasks, compares computer vision with the wider field of pattern recognition row by row, and reads the applications taught, from biometrics to smart cars and rescue drones, as evidence of what vision automates.
Generative AI and Prompt Engineering
Generative AI creates new text, images, code or audio from patterns in existing data.
The chapter lists the workplace uses the module names, sets out an iterative way of writing prompts, and states the course rules on declaring any use of these tools in the written assessments.
Risks, Bias and the Limits of Machine Learning
Five risks are named in the first lecture: hidden bias, lack of verifiability, lack of data, unintended consequences and errors.
The chapter pairs them with the five safeguards the lecture proposes, weighs errors that cost more than others, and brings in the deepfake frauds reported in Hong Kong.
Collaborative Intelligence and AI in Hong Kong
Machines bring automation, speed, scale and quantitative power; people bring creativity, empathy, communication and judgement.
The chapter shows how each assists the other, groups business examples under flexibility, speed, scale, decision making and personalisation, and connects them to AI in Hong Kong.
Python Programming Fundamentals
The last chapter covers the building blocks for the Python exercises and the exam: values, variables and operators, conditions and loops, lists and functions, and how to work in browser notebooks such as Google Colab and Kaggle.
How to use this guide
Work one chapter at a time and write your own answer to each practice item before reading ours.
The practice chapter mixes quick items of the online-test kind with a complete applied answer, and the final chapter is a revision pass for the morning of the exam.
Evidence and assessment control
Assessment labels and weights follow the course outline for Semester 1, 2026/27, which calls its schedule tentative.
Teaching explanations, worked examples and practice items are independently authored and are not University assessment material. Confirm dates, submission settings and any change to the assessment on the course Canvas site.
Worked example · free
Match a business question to the right AIDA tool
- 3Name the analytics level and learning type for the shortage question.
- 3Name the analytics level and learning type for the shopper-groups question.
- 2Name one risk for each, with a safeguard.
Key terms
- DIKW Pyramid
- A model of data, information, knowledge and wisdom in which every higher level asks a harder question of the data and is worth more for decisions.
- Data Analytics
- Working through data by inspecting, cleaning, transforming and modelling it, so that it yields useful information and supports decisions.
- Big Data
- Data sets too large or complex for traditional data-processing software, usually described by volume, velocity, variety and veracity.
- Veracity
- The uncertainty of data, arising from errors, inconsistency, incompleteness, ambiguity or deliberate deception.
- MapReduce
- A programming model that maps data into key-value pairs, shuffles them so each key reaches one reducer, and reduces each key to a result.
- Machine Learning
- The part of artificial intelligence in which a program improves at a task through experience instead of being explicitly programmed.
- Supervised Learning
- Machine learning trained on examples that carry targets, used for classification of categories and regression of numbers.
- Reinforcement Learning
- Machine learning in which an agent acts in an environment and adjusts its strategy from rewards and penalties.
- Overfitting
- A model that fits its training data closely but performs poorly on new data from the real situation.
- Pattern Recognition
- Identifying regularities in any kind of data, including images, text and audio, in order to classify it.
- Collaborative Intelligence
- Work designed so that machines contribute speed, scale and quantitative power while people contribute creativity, empathy and judgement.
MM1041 FAQ
Is the final exam open book?
No. The course outline lists the exam as closed book and worth forty per cent of the course mark, graded against the five learning outcomes, from foundational concepts through AIDA skills, business examples and ethical awareness to Python. Check Canvas and the examination timetable for the question format and the exam date, and prepare definitions you can write from memory together with applied examples.
How is the course assessed?
Six components total one hundred per cent: class attendance, the online AIDA test and the Python exercises at five per cent each, the individual AI case study at fifteen, the group project at thirty and the closed-book final examination at forty. The schedule in the outline is marked tentative, so confirm the week of each item on Canvas as the semester runs.
What is on the online AIDA test?
The test page describes twenty multiple-choice and true or false questions drawn from the first six lessons of the e-learning module, excluding the interactive playground, with a twenty-minute limit and the highest of several attempts counted. It also says you should answer at least thirteen correctly. The outline places the test in Week 6, so check the live window on Canvas.
Can I use ChatGPT or other AI tools on the assignments?
Yes, within the course rule. The outline allows generative AI tools in the written tasks, the individual and group assignments, provided the use is clearly declared. The first-week slides also forbid writing an assignment in another language and then using AI to translate the entire work into English, so draft in English and keep a record of how any tool was used.
What does the group project involve?
Groups of seven to eight students choose a real business problem or opportunity and apply the course concepts to make managerial recommendations, in three parts: case preparation, analysis with methods such as Python or machine learning, and suggestions. One member submits a fifteen-minute video through OneDrive and one submits the slides through Canvas in Week 12.
How is the AI case study marked?
Its rubric gives fifty per cent to questions one and two, forty per cent to questions three and four, and ten per cent to English. Top-band work meets every task requirement comprehensively, uses highly relevant, specific and detailed examples, is clearly structured with headings and paragraphs, and has accurate grammar and vocabulary.
Do I need to install Python?
No installation is needed. The course recommends Google Colab and Kaggle notebooks, which run in a web browser, and two scheduled sessions in Weeks 12 and 13 cover Python before the graded exercises. Practising in the same notebook tool before those sessions makes the exercises far less stressful.
Questions students ask in each MM1041 chapter
- What is the difference between data and information?
- Data is a single recorded quality or quantity, such as one transaction, and on its own it answers nothing. Information is data organised so that it answers a question, such as total weekend sales by branch. The shift from one to the other is work done by an analyst, which is why the pyramid treats each level as added value rather than added storage. Chapter 01 · Business Data and the DIKW Pyramid
- Is data analysis the same as data analytics?
- In this course, yes. The e-learning lesson uses the two terms as equivalents and gives one definition for both, built on four actions, inspecting, cleaning, transforming and modelling, aimed at useful information, sound conclusions and better decisions. Some textbooks draw a distinction, but an exam answer that follows the course definition is safe. Chapter 02 · Data Analytics: Process, Characteristics and History
- What are the four Vs of big data?
- Volume, velocity, variety and veracity. Volume is the scale of data, velocity is the speed of generation and required processing, variety is the range of data forms, and veracity is the uncertainty in the data. The first three come from the original description of data growth in e-commerce; veracity was added in the IBM version shown in the lesson. Chapter 03 · Big Data and the Four Vs
- Why couldn't large web companies just use a traditional database?
- Traditional relational databases are built to guarantee transactions, which costs speed. Companies handling streams of posts needed an extremely high rate of inserts and rarely needed those guarantees for such data. They therefore built NoSQL stores that trade strict consistency for throughput, and ran analysis on clusters rather than on one large server. Chapter 04 · Big Data Systems, MapReduce and Applications
- Is machine learning the same thing as artificial intelligence?
- No. Machine learning is one part of artificial intelligence. AI covers any system that performs tasks normally needing human intelligence, including systems built from hand-written rules. Machine learning is the part in which the system improves from examples instead of being explicitly programmed, and deep learning is a further subset that uses many-layered neural networks. Chapter 05 · AI, Machine Learning and Deep Learning
- What is the easiest way to tell supervised from unsupervised learning?
- Check whether the training examples carry the answer. If each example comes with a label that people supplied, such as a category or a number, the learning is supervised. If the data arrives with no labels and the model has to find structure by itself, it is unsupervised. The course summarises this as trained with targets against trained without targets. Chapter 06 · Three Types of Machine Learning
- What is the difference between computer vision and pattern recognition?
- Computer vision works only on visual input, images and video, and aims to understand features in them. Pattern recognition is broader: it identifies regularities in any data, such as speech, text, signals or images, in order to classify it. Every computer vision system recognises patterns, but a voice recogniser recognises patterns without seeing anything. Chapter 07 · Computer Vision and Pattern Recognition
- Am I allowed to use ChatGPT for the group project?
- Yes, with a condition. The course outline permits generative AI in the written tasks, including the individual and group assignments, provided the use is clearly declared. The first-week slides also forbid drafting in another language and having AI translate the whole work, so write in English and state which tools your group used and for what. Chapter 08 · Generative AI and Prompt Engineering
- Why can an AI system be biased if nobody programmed it to be?
- Because it learns from data, and data about past human decisions carries the patterns of those decisions. The lecture's example is a hiring system trained on recruiters' earlier choices, which can learn and repeat their racial, gender or ethnic biases. The bias enters through the training data, not through any explicit rule. Chapter 09 · Risks, Bias and the Limits of Machine Learning
- Will AI replace human workers?
- The course's answer is collaboration rather than replacement. Machines are strong at automated processes, speed, scale and quantitative work, while people are strong at creativity, empathy, communication, adaptability and leadership. The best-performing arrangements assign each side what it does best and design the hand-off between them. Chapter 10 · Collaborative Intelligence and AI in Hong Kong
- Do I need to install anything to program in Python for this course?
- No. The course recommends Google Colab and Kaggle notebooks, which run in a web browser with nothing to install. Sign in, create a notebook and type code into a cell. Practising in the same tool you will use for the graded exercises removes one source of stress on the day. Chapter 11 · Python Programming Fundamentals
How to study for the exam
Pick one Hong Kong business and carry it through every chapter, writing two lines per chapter on how the concept applies to it: which data it holds, which V of big data hurts it, which learning type fits its problem, which risk worries you and how its staff and systems should share the work. Then practise the four-part answer, concept, case, gain and limit, on paper under time.
Finally, open a notebook and type every Python example in the guide, because reading code and writing it are different skills.
Your AI Information Technology tutor for MM1041
Stuck on a hard MM1041 question? Sia is AskSia’s AI Information Technology tutor — ask any MM1041 Introduction to Artificial Intelligence and Data Analytics in Business question and get a clear, step-by-step explanation grounded in how the course is actually taught and assessed. Read this whole study guide free, then take your hardest questions to Sia.
