Study block A · reference semester 3
- Statistical inference
- Databases and data cleaning
← Programme and all study years
Bachelor of Science (BSc) — Data Science — Academic Support
Year 2
Semesters 3–4 (reference sequence)
Qualification: 3-year bachelor’s route / 4-year honours or research route, where offered. Current purchased LMS term: 12 months. This study section does not sell a new degree or alter an existing learner’s paid expiry.
What is available here: A year-specific reference study map and assignment, plus a shared original foundation workshop with worked reasoning and answered checks. The workshop is supplementary and may recur across related years. The live inventory below identifies existing LMS readings separately. This is not a claim that every textbook, video or semester subject has been completed or reviewed by the degree-awarding university.
Learning outcome and subject plan
Explain and apply statistical inference and databases and data cleaning, then demonstrate progress through a referenced assignment and a corrected practice portfolio.
Study block B · reference semester 4
- Algorithms
- Regression and visualisation
Confirm the exact university, award, admission batch, major/minor combination and current syllabus. A fourth-year route is conditional on the university's regulations and eligibility; it is not automatically included in every three-year award.
Original foundation reading and worked example
Computing workshop — specify, test and explain an algorithm
A program is more than code that runs once. State its inputs, output, assumptions and error conditions, then choose a representation and test cases. Correctness and efficiency are different properties. A faster result is not useful when the specification is wrong.
Practice situation
A fictional reading application receives lesson identifiers [12, 7, 12, 9, 7]. It must return each identifier once while preserving first appearance.
Worked reasoning
- Write the expected result before coding: [12, 7, 9]. Sorting to [7, 9, 12] would remove duplicates but violate the order requirement.
- Maintain an initially empty set of seen identifiers and an empty output list. Read each identifier in order. Add it to the output only when it is absent from the set, then mark it seen.
- Test empty input, one item, all-identical input, already unique input and the supplied mixed example. State the permitted identifier type and how invalid input is handled.
- With a typical hash-set implementation, expected processing is linear in the input length and extra storage is proportional to distinct identifiers. This is an implementation-dependent expectation, not an unconditional worst-case guarantee.
- Extend the exercise into a database or web application only after defining identity and permissions. Never use a demonstration dataset containing private student details, secrets or production credentials.
Five answered self-checks
Write your answer before opening the explanation.
1. What is the required output?
[12, 7, 9].
2. Why is sorting wrong here?
It changes first-appearance order.
3. What are the two working data structures?
A membership set and an ordered output list.
4. Name an important edge case.
Empty input or an input containing only duplicate identifiers.
5. Is a successful single test a proof?
No. Tests provide evidence; reasoning about the specification and all relevant cases remains necessary.
Your year 2 assignment
Implement and test the algorithm in your course language. For later years, compare a database solution, analyse complexity, document API behaviour and explain how you would protect learner records.
Apply the method to Statistical inference and compare it with Databases and data cleaning. Submit a source list, one worked application, your first answer, a corrected answer and a short explanation of the correction. For data science, keep the chosen topic, data and professional boundaries explicit.
Year-specific focus: Statistical inference; Databases and data cleaning; Algorithms; Regression and visualisation. Use your actual prescribed syllabus to choose the relevant paper. A completed foundation workshop alone does not complete these subjects.
Suggested study-session schedule
A repeatable self-study routine, not promised live classes: one session for reading and recall; one for explanation and a worked example; one for independent application; one for correction and cumulative review. Match your university’s semester calendar and available study time. No faculty date, live class or assessed laboratory has been invented.
How the study sessions work
Self-paced reading, tutor-guided explanation and independent practice. These are suggested learning sessions, not a published live faculty timetable. Teacher-led feedback is available only when separately confirmed for your programme.
- Read and recall: Read the selected unit, close the text, and state its main idea and two supporting reasons in your own words.
- Tutor-guided explanation: Ask Course Tutor to explain one specific difficulty and show a worked example. Check citations, calculations and legal/clinical statements against the primary source.
- Apply independently: Attempt a new problem, case analysis, code task or evidence-based essay before opening the model approach.
- Check and correct: Compare your result with the marking criteria. Keep the original answer, correction, reason for the error and one transfer question.
- Review and assess: Revisit the unit in the next study session and at the end of the study block. Combine short recall, application and a cumulative assignment.
Assessment and feedback
Use a short diagnostic, unit practice, a study-block assignment and a cumulative review. Each academic year page includes a worked foundation workshop and answered self-checks; internship sections instead explain institutional requirements. Course Tutor feedback is AI-generated and needs verification; no human marking service is implied.
- Concept accuracy
- 30%
- Reasoning and application
- 30%
- Evidence, sources and method
- 20%
- Clarity and professional presentation
- 10%
- Correction and reflection
- 10%
This is ZELVU’s suggested self-assessment rubric, not the university’s official marking scheme. University examinations, laboratory or field assessment and credit awards remain institutional.
Available protected course reading
18 published reading records in this course; 3 explicitly assigned to this year view. A record count does not establish complete syllabus coverage. Unassigned and legacy resources remain shared course reading.
Assigned year 2 reading
- Statistical Inference: Sampling, Confidence Intervals, Hypothesis Tests and Responsible Conclusions
- Algorithms and Data Structures: Correctness, Searching, Sorting, Graphs and Dynamic Programming
- Database Systems: Relational Design, SQL Queries, Constraints and Concurrent Transactions
13 shared / legacy reading records — not a complete year syllabus
- Course roadmap: the data science lifecycle, types of data and how the BSc syllabus fits togetherStart Here — Learning Roadmap
- Descriptive statistics: measures of centre, spread, position and shapeCore Learning — Part 1
- Probability for data science: rules, conditional probability, independence and Bayes' theoremCore Learning — Part 1
- Linear algebra for data science: vectors, norms, dot products, matrices, inverses and eigenvaluesCore Learning — Part 1
- Data wrangling in Python with pandas: inspecting, cleaning, handling missing values, joining and groupingCore Learning — Part 2
- Simple linear regression: least squares, correlation, R-squared and residual analysisCore Learning — Part 2
- Classification and model evaluation: logistic regression, the confusion matrix, precision, recall and cross-validationCore Learning — Part 2
- Quick Revision: SQL and relational databases — keys, joins, GROUP BY, NULLs and normalisationStudy Pack — Revision, Glossary & Practice
- Key Terms Glossary: Unsupervised Learning — Distance, k-Means, Hierarchical Clustering, DBSCAN and PCAStudy Pack — Revision, Glossary & Practice
- Practice Set: Hypothesis Testing — t-test, Two-Proportion z-test for A/B Tests and Chi-square Test of IndependenceStudy Pack — Revision, Glossary & Practice
- Final Revision: Data Visualisation Principles, Responsible Data Use and Personal Data ProtectionStudy Pack — Revision, Glossary & Practice
- Practice Lab 01 — Retrieval, Application & Error AnalysisPractice & Mastery Lab
- Practice Lab 02 — Mixed Assessment & Worked-Solution MethodPractice & Mastery Lab
Source and scope check
This is supplementary academic learning support, not admission to or award of a university degree. The listed sources support duration or reference structure, not endorsement of ZELVU’s material. Institutional paper order, assessment and academic credits remain separate.