Practise vocabulary for personalisation platforms: user profile stores, feature stores, real-time vs. batch personalisation, and platform architecture.
0 / 10 completed
1 / 10
A ___ store maintains each user's preferences, interaction history, and computed attributes used for personalisation.
A user profile store is a database (Redis, DynamoDB, Cassandra) holding user state: demographics, preference vectors, interaction history, and computed attributes (last 10 items viewed, favourite genres). It's queried at inference time to personalise recommendations.
2 / 10
A ___ store is a centralised repository of pre-computed feature vectors used by multiple ML models — including recommendation models.
A feature store (Feast, Tecton, Hopsworks) stores and serves ML features with point-in-time correctness. It enables consistent feature computation between training and serving, and allows multiple models to share the same feature pipelines.
3 / 10
___ personalisation generates recommendations at request time using the user's current context — enabling session-aware recommendations.
Real-time (online) personalisation computes recommendations at serving time using the user's current session context (page just visited, items in cart). It's more relevant than batch personalisation but requires low-latency infrastructure.
4 / 10
___ personalisation pre-computes recommendations for all users in advance (e.g., nightly jobs), storing results for fast retrieval.
Batch personalisation runs recommendation scoring for all users offline (nightly, hourly) and stores pre-computed results. It's cheaper and simpler than real-time but can't incorporate the user's current session behaviour.
5 / 10
A ___ layer sits between the recommendation model and the user, applying business rules: filtering out-of-stock items, boosting sponsored content, and enforcing diversity.
A post-processing layer applies business rules that the ML model shouldn't learn: exclude out-of-stock items, boost items with high margins, ensure category diversity, and handle regulatory exclusions. This separates business logic from model optimisation.
6 / 10
David from the Recommendation Team posted this Slack message: 'Just ran a query against the UserProfileCache and it's returning significantly higher latency than usual. Looks like we're hitting the index for 'preferred_genres' really hard. Should we investigate adding an approximate nearest neighbor (ANN) index to that field?' What does David likely mean by 'UserProfileCache'?
David is referring to a caching mechanism – the UserProfileCache – which stores pre-computed user profiles for fast retrieval. This avoids repeatedly querying slower data sources like a database. Using an ANN index would improve search speed within that cached profile.
7 / 10
During a code review, Sarah comments on a PR: 'This section uses the `get_personalized_product_list` API endpoint. We should add more detailed logging to track which features are being used to drive the personalisation – this will help us understand if our current model is performing as expected and identify areas for improvement.' What does Sarah imply about the purpose of the get_personalized_product_list endpoint?
Sarah highlights the importance of logging *how* the endpoint is being used – it's not just about retrieving data. The API exposes the personalisation logic itself, allowing analysis of feature usage and model effectiveness. This contrasts with a simple transactional or database access layer.
8 / 10
You're reviewing the description for a nightly job that updates the UserPreferenceMatrix: 'This job calculates user preference vectors based on their recent activity and stores them in the UserPreferenceMatrix. The resulting matrix is then used by the recommendation engine to generate personalized product recommendations.' What role does the UserPreferenceMatrix play?
The description explicitly states that the matrix stores *pre-calculated* vectors. This is key – it's not a live database of current preferences but rather a pre-computed representation enabling fast recommendations. This contrasts with a real-time database or log file.
9 / 10
Ben, the Lead Data Scientist, asks: 'I need to implement a rule to boost all products from our 'New Arrivals' category by 20% in the recommendation feed. How should I achieve this?' What is the most appropriate approach?
Introducing business rules at a layer *between* the recommendation model and the UI is best practice. This separates concerns – the model focuses on algorithmic recommendations while the rule layer handles business logic like category boosts. Directly modifying the model or altering the preference matrix would be less maintainable.
10 / 10
You're debugging a performance issue with a recommendation system and receive this API response from the model server: `{'status': 'success', 'response_time': 0.35, 'recommendations': [{'product_id': '1234', 'score': 0.9}, {'product_id': '5678', 'score': 0.8}]}`. What does the `score` field indicate?
The `score` field represents the model's *confidence* – a higher score indicates that the model believes the recommended product is highly relevant to the user based on their profile and context. It's not inventory or price; it's a measure of predicted preference.
What will I practice in "Personalisation Platform Vocabulary"?
This is a Recommendation Systems Language exercise set. It walks through 10 scenario-based multiple-choice questions built around real usage of recommendation systems language terminology that IT professionals encounter on the job.
Is this exercise free to use?
Yes. Every exercise on CoderSlingo, including this one, is free to complete with no account, sign-up, or paywall.
How many questions are in this exercise?
This set contains 10 questions. Each one shows immediate feedback and a detailed explanation after you answer, so you learn the correct usage right away rather than waiting for a final score.
Do I need prior experience to complete this exercise?
No prior experience is required. Each question includes a full explanation covering the reasoning behind the correct answer, so the exercise itself teaches the recommendation systems language vocabulary as you go.
Can I retry the exercise if I get questions wrong?
Yes — use the "Try again" button on the results screen to reset your answers and go through all the questions again. There is no limit on attempts.
Is my progress saved?
Your answers and score for the current session are tracked in the browser as you go. No account or login is needed, and there is nothing to install.
What if I don't understand a term used in a question?
Read the explanation shown after you answer each question — it breaks down the correct term in plain English with a real-world example. You can also check the site Glossary for quick definitions.
How is this different from reading a blog article on the topic?
Exercises like this one are interactive drills that test and reinforce specific vocabulary through multiple-choice questions, while blog articles explain concepts in prose. Practising here after reading builds active recall, not just passive recognition.
Where can I find more Recommendation Systems Language exercises?
See the Recommendation Systems Language exercises hub for the full set of related pages, or browse all exercise categories from the main Exercises index.
Can I use this exercise to prepare for a technical interview?
Yes — recommendation systems language vocabulary comes up often in technical discussions and interviews. Pair this exercise with our dedicated Interview Preparation section for role-specific practice.