---
title: "Recommender system design and types"
subject: "Information System"
chapter: 7
key: "semester_8/information_system:recommender-design-types"
source: https://www.bctengineer.com/q/semester_8/information_system/recommender-design-types
site: BCT Engineering
---

# Recommender system design and types

Information System · Chapter 7: Web based information system and navigation

- Asked in: 2074 Magh, 4+4 marks
- Video: [Recommendation System : Content Based Recommendation and Collaborative Filtering Explained in Hindi](https://www.youtube.com/watch?v=jhFFOmvPne4) — 5 Minutes Engineering
- Answer checked by Claude Opus 5.5 (high effort) on 2026-10-14
- Page: https://www.bctengineer.com/q/semester_8/information_system/recommender-design-types
- Whole chapter: https://www.bctengineer.com/md/semester_8/information_system/chapter-7

## Question

Describe the characteristics for the design of the typical recommender system. Explain the different types of recommender systems.

## Short answer

Recommender design requires accuracy, scalability, sparsity handling, and real-time personalization. Types include content-based (item similarity), collaborative filtering (user similarity), hybrid, knowledge-based (explicit rules), demographic, popularity-based, and context-aware systems.

## Answer

#### Characteristics for the design of a typical recommender system
- **Data collection:** explicit (ratings, reviews) and implicit (clicks, purchases, viewing time) feedback, plus user profiles and item attributes.
- **Accuracy and relevance:** predictions must match user taste.
- **Scalability:** handle millions of users and items in real time.
- **Handling sparsity and cold start** for new users and items.
- **Diversity, novelty and serendipity:** avoid showing only the obvious.
- **Personalisation and context awareness:** time, place and device.
- **Real-time response and updating** with new data.
- **Robustness and security:** resist fake ratings and attacks.
- **Privacy:** protect user data.
- **Explainability and trust:** tell why an item is recommended.
- **Evaluation:** precision, recall, RMSE and user satisfaction.

#### Types of recommender systems
| Type | Principle | Example |
|---|---|---|
| **Content-based** | Recommends items similar to those the user liked, using item features | News with similar topics |
| **Collaborative filtering** | Uses preferences of similar users or items | Amazon "customers also bought" |
| **Hybrid** | Combines two or more methods to reduce weaknesses | Netflix |
| **Knowledge-based** | Uses explicit requirements and domain rules | Choosing a laptop by budget and RAM |
| **Demographic** | Uses age, gender, location | Recommending local products |
| **Popularity-based** | Recommends most popular items | Top 10 lists |
| **Context-aware** | Considers time, location, mood | Restaurants nearby at lunch |

## Check yourself

The answer's key points as multiple-choice questions, with the correct option and why.

1. Which design characteristic avoids showing only obvious items?
   A. Scalability
   B. Privacy
   C. Diversity, novelty and serendipity
   D. Explainability
   Answer: C. These aim for surprising and varied recommendations. (Key point: Design characteristic)

2. How does a content-based recommender work?
   A. It uses ratings of other users only
   B. It recommends items similar to those the user liked, using item features
   C. It recommends the most popular items
   D. It uses only age and gender
   Answer: B. Collaborative filtering uses other users' preferences. (Key point: Content-based)

3. Why use a hybrid recommender?
   A. To combine methods and reduce their weaknesses
   B. To avoid using any data
   C. To show only top 10 lists
   D. To remove the need for users
   Answer: A. Netflix is a hybrid example. (Key point: Hybrid)

4. Choosing a laptop by budget and RAM is an example of which type?
   A. Popularity-based
   B. Demographic
   C. Content-based
   D. Knowledge-based
   Answer: D. It uses explicit requirements and domain rules. (Key point: Knowledge-based)

## Related in Information System

- [Recommender systems and examples](https://www.bctengineer.com/question.md?key=semester_8%2Finformation_system%3Arecommender-how-works-examples) — Information System, chapter 7
- [Generic to personalized recommender](https://www.bctengineer.com/question.md?key=semester_8%2Finformation_system%3Arecommender-generic-to-personalized) — Information System, chapter 7
- [DSS and GDSS characteristics](https://www.bctengineer.com/question.md?key=semester_8%2Finformation_system%3Adss-gdss-characteristics) — Information System, chapter 4

## Related in other subjects

- [Typical water supply system and components](https://www.bctengineer.com/question.md?key=semester_5%2Fwater_supply_engineering%3Atypical-ws-system-components) — Water Supply Engineering, chapter 1, asked 4×
- [Features of a typical operating system](https://www.bctengineer.com/question.md?key=semester_4%2Fmicroprocessor%3Afeatures-of-operating-system) — Microprocessor, chapter 6, asked 3×
- [Typical power system network and components](https://www.bctengineer.com/question.md?key=semester_4%2Fpower_system_analysis_i%3Atypical-power-system-network-components) — Power System Analysis I, chapter 1
- [Typical power system with voltage levels](https://www.bctengineer.com/question.md?key=semester_4%2Fpower_system_analysis_i%3Atypical-power-system-voltage-levels) — Power System Analysis I, chapter 1