# Basic Recommender Systems

{{tag>Basic Recommender Systems}}


## Learning Objectives

1. Basic Concepts
  - Distinguish the different input data of a Recommender System
  - Classify the different families of Recommender System
  - Distinguish between implicit and explicit user feedback
  - Express the Global Effects formul

2. Requirements and Evaluation
  - Decide the most appropriate evaluation metric for the recommendation task
  - Decide the most appropriate splitting strategy
  - Distinguish between error and accuracy metrics
  - Discuss the importance of beyond accuracy metrics

3. Content-Based filtering
  - Define a content-based recommender system
  - Describe how to improve the different similarity functions
  - Discuss the importance of weighting the attributes
  - Describe advantages and disadvantages of Content-Based Filtering vs Collaborative Filtering

4. Collaborative Filtering
  - Select the most appropriate similarity function
  - Describe the difference between item-based and user-based models
  - Describe association rules based recommenders
  - Decide the most appropriate approach based on implicit or explicit feedback






## Basic Concepts

![](https://i.imgur.com/AFwBL1K.jpg)

### Taxonomy of Recommender Systems

  - Algorithems
    - Non-Personalized
    - Personalized
      - Content-Based Filtering ([[CBF]])
      - Collaborative Filtering ([[CF]])
        - User Based
        - Item Based
        - Matrix Factorization
        - Others (with CARS, Hybrids)
          - Factorization Machine
          - Deep Learning
      - Context-Aware ([[CARS]])
      - Hybrids Side Information

![](https://i.imgur.com/efLbS9x.jpg)

### Item Content Matrix





## 출처

- https://www.coursera.org/learn/basic-recommender-systems/home/welcome