avatarFabio Chiusano

Summary

The web content provides insights into various recommender system approaches, emphasizing the application of NLP techniques in content-based filtering and the challenges of collaborative filtering, including the cold start problem.

Abstract

The article discusses different types of recommender systems, focusing on their application in suggesting Medium articles to users. It details content-based filtering, which leverages item descriptions and NLP techniques for recommendations without requiring user interaction history. Collaborative filtering, on the other hand, relies on past user interactions to predict future preferences, facing the cold start issue. The author expresses a preference for collaborative filtering when sufficient data is available. User-based filtering is also mentioned, though it generally underperforms compared to the other methods. The article concludes by advocating for hybrid approaches that combine the strengths of various recommender systems.

Opinions

  • The author finds collaborative filtering to be more effective than content-based filtering when there is adequate user interaction data.
  • Content-based filtering is advantageous from the outset for new users as it doesn't require a history of user interactions.
  • Collaborative filtering is praised for its ability to recommend complex items without needing to understand the items themselves.
  • User-based filtering is noted to typically perform worse than content-based and collaborative filtering methods.
  • The author suggests that hybrid recommender systems are commonly used to mitigate the limitations of individual approaches.

Two minutes NLP — Tips for Recommender Systems with NLP

Content-based and User-based Filtering, Collaborative Filtering, and Hybrid Approaches

How suggestions work in different types of recommender systems. Image by the author.

There are several types of recommender systems, but not all of them are suitable to be implemented with NLP techniques. Suppose we are building a recommender system for Medium, where our goal is to suggest articles to the users.

Content-based filtering

How suggestions work in content-based filtering recommender systems. Image by the author.

Content-based filtering methods are based on descriptions of the items to be recommended. They are best suited to situations where there is known data on the items (like name, description, etc.), but not on the users (like his/her previously read articles). These algorithms try to recommend items similar to those that a user liked in the past or is examining in the present.

A key advantage of content-based filtering is that it doesn’t need to know a list of items the user has interacted with in the past, which is usually collected after the user has interacted with the service for some time. As a consequence, content-based filtering works well from day one of a new user.

If we have text data that describe the items, we can leverage NLP techniques to compute items' similarity with document embeddings, like the ones obtained with Doc2vec.

Collaborative filtering

How suggestions work in collaborative filtering recommender systems. Image by the author.

Collaborative filtering is based on the assumption that people who agreed in the past will agree in the future, and that they will like similar kinds of items as they liked in the past.

A key advantage of the collaborative filtering approach is that it does not rely on items’ descriptions and therefore it is capable of accurately recommending complex items such as movies without requiring an “understanding” of the item itself. However, this approach needs to know a list of items the user has interacted with in the past, thus suffering from the cold start problem.

As collaborative filtering does not rely on descriptive features of items and users, it’s not possible to leverage NLP techniques.

I personally found the collaborative filtering approach to outperform the content-based approach when enough data is available.

User-based filtering

How suggestions work in user-based filtering recommender systems. Image by the author.

It is possible to create recommender systems that are based on similarities between users as well, though they commonly perform worse than content-based and collaborative filtering.

Similar to content-based filtering, if we have text data that describe the users, we can use it to compute users similarity leveraging embeddings.

Hybrid filtering

How suggestions work in hybrid filtering recommender systems. Image by the author.

Given the pros and cons of the different types of recommender systems, it is very common to use a hybrid approach.

Thank you for reading! If you are interested in learning more about NLP, remember to follow NLPlanet on Medium, LinkedIn, and Twitter!

Two minutes NLP related posts

NLP
Naturallanguageprocessing
Data Science
Artificial Intelligence
Recommender Systems
Recommended from ReadMedium
avatarRahul Kumar
NLP Hands-On with Summarization

This post is a part of the NLP Hands-on series and consists of the following tasks: 1. Text Classification 2. Token Classification 3…

2 min read
avatarMohamed Elrefaey
Understanding Local Loss, Focal Loss, and Gradient Blending in Multi-Task Learning

Overview

6 min read
avatarMdabdullahalhasib
A Complete Guide to Embedding For NLP & Generative AI/LLM

Understand the concept of vector embedding, why it is needed, and implementation with LangChain.

11 min read
avatarVipra Singh
LLM Architectures Explained: NLP Fundamentals (Part 1)

Deep Dive into the architecture & building of real-world applications leveraging NLP Models starting from RNN to the Transformers.

39 min read
avatarMaya Murad
How to Build a General-Purpose LLM Agent

A Step-by-Step Guide

11 min read
avatarOkan Yenigün
Fine-Tuning ABSA Models with SetFit: A Practical Guide

Step-by-Step Fine-Tuning of ABSA Models with SetFit

6 min read