Skip to content

What makes a Search Engine different from a Recommender System?

October 1, 2013

Introduction

Everyone who spends time surfing the web comes into regular contact with both search engines and recommender systems, whether they know it or not.  But what is it that makes them different from one another?  This is a question that I’ve been asking myself more and more as of late so I thought that I’d start to put some of my thoughts down.

The Underlying Technologies

I know plenty of folk who begin to answer this question by saying that search engines and recommender systems are different technologies.  In the early days of their development the people working on these systems split largely into two different communities, one of which focussed more on Information Retrieval and the other on Information Filtering.  As a result, different avenues of research were pursued, which, despite some cross-fertilisation, shaped the fields, and resulting technologies, in different ways.

These two communities are increasingly coming back together as advances in search engines include lessons learned from Information Filtering techniques (e.g. collaborative search) and recommender systems start exploiting well established Information Retrieval techniques (e.g. learning to rank).  As such, it’s becoming less relevant to distinguish search engines and recommender systems based on their underlying technologies.

So how should we distinguish them?  I think that their primary difference is found in how users interact with them.

How to Spot a Search Engine

  1. You see a query box where you type in what you’re looking for and they bring back a list of results.  From gigantic search engines like Google to the discrete search boxes that index the contents of a single page blog, they all have query boxes.
  2. You start with a query that you enter into a query box.  You have an idea of what you’re looking for.  That thing may or may not exist but if it does then you want the search engine to find it for you.
  3. You find yourself going back to reformulate your query as you see what results it gives and you widen or narrow the search.
  4. You may even generate a query by clicking on items that interest you (e.g. movies that you like).  The search engine will then retrieve more movies for you.

How to Spot a Recommender System

  1. Some content has just appeared on your screen that is relevant to you but you didn’t request it.  Where did that magic come from?  That would be a recommender system.
  2. You don’t build a query and request results.  Recommendations engines observe your actions and construct queries for you (often without you knowing).
  3. They tend to power adverts, which can give them a bad image, especially if the content is embarrassing.

Conclusion

Search Engines are not the same as Recommender Systems.  Both can provide personalised content that matches your needs but it’s not what they do or the techno-magic that they use to do it but more how you interact with them that distinguishes them from one another.

Advertisements

From → Uncategorized

3 Comments
  1. Nice post, Kris. I have come across with exactly the same problem recently. The term “recommendation” has become a bit “semantically overloaded”, with all the typical hype of new stuff. To the point that, for some, a search result is a “recommendation”, and filtering the results is a “recommendation constraint” :S

    Thanks again – your post will come in handy 🙂

  2. it’s an interesting question, indeed. from a user modelling perspective i would argue:

    Typically, search engines have only sparse user model of their users (i.e. the search term), and this user model is created explicitly by the user (by typing it to the search box). in contrast, a recommender system usually (but not necessarily) has more comprehensive user models that are usually (but not necessarily) created be the recommender system through analyzing the user’s behaviour or existing data. In other words, there are two dimensions that distinguish search engines from recommender systems: the size of the user models and the way of creating the user models.

    The problem is that there is no real boundary between the two types of systems because there are also recommender systems utilizing very small user models and in some cases, recommender systems even do not create the models automatically but let the users specify their interests. So, what’s the difference between a user entering his interest(s) into a recommender system manually and a user entering his information need into the search box of a search engine? imho, there is hardly any difference.

    btw. the question becomes even more interesting, if you ask “what’s the difference between search engines, *personalized* search engines, and recommender systems”? 🙂

  3. bernielandry permalink

    Thanks some serious info on doing research
    Will put this to use this weekend for a workshop I am preparing to give. B

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: