jump to navigation

Are User-Based Recommenders Biased by Search Engine Ranking? September 28, 2010

Posted by Andre Vellino in Collaborative filtering, Recommender, Recommender service, Search, Semantics.
trackback

I have a hypothesis (first emitted here) that I would like to test with data from query logs: user-based recommenders – such as the ‘bX’ recommender for journal articles – are biased by search-engine language models and ranking algorithms.

Let’s say you are looking for “multiple sclerosis” and you enter those terms as a search query. Some of the articles that were presented to you from the search results will likely be relevant and you download a few of the articles during your session. This may be followed by another, semantically germane query that yeilds more article downloads. As a consequence, the usage-log (e.g. the SFX log used by ‘bX’) is going to register these articles as having been “co-downloaded”.  Which is natural enough.

But if this happens a lot, then a collaborative filtering recommender is going to generate recommendations that are biased by the ranking algorithm and language model that produced the search-result ranking: even by PageRank, if you’re using Google.

In contrast, a citation-based (i.e. author-centric) recommender (such as Sarkanto) will likely yield more semantically diverse recommendations because co-citations will have (we hope!) originated from deeper semantic relations (i.e. non-obvious but meaningful connections between the items cited in the bibliography).

Comments»

1. xlvector - September 28, 2010

Yes, user-based CF always produce popular results because popular items tend to be liked by many users have similar taste with you.

Andre Vellino - October 3, 2010

What I was trying to say is that user-based journal article recommenders that are based on web-based usage logs (such as ‘bX’) are likely to be biased by the language-model and ranking method of the search engine(s) that were used to obtain the clicked-on links in the first place.

In contrast, ratings-based recommenders (e.g. for movies or songs) are not (generally speaking) strongly influenced by search engines or how they rank results.


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: