• A
  • A
  • A
  • ABC
  • ABC
  • ABC
  • А
  • А
  • А
  • А
  • А
Regular version of the site
Student
Title
Supervisor
Faculty
Educational Programme
Final Grade
Year of Graduation
David alejandro Cortes rivera
Learning to Rank trough User's Preference Data
Big Data Systems
(Master’s programme)
9
2016
This work examines the problem of arranging items in electronic catalogs. The intuitive idea is to devise orderings that respect aggregated user preferences. Hypotheses about how user preferences drive browsing behavior are reviewed and some simple rules from deducing pairwise preferences between items from usage logs are explored, which are shown to agree to a high degree with explicitly-stated preferences.

Problems and properties of ranking items solely with pairwise preferences are explored, and some criteria for ranking with implicitly-deduced preferences are established.

Some algorithms were proposed for ranking items based on these preferences, and these were compared to existing algorithms for similar problems. One of these algorithms was found to produce results that are better or at least as good as existing algorithms. Then, a different approach was proposed, based on representing items by their attributes rather than as whole elements.

The advantages and disadvantages of both methods were examined, as well as the cases where they could and could not be used and possible extensions based on variables such as time.

These ranking methods were tried with data provided by a retail company and were found to generate orderings that respect user preferences to a greater extent than popular approaches such as sorting by clicks.

Student Theses at HSE must be completed in accordance with the University Rules and regulations specified by each educational programme.

Summaries of all theses must be published and made freely available on the HSE website.

The full text of a thesis can be published in open access on the HSE website only if the authoring student (copyright holder) agrees, or, if the thesis was written by a team of students, if all the co-authors (copyright holders) agree. After a thesis is published on the HSE website, it obtains the status of an online publication.

Student theses are objects of copyright and their use is subject to limitations in accordance with the Russian Federation’s law on intellectual property.

In the event that a thesis is quoted or otherwise used, reference to the author’s name and the source of quotation is required.

Search all student theses