KIIS COLLOQUIA - MOD 1

This course is part of the PhD program in Computer Science. It aims at investigating the state of the art methods and algorithms in the are of Web Mining, and in particular of learning to rank.

Students will achieve the following learning outcomes:

Knowledge and understanding: understanding the most advanced learning to rank models and algorithms

Applying knowledge and understanding: being able to design novel algorithms for learning to rank tasks

Making judgements: being able to analyze different methods and compare them with state of the art solutions

Basic notions Machine Learning and Information Retrieval.

Helping a user in finding interesting pieces of information among the billions pages of the Web is a very challenging problem, an open research area and an application field for the most advanced techniques in the areas of Web Search and Data Mining. This course reviews the most recent advances in the area of learning to rank.

Topics include:
- Introduction to the Ranking Problem
- Machine Learning methodology
- Information Retrieval Evaluation measures
- Ranking with BM25F
- Line Search optimization
- Pairwise approaches: RankNet
- Pairwise approaches: RankingSVM
- Genetic Algorithms
- Gradient Boosting Machines

Scientific papers.
Lecture Notes.

Oral exam with presentation and discussion of one scientifica paper selected by the teacher.

Scientific papers.
Lectures.

oral