Please enable JavaScript.
Coggle requires JavaScript to display documents.
LTR IR (Examples (Definition Finding, Sentiment Analysis, Key-term…
LTR IR
-
Relevance Judgments
-
in practice, always unjudged documents exist
In practice, these documents are IRRELEVANT when going to judge any algorithm.
Pointwise Approach
-
-
Note that the pointwise approach does not consider the interdependency
among documents, and thus the position of a document in
the final ranked list is invisible to its loss function
MODELS
Query-Dependent
-
-
e.g, using TF-IDF to generate features vectors
-
-
-
-
The capability of combining a large number of features is a very
important advantage of learning to rank. Such a capability is highly
demanding for real search engines, since it is almost impossible to use
only a few factors to satisfy complex information needs of Web users
Discriminative training is an automatic learning process based on
the training data. This is also highly demanding for real search engines,
because everyday these search engines will receive a lot of user feedback
and usage logs indicating poor ranking for some queries or documents.
It is very important to automatically learn from feedback and constantly
improve the ranking mechanism