With the advent of web search engines and the huge query logs that are collected
from user interactions, the amount of potential training data is enormous.
This has led to the development of new techniques that are having a significant
impact in the field of information retrieval and on the design of search engines.
In the next section, we describe techniques for learning ranking algorithms that
can combine and weight the many pieces of evidence that are important for web
search.
Another very active area of machine learning has been the development of sophisticated
statistical models of text. In section 7.6.2, we describe how these models
can be used to improve ranking based on language models.