System and method for using anchor text as training data for classifier-based search systems

A method of integrating a new document into a corpus of documents, the method comprising:

• generating the new document;
• modifying at least one existing document in the corpus to have anchor text relating to the new document;
• training a machine learning classifier using at least the anchor text;
• and
• publishing the new document and the at least one modified document.

A method of training a machine learning classifier for retrieving documents from a corpus, the method comprising:

• extracting a first set of training data relative to at least some of the documents;
• extracting a second set of training data relative to at least some of the documents;
• generating a combined training set using at least some of the first set of training data and the second set of training data, wherein the first and second sets of training data are of different types of data;
• and
• training the machine learning classifier with the combined training set.