Preliminary ranker for scoring matching documents

One or more computer storage media storing computer-usable instructions that, when used by one or more computing devices, cause the one or more computing devices to perform operations, the operations comprising:

• identifying a first set of documents as matching one or more terms from a search query;
• computing a preliminary rank score for each of at least a portion of the documents in the first set of documents, the preliminary rank score for each document being computed by:
• accessing a table associated with the document that stores pre-computed data used to derive a frequency for each of a plurality of terms in the document;
• algorithmically determining the frequency of at least one term from the search query based on the pre-computed data from the table;
• and
• computing the preliminary rank score of the document based at least on the frequency;
• selecting a subset of documents from the first set of documents based on the preliminary rank scores;
• generating a final rank score for at least a portion of the documents in the subset of documents;
• and
• providing search results based on the final rank scores.

A computer system comprising:

• one or more processors;
• and
• one or more computer storage media storing computer-usable instructions that, when used by the one or more processors, cause the one or more processors to perform operations comprising:
• computing a preliminary rank score for each document from a plurality of documents matching a search query, the preliminary rank score for each document being computed by:
• accessing a table associated with the document that stores pre-computed frequency data for each of a plurality of terms in the document, wherein the table stores pre-computed frequency data only for terms that occur more than once in the document;
• and
• computing the preliminary rank score of the document based at least on pre-computed frequency data from the table for at least one term from the search query;
• selecting a subset of documents from the plurality of documents based on the preliminary rank scores;
• generating a final rank score for at least a portion of the documents in the subset of documents;
• and
• providing search results based on the final rank scores.