Ranking parser for a natural language processing system

One or more computer-readable storage media having computer-executable instructions that, when executed by a computer, perform a method to determine language usage probabilities of a natural language based upon a training corpus, the method comprising:

• examining a training corpus, wherein such corpus includes phrases parsed in accordance with a set of grammar rules, the phrases having been parsed, at least partially, automatically and without human intervention;
• computing probabilities of usage of combinations of linguistic features based upon empirical tracking of appearances of instances of such combinations in phrases within the training corpus;
• and
• determining a goodness measure for one or more phrase of the training corpus based on a number of times that a particular headword included in the one more phrases is modified by a particular grammar rule and based on a number of times that the particular headword included in the one or more phrases is not modified by the particular grammar rule.

An apparatus comprising:

• a processor;
• a natural-language-usage parser including instructions executable by the processor to:
• generate at least one parse tree representing a syntactically valid parse of a phrase, wherein the parse tree has hierarchical nodes;
• calculate a syntactic history for each node;
• compute the probability for a node based upon the syntactic history calculated for that node.

An apparatus comprising:

• one or more processors;
• a data-acquisition device including instructions executable by the one or more processor to receive language-usage probabilities based upon appearances of instances of combinations of linguistic features within a training corpus;
• and
• a probability calculator including instructions executable by the one or more processors to calculate a probability at a node of a parse tree based upon linguistic features of the node and the language-usage probabilities.
• wherein the combinations of linguistic features comprises:
• (transition, headword, phrase level, syntactic history, segtype);
• (headword, phrase level, syntactic history, segtype);
• (modifying headword, transition, headword);
• and
• (transition, headword).

An apparatus comprising:

• one or more processors;
• a data-acquisition device including instructions executable by the one or more processors to receive language-usage probabilities based upon appearances of instances of combinations of linguistic features within a training corpus, wherein the training corpus includes phrases parsed in accordance with a set of grammar rules, the phrases having been parsed, at least partially, automatically and without human intervention;
• and
• a probability calculator including instructions executable by one or more processors to calculate a probability at a node of a parse tree based upon linguistic features of the node and the language-usage probabilities, wherein the probability is related to a null transition associated with the node.

One or more computer-readable storage media including instructions executable by a computer having a processor and a memory, to perform acts comprising:

• generating a data structure comprising:
• a corpus comprising one or more phrases in a natural language;
• parse trees having hierarchical nodes, each tree representing at least one syntactically valid parse of each phrase in a subset of the corpus;
• wherein each of one or more nodes have a syntactic history and a probability associated therewith, wherein the probability of a node is based on the syntactic history associated with a respective node, and wherein the syntactic history includes information related to a sibling node, a parent node, a child node. and any combination thereof, of the respective node.