Abstract: An example embodiment may involve a software application executable on a computing device of a computational instance of a remote network management platform associated with a managed network. The software application may receive, from a client device of the managed network, a natural language query, determine that a first term of the query does not match any term in a list of predetermined terms, and revise the query by (i) applying a set of linguistic rules that modify the first term to generate a replacement term, (ii) determining whether the replacement term matches at least one term in the list, and (iii) after determining that the replacement term matches at least one term in the list, replacing the first term in the query with the replacement term. The software application may then parse the revised query, search the database, and transmit a result to the client device.

Abstract: An example embodiment may involve a software application executable on a computing device of a computational instance of a remote network management platform associated with a managed network. The computational instance may contain a database from which information associated with the managed network can be delivered to the client device. The software application may receive, from a client device of the managed network, a natural language query, parse the query using context-free grammar rules in Backus-Naur form (BNF), generate a search sentence based on the parsed query, search the database using the search sentence as a search query, and transmit, to the client device, a result of the search query. The context-free grammar rules may include an extended context-free grammar rule that defines a non-terminal symbol and comprises a Boolean NOT operator that eliminates one or more potential definitions of the non-terminal symbol from consideration when parsing the query.

Abstract: An example embodiment may involve a software application executable on a computing device of a computational instance of a remote network management platform associated with a managed network. The software application may receive, from a client device of the managed network, a natural language query, determine that a first term of the query does not match any term in a list of predetermined terms, and revise the query by (i) applying a set of linguistic rules that modify the first term to generate a replacement term, (ii) determining whether the replacement term matches at least one term in the list, and (iii) after determining that the replacement term matches at least one term in the list, replacing the first term in the query with the replacement term. The software application may then parse the revised query, search the database, and transmit a result to the client device.

Abstract: An example embodiment may involve a software application executable on a computing device of a computational instance of a remote network management platform associated with a managed network. The computational instance may contain a database from which information associated with the managed network can be delivered to the client device. The software application may receive, from a client device of the managed network, a natural language query, parse the query using context-free grammar rules in Backus-Naur form (BNF), generate a search sentence based on the parsed query, search the database using the search sentence as a search query, and transmit, to the client device, a result of the search query. The context-free grammar rules may include an extended context-free grammar rule that defines a non-terminal symbol and comprises a Boolean NOT operator that eliminates one or more potential definitions of the non-terminal symbol from consideration when parsing the query.

Abstract: Architecture that automatically determines relationships between vector spaces of disparate media types, and outputs ranker signals based on these relationships, all in a single process. The architecture improves search result relevance by simultaneously clustering queries and documents, and enables the training of a model for creating one or more ranker signals using simultaneous clustering of queries and documents in their respective spaces.

Abstract: Architecture that automatically determines relationships between vector spaces of disparate media types, and outputs ranker signals based on these relationships, all in a single process. The architecture improves search result relevance by simultaneously clustering queries and documents, and enables the training of a model for creating one or more ranker signals using simultaneous clustering of queries and documents in their respective spaces.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for classifying resources using scores from multiple classifiers. In general, one aspect of the subject matter described in this specification can be embodied in methods that include the actions of receiving identifying a collection of documents to classify; receiving a plurality of classifiers for scoring a document with respect to a specified property; for each document in the collection, applying each of the plurality of classifiers, each classifier generating a score associated with a likelihood that the document has the specified property, combining the scores from each classifier including applying a multiple classifier model that uses monotonic regression to combine the plurality of classifiers, and classifying the document as having the specified property based on the combined score.

Abstract: Methods, systems, and apparatus, including computer program products, for constructing text classifiers. The method includes receiving a collection of candidate phrases for a given topic; filtering the received candidate phrases to remove erroneously included candidate phrases; assigning weights to the candidate phrases including scoring each candidate phrase using an initial classifier and assigning weights to the candidate phrases based on the scores; and generating a linear classifier using the filtered and weighted candidate phrases, where the linear classifier varies the weights for each phrase candidate depending on the length of the document being classified.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for classifying resources using scores from multiple classifiers. In general, one aspect of the subject matter described in this specification can be embodied in methods that include the actions of receiving identifying a collection of documents to classify; receiving a plurality of classifiers for scoring a document with respect to a specified property; for each document in the collection, applying each of the plurality of classifiers, each classifier generating a score associated with a likelihood that the document has the specified property, combining the scores from each classifier including applying a multiple classifier model that uses monotonic regression to combine the plurality of classifiers, and classifying the document as having the specified property based on the combined score.

Abstract: Architecture that automatically determines relationships between vector spaces of disparate media types, and outputs ranker signals based on these relationships, all in a single process. The architecture improves search result relevance by simultaneously clustering queries and documents, and enables the training of a model for creating one or more ranker signals using simultaneous clustering of queries and documents in their respective spaces.

Abstract: Methods, systems, and apparatus, including computer program products, for constructing text classifiers. The method includes receiving a collection of candidate phrases for a given topic; filtering the received candidate phrases to remove erroneously included candidate phrases; assigning weights to the candidate phrases including scoring each candidate phrase using an initial classifier and assigning weights to the candidate phrases based on the scores; and generating a linear classifier using the filtered and weighted candidate phrases, where the linear classifier varies the weights for each phrase candidate depending on the length of the document being classified.