Methods, systems, and apparatus, including computer programs encoded on computer storage media, for re-ranking a collection of documents according to a first metric and subject to a constraint on a function of one or more second metrics. One of the methods includes: obtaining, for each document in the first collection of documents, a respective first metric value corresponding to the first metric and respective one or more second metric values corresponding to the one or more second metrics; re-ranking the first collection of documents, comprising: determining the constraint on the function of one or more second metrics by computing a first threshold value using a variable threshold function that takes as input second metric values for the documents in the first collection of documents; and determining the re-ranking for the first collection of documents by solving a constrained optimization for the first metric constrained by the first threshold value.

An apparatus comprises a processing device configured to receive a query to extract information from a document, and to perform two or more iterations of utilizing a machine learning-based information extraction model to extract portions of unstructured text data from the document. In each iteration, a portion of the unstructured text data extracted from the document and an associated relevance score are output. In a first iteration, the query and document are input while in subsequent iterations the query and modified versions of the document are input, the modified versions having previously-extracted portions of the unstructured text data removed therefrom. The processing device is also configured to generate a response to the query comprising a subset of the portions of the unstructured text data extracted from the document determined to have associated relevance scores exceeding a threshold relevance score and at least a threshold level of similarity to the query.

A method of disambiguating user queries in a multi-turn dialogue including a set of user utterances. The method includes using a predefined language model to recognize an ambiguous entity in an unresolved user utterance from the multi-turn dialogue, and using the predefined language model to recognize entity constraints of the ambiguous entity. The method further includes, in a computer-accessible conversation history of the multi-turn dialogue, searching a set of previously-resolved entities for a candidate entity having entity properties with a highest confidence correspondence to the entity constraints of the ambiguous entity. The unresolved user utterance is rewritten as a rewritten utterance that replaces the ambiguous entity with the candidate entity. The rewritten utterance is output to one or more query answering machines.

The present disclosure provides a picture information processing method based on a mobile terminal, where the picture information processing method based on a mobile terminal includes: generating, by the mobile terminal, a query request information according to a information of a picture of an object taken by a user and a shooting positional information of the object, and sending a query request information to the back-end server; searching and matching, by the back-end server, corresponding introductory information of the object, and sending the introductory information of the object back to the mobile terminal. The present disclosure makes the user easily obtain the introductory information of the object, which provides convenience for the user.

A graphical platform enables users to leverage unstructured data in many different ways via access with many different types of devices that can be located practically anywhere with network connectivity. The graphical platform provides users with flexibility in how they create, maintain, and access the unstructured data, while providing powerful tools to enable the users to quickly extract meaningful information from the unstructured data. The graphical platform leverages location information in the unstructured data to enable providing a map and/or other graphical interfaces that show at least some of the records in the unstructured data. Users can then filter the information to narrow down the records to those of interest to respective different users. From the resulting display of records, users can select a specific record, access additional secure documents associated with the record, and/or perform other interactions with the data in the record.

Techniques are disclosed for processing a data set that is not organized according to a schema being used for organizing data (referred to herein as an “unstructured data set”). An unstructured data set is analyzed based on a group of structured data sets that are organized according to the schema. A particular structured data set is determined to be associated with the unstructured data set. The unstructured data set is stored in association with the particular structured data set. Periodically, the unstructured data set is re-analyzed based on a current version of the group of structured data sets. Additionally or alternatively, an unstructured data set is analyzed based on a particular schema of a set of schemas. A subset of information is extracted from the unstructured data set, and stored in accordance with the particular schema. Periodically, the unstructured data set is re-analyzed based on a current version of the set of schemas.

The current document is directed to a query-as-a-service system (“QAAS system”) that collects enormous volumes of data from network-connected entities, referred to as “Things” in the phrase “Internet of Things,” persistently stores the collected data and provides a distributed-query-execution engine that allows remote clients to continuously execute queries against the collected data. In a described implementation, both the raw data and query results are persistently stored in the QAAS system, with the raw data stored for significantly longer periods of time. Query results generated by the query-processing engine are securely transmitted to QAAS remote clients for distribution to file systems, storage appliances, applications, and other data sinks within client systems.

Apparatus, systems and methods for contextual search term evaluation are disclosed. A current search query is received during a search session. A predicate subsequence in the search query is identified. A subsequent search term in the query is identified. The search term attributes of the subsequent search term are adjusted.

An apparatus for predicting a command in a command line interface includes a template command module, a parameter derivation module, and a parameter substitution module. The template command module is configured to determine a template command based on a command line history. The template command includes a command name and a parameter and the command line history includes two or more previously entered commands. The parameter derivation module is configured to determine a parameter derivation rule for deriving the parameter in the template command based on the command line history. The parameter substitution module is configured to substitute a substitute parameter for the parameter of the template command according to the parameter derivation rule.

Disclosed are methods, systems, devices, apparatus, media, design structures, and other implementations, including a method that includes receiving, at a local device from a remote device, query data representative of a question relating to source content of a source document, and determining whether one or more pre-determined questions stored in a question-answer cache maintained at the local device matches the query data according to one or more matching criteria. The method further includes obtaining from the question-answer cache, in response to a determination that at least one of the pre-determined questions matches the query data received from the remote device, at least one answer data item, associated with at least one pre-determined question, corresponding to an answer to the question relating to the source content.