Techniques are disclosed for classifying a user search query for a database system. In disclosed techniques, a computing system receives a user search query for the database system and determines, based on the user search query, a database query that is compatible with an interface of the database system to implement the user search query. In some embodiments, the computing system causes a query remediation interface to be presented to a user that entered the user search query. In some embodiments, the interface includes: a classification of the user search query that specifies attributes of the database query, and one or more elements selectable to alter the database query. The computing system receives, from the user via the query remediation interface, input for altering the database query and determines an updated database query based on the input. The computing system may access the database system using the updated database query.

Methods, apparatuses, and systems for intention based search techniques are described herein. An example method comprises receiving data indicative of an intent of a user to perform an action via at least one application, the data comprising a description of the action that is input via a client device application on a client device, identifying, by the computing device, the at least one application based on a comparison of the description of the action and one or more records, the one or more records being indicative of prior intents of one or more users to perform actions and indicative of at least one selection by the one or more users to perform the actions with a selected application, and providing, by the computing device, output to the client device to enable performance of the action via the at least one application.

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for determining query refinements using search data. In one aspect, a method includes receiving a first query and a second query each comprising one or more n-grams for a user session, determining a first set of query refinements for the first query, determining a second set of query refinements from the first set of query refinements, each query refinement in the second set of query refinements including at least one n-gram that is similar to an n-gram from the first query and at least on n-gram that is similar to an n-gram from the second query, scoring each query refinement in the second set of query refinements, selecting a third query from a group consisting of the second set of query refinements and the second query, and providing the third query as input to a search operation.

Systems and methods for providing a search-directed user interface for online banking applications. An example method may comprise: receiving, via a graphical user interface (GUI) session associated with an authenticated user, a search argument comprising a character string; executing, by a processing device, a search query by matching the character string to account data of one or more accounts that the authenticated user is authorized to access, the account data comprising a plurality of data items representing at least one of: financial product types, financial product identifiers, financial transaction types, financial transaction descriptions, financial transaction amounts, portfolio types, accounts, and aggregated financial indicators; and causing a data set produced by executing the search query to be visually represented via the GUI session.

A computer-implemented method includes: reading a vector of a first table in a database, the vector including counts of a plurality of keywords in the first table, the plurality of keywords including a first keyword and a second keyword; determining a first custom attribute describing the first table, the first custom attribute having a vector including counts of at least a first portion of the plurality of keywords in the first table; determining a multiplier of the first custom attribute, the multiplier being a number of other tables that reference the first custom attribute; and revising the vector of the first table based on the first custom attribute.

An information output receives a conversational sentence, parameterizes one or more entities included in one or more received conversational sentences, searches a database using the to acquire a search result, and outputs the search result.

Described herein is a regulatory parser that downloads and efficiently processes regulatory documents. The regulatory documents may be from different sources and may have different formats. The regulatory parser parses all of the text in the regulatory documents and converts into a predetermined, single format for downstream applications. The text is organized and stored in a structured tree, organized into one or more hierarchies with nodes storing segments of text from a regulatory document. In some embodiments, each node in the regulatory tree may represent a segment of text. Partitioning the text of a regulatory document into segments of text may make the storage and querying of the regulatory documents more manageable. The organization and structure of the structured tree may reduce the times and resources needed for accessing and searching for a regulatory citation. The structured tree may allow a user to manipulate a regulatory document or text.

The present invention relates generally to the field of providing a computer-implemented system and method that supports ad hoc multilingual Natural Language database questions that are transformed in realtime to Database queries using a suite of cascading Deep Neural Networks which are initially and continuously trained through schema information and question-query examples.

In one embodiment, a method includes receiving a user input from a first user at the first client system, determining that the user input is a sharing request to share content, determining multiple second users the sharing request is directed to, determining, for each second user, modalities associated with the respective second user based on the content, a user profile associated with the respective second user, and modalities supported by a second client system the respective second user is currently engaged with, the respective second user being associated with two or more second client systems, and sending, to one or more second client systems currently associated with the second users, instructions for accessing the content based on the determined modalities for each second user.

Among other things, a developer of an interaction application for an enterprise can create items of content to be provided to an assistant platform for use in responses to requests of end-users. The developer can deploy the interaction application using defined items of content and an available general interaction model including intents and sample utterances having slots. The developer can deploy the interaction application without requiring the developer to formulate any of the intents, sample utterances, or slots of the general interaction model.