A method for profile matching includes receiving a plurality of user profiles, each user profile comprising traits of a respective user. The method includes receiving a preference indication for a first user profile of the plurality of user profiles. The method also includes determining a potential match user profile of the plurality of user profiles based on the preference indication for the first user profile. The method also includes presenting the potential match user profile to a second user.

A non-transitory computer-readable recording medium storing an estimation program that causes a processor included in a computer to execute a process. The process includes receiving a message of an estimation target of a forwarding destination, searching past messages for a message similar to the received message, estimating, when there is no similar message among the past massage, the forwarding destination of the message using a model of estimating the forwarding destination based on a keyword that is identified based on a word included in the received message, and outputting a forwarding destination of the searched similar message or the estimated forwarding destination.

A system and method for creating a machine learning (ML) classifier for a database uses a weakly-supervised training data set created automatically from database items on the basis of a human-created keyword set. The automatically created training data set is used to construct one or more deep learning classifier checkpoints, which can then be compared with one another and with a classifier based on the original keyword set in order to select a classifier for use by other users viewing the database.

A computer-implemented method for determining an origin of a target source code for a computer program or a part thereof. The method involves: Searching a plurality of software archives from different sources in a global computer network to find occurrences of the target source code among code files in said software archives. For every found occurrence of the target source code, the method further involves collecting key information about the matching source code files and, from the key information collected a frequency map is built that contains, for each keyword found in the key information, a keyword count value being indicative of the number of times the keyword occurs in the key information. The method may further involves applying a scoring scheme to the matching source code files based on the built frequency map (310), determining a highest score (Vmax) among the matching source code files after the scoring scheme has been applied, and determining the origin of the target source code as the matching source code file having the highest score.

Methods, apparatus, systems, and computer-readable media are provided for providing context specific schema files that allow an automated assistant to broker human-to-computer dialogs between a user and an application that is separate from the automated assistant. The context specific schema file can provide the automated assistant with sufficient data to be responsive to user queries without necessarily communicating with a remote device, such as a server. Multiple different context specific schema files can be made available to the automated assistant according to a context in which a user is interacting with the automated assistant. In this way, latency otherwise exhibited by the automated assistant can be mitigated by providing the automated assistant with the information needed to respond to a user without continually retrieving the information over a network.

A method for profile matching includes receiving a plurality of user profiles, each user profile comprising traits of a respective user. The method includes receiving a preference indication for a first user profile of the plurality of user profiles. The method also includes determining a potential match user profile of the plurality of user profiles based on the preference indication for the first user profile. The method also includes presenting the potential match user profile to a second user.

A relevance searching method performed by a computer, the relevance searching method includes generating a combined database by combining a plurality of databases each including a plurality of elements and relevance information indicating direct relevance between two elements in the plurality of elements; and searching for relevance between two elements that do not have direct relevance by using the combined database.

An illustrative embodiment includes a method for post-retrieval query performance prediction using hybrid document-passage information. The method includes: obtaining a set of documents responsive to a specific query; extracting document-level information regarding respective documents within the set; extracting passage-level information regarding respective passages of documents within the set; and estimating a likelihood that the set of documents includes relevant information to the specific query using both the document-level information and the passage-level information.

An audio file format of an audio portion of a natural language content is determined. Using a trained audio language identification model, a human language included in the audio portion is identified. Using a trained audio to text model trained on the human language, the audio portion is converted to a corresponding set of text data. The set of text data is indexed. Using the indexed set of text data responsive to a search query, a search result is generated, the search query specifying a search including a non-textual portion of the natural language content.

Provided are systems, methods, and computer-program products for a proxy network that can determine, for a set of objects, an initial differential, where the initial differential is determined using a rules data store of a host network. The proxy network can further determine a supplemental differential for the set of objects. The proxy network can further determine a final differential that is the sum of the initial differential and the supplemental differential. The proxy network can further determine a final sum for the set of objects that is the object value less the final differential. The proxy network can further generate an outbound data packet that includes values corresponding to the set of objects and the final sum. When the host network receives the outbound data packet, the host network can modify an object data store using the values corresponding to the set of objects and the final sum.