Abstract: A visiting destination prediction device includes: a user position information acquiring unit configured to acquire user position information indicating a position of a user; an action history acquiring unit configured to acquire information on an action history of the user; and a visiting POI estimating unit configured to estimate a visiting POI which is a visiting destination of the user on the basis of a relationship between an acquired user position and a previously stored POI position and a relationship between the acquired action history of the user in a predetermined period and previously stored POI relevant information.

Abstract: Information related to apparently successful matches between two entities is collected, and culled based on a later indication that the match failed. Matches between two entities may be generated based on comparative information with other entities who appear to share some characteristics or preferences. Matches may be based on actual actions, in contrast to expressed preferences. Actual actions may be taken into account in addition to expressed preferences. Generation of matches may take into account geographical and/or temporal proximity and/or likelihood of receiving a response, in addition to other attributes of an entity. Matching algorithms may be updated based on entity input. Potential matches may be presented to third party entities for evaluation.

Abstract: Disclosed are systems, methods, and devices for pattern extraction. A graph data structure is generated, reflective of a directed graph having vertices representative of entities and edges representative of relationships between two entities. A subgraph data structure is generated, reflective of a plurality of subgraphs of the directed graph. A similarity matrix data structure is generated by applying a graph kernel to obtain a subgraph similarity matrix including a plurality of entries, each entry providing a score of the similarity between two subgraphs of the plurality of subgraphs. A clustering data structure is generated, reflective of a plurality of groups of the plurality entities upon processing the similarity matrix data structure. For at least a given one of the plurality of groups, a common pattern data structure is generated corresponding to a subgraph that is similar to subgraphs in the given group.

Abstract: Multiple stage filtering may be implemented for natural language query processing pipelines. Natural language queries may be received at a natural language query processing system and processed through a query language processing pipeline. The query language processing pipeline may filter candidate linkages for a natural language query before performing further filtering of the candidate linkages in the natural language query processing pipeline as part of generating an intermediate representation used to execute the natural language query.

Abstract: A fuzzy matching system matching data records in one or more data sets based on user-customized selection of multiple fuzzy matching algorithms. Possible matches may be displayed to a user, who provides feedback on the accuracy of the matches, which may then be used by a machine learning algorithm to update weightings and parameters of the multiple fuzzy matching algorithms, such as based on machine learning analysis of the matching results and the user feedback.

Abstract: An example method is provided in according with one implementation of the present disclosure. The method comprises generating, via a processor, a set of hashes for each of a plurality of objects. The method also comprises computing, via the processor, a high-dimensional sparse vector for each object, where the vector represents the set of hashes for each object. The method further comprises computing, via the processor, a combined high-dimensional sparse vector from the high-dimensional sparse vectors for all objects and computing a hash suppression threshold. The method also comprises determining, via the processor, a group of hashes to be suppressed by using the hash suppression threshold, and suppressing, via the processor, the group of selected hashes when performing an action.

Abstract: The present invention discloses numerous implementations of system and method which receives a user request and, using methods of natural language processing including part of speech tagging, analyses the user request to generate a query to a database of information. Based on the machine understanding, the system presents an interactive representation of the uttered request back to the user. This provides context to the user, which explains the machine understanding of the request and acts as an interface to iteratively refine or adjust the machine understanding by altering specific elements of the uttered language. The methods of altering specific elements of the uttered language may vary depending on the element and a variety of user selectable interfaces may be used to display one or more queried elements along with alternative elements pertaining to the queried element. The user could select an alternative element and change the database query.

Abstract: A central processing unit can offload table lookup or tree traversal to an offload engine. The offload engine can provide hardware accelerated operations such as instruction queueing, bit masking, hashing functions, data comparisons, a results queue, and a progress tracking. The offload engine can be associated with a last level cache. In the case of a hash table lookup, the offload engine can apply a hashing function to a key to generate a signature, apply a comparator to compare signatures against the generated signature, retrieve a key associated with the signature, and apply the comparator to compare the key against the retrieved key. Accordingly, a data pointer associated with the key can be provided in the result queue. Acceleration of operations in tree traversal and tuple search can also occur.

Abstract: A method includes receiving a plurality of product query arrays each including a plurality of individual product queries received during a single user search session. The method further includes inputting the plurality of product query arrays into the query rewrite model. Text of each of the plurality of individual product queries in each product query array is treated as a whole token. The method further includes receiving a product query from a user electronic device. The method further includes determining a query rewrite for the product query using the query rewrite model and determining search results for the product query using the query rewrite. The method further includes sending information for presenting the search results on a display of the user electronic device responsive to the product query.

Abstract: Aspects of the disclosure relate to processing queries and commands. A computing platform may receive a command input configured in a first format. The computing platform may input the command input into a natural syntactic language processor configured to extract parameters from the command input. The computing platform may input the command input into a decision system configured to identify a command type of the command input. The computing platform may generate, using a mapping catalogue and based on the parameters and the command type, an output command formatted in a second format. The computing platform may execute, against a binary store configured in the second format, the output command to obtain results to the command input. The computing platform may send commands directing an enterprise user device to display the results, which may cause the enterprise user device to display the results.

Abstract: Computer-implemented methods and systems for quantifying appropriate machine learning model complexity corresponding to training dataset are provided. The method comprises monitoring, using one or more processors, N observed variables, v1 through vN, of a training dataset for a machine learning model; translating the N observed variables into m equisized bin indexes which generate mN possible equisized hypercells to estimate a fundamental dimensionality for the dataset; generating one or more samples by assigning a record in the dataset with numbers j through k as set id; generating a merged sample Si, for one or more values of the set id i, where i goes from j to k; and computing a fractal dimension of the equisized hypercube phase space based on count of cells with data coverage of at least one data point.

Abstract: Systems, methods, computing platforms, and storage media for comparing non-adjacent data subsets are disclosed. Exemplary implementations may: receive an input data set, the input data set including information to be analyzed; generate at least one list of data subsets of the input data set; determine whether at least one data subset of the at least one list of data subsets contains a notable characteristic; identify at least one data pattern in the at least one list of data subsets of the input data set; sort data subsets of the at least one list of data subsets of the input data set; and display, via a user interface, the data subsets of the at least one list of data subsets, based on the sorting.

Abstract: Example artificial intelligence systems and methods provide parallel storage of data to primary storage and notification to a model server supported by the primary storage. A primary storage system receives operations on a training data set from a model trainer and sends a model instance of a computational model to a model server. When a new data element is received by a data ingester, the model server is initiated to evaluate the new data element using the model instance while the primary storage system stores the new data element in parallel.

Abstract: A search engine optimizer which works independently and in parallel with a browser and search engine supercomputer to gather, analyze, and distill input information interactively. The optimizer reorganizes the input, and providing an optimized version as an output. The optimized version of the input (e.g. output) is sent to the search engine which responds to the end user with search results. The optimizer recognizes each request as a pattern and stores the pattern in an advanced Glyph format. This permits the optimizer to identify a left and ride side check mate combination required to achieve certitude.

Abstract: An interactive interpretation session with respect to a first version of a machine learning model is initiated. In the session, indications of factors contributing to a prediction decision are provided, as well indications of candidate model enhancement actions. In response to received input, an enhancement action is implemented to obtain a second version of the model. The second version of the model is stored.

Abstract: Partial initial construction of a deferred object model. This is done using a map that correlates positions of a hierarchically structured definition and corresponding hierarchical positions within an object model. The map is accessed and used to initially construct a deferred object model that even leaves some of the deferred objects unpopulated. The map is used to determine which parts of the hierarchical object definition no longer need to be parsed in order to construct this initial form of the deferred object model. If a request for an object is detected at some point after the initial construction, and that requested object is not represented even in deferred form in the deferred object model, the system uses the map to find the position of the corresponding object definition in the hierarchical object definition. At that point, the system parses that position, and constructs the object.

Abstract: A method can include obtaining target data corresponding to a link. The method can include determining, based on the target data, that the link is valid. The method can include obtaining, in response to the determining that the link is valid, source data corresponding to the link. The method can include determining, by comparing the target data to the source data, a degree of relevance between the target data and the source data. The method can include determining, based on the degree of relevance, that the link is incorrect. The method can include generating, in response to the determining that the link is incorrect, a corresponding notification. The method can include initiating a transfer of the notification to a user device.

Abstract: Methods and systems for redirection of requests received by a service or an API are provided. According to one embodiment, a request, relating to a domain or a focal point of related data (e.g., an OData entity), is received from a client by a service (e.g., an OData service) operable to serve data from a data source or an in-memory cache. When the request involves manipulation of a portion of the data, automatic updating of values of the data that are dependent thereon is facilitated by directing fulfilment of the request to the cache; otherwise: it is determined whether to fulfill the request based on the in-memory cache. When the determination is affirmative, then the portion of the data is retrieved from the in-memory cache. When the determination is negative, then the portion of the data is retrieved from the data source.

Abstract: A request is received from a user to search a document for at least one search term. The document is analyzed to determine contextual meanings associated with the at least one search term. An ordered set of contexts of the at least one search term is constructed. An option of disambiguation is then prompted to the user to select a context of the ordered set of contexts. Search results are presented to the user associated with the selected context of the at least one search term.

Abstract: Methods and system for item matching are described. In one embodiment, compatibility-based text for an item may be accessed. A compatibility identifier may be identified based on the compatibility-based text. The compatibility identifier may be associated with an item cluster. The compatibility identifier may be used to identify a plurality of matching items. A result may be provided based on identification of the plurality of matching items. Additional methods and systems are disclosed.