Abstract: Systems and methods are disclosed for processing digital images to predict at least one continuous value comprising receiving one or more digital medical images, determining whether the one or more digital medical images includes at least one salient region, upon determining that the one or more digital medical images includes the at least one salient region, predicting, by a trained machine learning system, at least one continuous value corresponding to the at least one salient region, and outputting the at least one continuous value to an electronic storage device and/or display.

Abstract: Systems and methods are disclosed for processing digital images to predict at least one continuous value comprising receiving one or more digital medical images, determining whether the one or more digital medical images includes at least one salient region, upon determining that the one or more digital medical images includes the at least one salient region, predicting, by a trained machine learning system, at least one continuous value corresponding to the at least one salient region, and outputting the at least one continuous value to an electronic storage device and/or display.

Abstract: Techniques are disclosed relating to the serialization of data of a database. A computer system may maintain a database that stores data as a plurality of triples. A triple may include a first data entity that is connected to a second data entity via an edge, and triples of the plurality of triples may be connected to form graphs. The computer system may store metadata defining types of edges between data entities. An edge's type may indicate a procedure for serializing a target data entity connected to that edge's type. The computer system may receive a request to access a data entity associated with a plurality of descendant data entities connected to the data entity via edges. The computer system may generate an output having the data entity and one or more of the plurality of descendant data entities that are serialized according to the types of edges.

Abstract: Techniques are disclosed relating to the garbage collection of data of a database. A computer system may maintain a database that stores data as a plurality of triples. A given triple may include a first data entity that is connected to a second data entity via an edge, and triples of the plurality of triples may be connected to form a plurality of graphs. The computer system may store metadata that defines strengths of edges between graphs of the plurality of graphs. The computer system may receive a request to delete a particular triple or graph of the plurality of graphs. The particular graph may include a set of edges originating in the particular graph to other graphs of the plurality of graphs. The computer system may determine, based on the stored metadata, which graphs of the other graphs to delete in addition to the particular graph.

Abstract: The technology disclosed describes systems and methods for generating feature vectors from resource description framework (RDF) graphs. Machine learning tasks frequently operate on vectors of features. Available systems for parsing multiple documents often generate RDF graphs. Once a set of interesting features to be considered has been established, the disclosed technology describes systems and methods for generating feature vectors from the RDF graphs for the documents. In one example setting, a machine learning system can use generated feature vectors to determine how interesting a news article might be, or to learn information-of-interest about a specific subject reported in multiple articles. In another example setting, viable interview candidates for a particular job opening can be identified using feature vectors generated from a resume database, using the disclosed systems and methods for generating feature vectors from RDF graphs.

Abstract: A golf putter head having adjustability in weight and center of gravity. The putter head has a ball-striking front portion with a club face defining its width and height, a heal, a toe, and top and bottom surfaces defining the length of the front portion. The putter head has a front central axis extending through the center of gravity of the front portion and parallel to the top and bottom surfaces. A weight-and-balance-adjustment rear portion is connected to the front portion, which has a plurality of lengthwise-extending housings. Each housing includes an internal cavity constructed and arranged to removably contain at least one weighted slug. The putter head has a first center of gravity when the internal cavities are empty. The first center of gravity can be moved along the length, width and height of the putter by inserting weighted slugs in one or more of the cavities.

Abstract: A liquid chromatography system and method utilizes a mobile phase comprising liquified compressible gas and miscible organic solvents. The compressible fluid may be carbon dioxide (CO2). Liquid CO2 tapped from an existing source is depressurized through a flow control metering station before adding solvent. The mobile phase flows through a sample vessel containing analytes and chromatography column for sample separation. A back pressure regulator maintains a set elution pressure in the chromatography column. CO2 advantageously remains in liquid phase for elution in the column, thereby avoiding two-phase conditions adversely affecting analyte resolution. An equilibration bypass flow loop may be provided to separate normal sample elution from initial CO2 flow equilibration, thereby allowing rapid exchange of samples with minimal downtime.

Abstract: A system receives a character sequence entered in a search box, identifies a first item that includes the character sequence and a second item that includes the character sequence, identifies a first item set that includes the first item and a second item set that includes the second item; and outputs the first item set and the second item set to a location associated with the search box. The system receives a selection of a third item from the first item set, identifies a third item set that includes the third item and a fourth item set that includes the third item, and outputs the third item set and the fourth item set to the location associated with the search box. The system receives a selection of any item set from the location associated with the search box, and executes a search based on the selected item set.

Abstract: Systems and methods for processing user-submitted updates based on user reliability scores. An update score is determined for an update submitted by a user based on a similarity of a field value provided by the update to corresponding field values in identified records. A user score is determined based on update scores, including the update score, determined for corresponding updates submitted by the user. The update is then processed based on the user score.

Abstract: The technology disclosed relates to improving parallel functional processing using abstractions and methods defined based on category theory. In particular, the technology disclosed provides a range of useful categorical functions for processing large data sets in parallel. These categorical functions manage all phases of distributed computing, including dividing a data set into subsets of approximately equal size and combining the results of the subset calculations into a final result, while hiding many of the low-level programming details. These categorical functions are extraordinarily well-ordered and have a sophisticated type system and type inference, which allows for generating maps and reducing them in an elegant and succinct way using concise and expressive programs that can significantly efficientize a whole software development process.

Abstract: A system creates a graph of nodes connected by arcs, and identifies a first compound attribute associated with contacts purchased by a current user. The first compound attribute includes a first attribute associated with a first value and a second attribute associated with a second value. The system identifies a directed arc from a first node to a second node. The directed arc is associated with a probability that previous users who purchased a first contact associated with the first compound attribute also purchased a second contact associated with a second compound attribute. The second compound attribute includes the first attribute, associated with a third value which matches the first value, and the second attribute, associated with a fourth value, which lacks a match with the second value. The system outputs a recommendation for the current user to purchase contacts associated with the second compound attribute if the probability exceeds a threshold.

Abstract: A system receives an association of first item with first system user, generates first hash value by applying first hash function associated with first system user to first item identifier associated with first item, and sets a bit corresponding to first hash value in array. The system receives an association of second item with second system user, generates second hash value by applying second hash function associated with second user to second item identifier associated with second item, and sets a bit corresponding to second hash value in array. The system receives a request to determine whether third item is associated with first system user, generates third hash value by applying first hash function to third item identifier associated with third item, and outputs message that third item is not associated with first user if a bit corresponding to third hash value is not set in array.

Abstract: A system determines a count of each item in each item set, sorts each count into ascending order, assigns an ascending identifier to each item corresponding to each sorted count, and sorts each identifier in each item set in descending order. The system partitions item sets into a first group of item sets and a second group of item sets, each item set in the first group including a common largest identifier, determines a count for each subset of each item set of the first group, and determines a count of each subset of each item set by summing each count for each subset of each item set of the first group with each corresponding count for each corresponding subset of each item set of the second group. The system outputs a recommended item set based on the count of each subset of each item set.

Abstract: Suggesting query items based on database fields is described. A database system receives a character sequence entered in a search box. The database system identifies a first distribution of first field-based items that include the character sequence, and a second distribution of second field-based items that include the character sequence. The database system identifies a first item based on combining the first distribution with a distribution of queried fields, and a second item based on combining the second distribution with the distribution of queried fields. The database system outputs the first item and the second item to a location associated with the search box. The database system executes a search based on any requested item, in response to receiving a request to search for any item output to the location associated with the search box.

Abstract: Methods and systems are provided for evaluating standing queries against updated contact entries configured as a stream of facts. The method includes resolving the standing queries into an array of rules, each rule having a first and a second condition; sorting one of the facts into a first property and a second property; comparing the first property of the fact to the first condition of each rule in the array of rules to produce a first subset of matching rules; comparing the second property of the fact to the second condition of each rule in the first subset of rules to produce a second subset of matching rules; and reporting at least one of the second subset of rules to an author of the matching rule. The method further includes populating a first hash with indicia of the first subset, and populating a second hash with the second subset.

Abstract: The technology disclosed describes systems and methods for generating feature vectors from resource description framework (RDF) graphs. Machine learning tasks frequently operate on vectors of features. Available systems for parsing multiple documents often generate RDF graphs. Once a set of interesting features to be considered has been established, the disclosed technology describes systems and methods for generating feature vectors from the RDF graphs for the documents. In one example setting, a machine learning system can use generated feature vectors to determine how interesting a news article might be, or to learn information-of-interest about a specific subject reported in multiple articles. In another example setting, viable interview candidates for a particular job opening can be identified using feature vectors generated from a resume database, using the disclosed systems and methods for generating feature vectors from RDF graphs.

Abstract: Contact recommendations based on purchase history are described. A system creates a directed graph of nodes in which at least some of the nodes are connected by directed arcs, wherein a directed arc from a first node to a second node represents a conditional probability that previous users who purchased a first contact also purchased a second contact. The system identifies a set of contacts purchased by a current user. The system estimates a prospective purchase probability based on a historical probability that previous users purchased a specific contact and a related probability that previous users who purchased the specific contact also purchased a contact in the set of contacts, for each candidate contact. The system outputs a recommendation for the current user to purchase a recommended candidate contact based on a corresponding prospective purchase probability.

Abstract: The technology disclosed describes systems and methods for generating feature vectors from resource description framework (RDF) graphs. Machine learning tasks frequently operate on vectors of features. Available systems for parsing multiple documents often generate RDF graphs. Once a set of interesting features to be considered has been established, the disclosed technology describes systems and methods for generating feature vectors from the RDF graphs for the documents. In one example setting, a machine learning system can use generated feature vectors to determine how interesting a news article might be, or to learn information-of-interest about a specific subject reported in multiple articles. In another example setting, viable interview candidates for a particular job opening can be identified using feature vectors generated from a resume database, using the disclosed systems and methods for generating feature vectors from RDF graphs.

Abstract: An error checking technique for database records. A record is selected and its entities are compared with the entities of other records stored in the database to determine a likelihood that the labels associated with the entities of the selected record are correct. The likelihood for each entity of the selected record being correctly labeled can be determined by comparing the number of times that the entity appears in the database records with that label to the number of times that the entity appears in the database records with any other label. If the likelihood does not exceed a threshold, then an error is likely, and action can be taken to correct the record.

Abstract: The technology disclosed relates to methods for partitioning sets of features for a Bayesian classifier, finding a data partition that makes the classification process faster and more accurate, while discovering and taking into account feature dependence among sets of features in the data set. It relates to computing class entropy scores for a class label across all tuples that share the feature-subset and arranging the tuples in order of non-decreasing entropy scores for the class label, and constructing a data partition that offers the highest improvement in predictive accuracy for the data set. Also disclosed is a method for partitioning a complete set of records of features in a batch computation, computing increasing predictive power; and also relates to starting with singleton partitions, and using an iterative process to construct a data partition that offers the highest improvement in predictive accuracy for the data set.