公开(公告)号：US20240220538A1

公开(公告)日：2024-07-04

申请号：US18604990

申请日：2024-03-14

申请人： Google LLC

发明人： Martin Probst

IPC分类号： G06F16/84 ,  G06F16/21 ,  G06F16/83 ,  G06F16/835 ,  G06F16/33

CPC分类号： G06F16/86 ,  G06F16/213 ,  G06F16/83 ,  G06F16/835 ,  G06F16/33

摘要： Methods, systems, and apparatus, including computer programs encoded on computer storage media, for storing semi-structured data. One of the methods includes maintaining a plurality of schemas; receiving a first semi-structured data item; determining that the first semi-structured data item does not match any of the schemas in the plurality of schemas; and in response to determining that the first semi-structured data item does not match any of the schemas in the plurality of schemas: generating a new schema, encoding the first semi-structured data item in the first data format to generate the first new encoded data item in accordance with the new schema, storing the first new encoded data item in the data item repository, and associating the first new encoded data item with the new schema.

公开(公告)号：US11887014B2

公开(公告)日：2024-01-30

申请号：US16047223

申请日：2018-07-27

申请人： SAP SE

发明人： Jeremiah Reeves ,  Nithya Rajagopalan ,  Abhishek Chaturvedi ,  Sunil Gornalle ,  Prasad Karani ,  Surendranath Gopinathan ,  Gurudayal Khosla

IPC分类号： G06F16/00 ,  G06N5/04 ,  G06Q30/0601 ,  G06N3/006 ,  G06N20/00 ,  G06F16/27 ,  G06F16/81 ,  G06F16/838 ,  G06F16/835

CPC分类号： G06N5/04 ,  G06F16/27 ,  G06F16/81 ,  G06F16/838 ,  G06F16/8373 ,  G06N3/006 ,  G06N20/00 ,  G06Q30/0625

摘要： In an example embodiment, a request is received, via a graphical user interface, to add a new object to a directory of objects, the new object having a first category in a hierarchical taxonomy of categories and objects. Then one or more questions previously assigned to the first category and/or one or more existing objects within the first category are retrieved. Each of the retrieved one or more questions and information about the new object are then fed into a first machine learned model trained to output a probability that a question is applicable to an object. One or more questions are generated for the new object based on the probability for each of the retrieved one or more questions. At least one of the one or more generated questions is then assigned to the new object.

公开(公告)号：US11676070B1

公开(公告)日：2023-06-13

申请号：US17111477

申请日：2020-12-03

申请人： Momentum NA, Inc.

发明人： Jason Alan Snyder ,  Manuel De Araujo Pedreira Neto ,  Elena Klau Silverman ,  Stephen Michael Gorman ,  Michael Clark

IPC分类号： G06N20/00 ,  G06F16/835 ,  G06F16/242

CPC分类号： G06N20/00 ,  G06F16/2433 ,  G06F16/835

摘要： The Multidimensional Machine Learning Data and User Interface Segment Tagging Engine Apparatuses, Methods and Systems (“MLUI”) transforms ambient condition data, sales data, user interface selections, cognitive intelligence question input inputs via MLUI components into project projections, campaigns, user interface visualizations, cognitive intelligence question output outputs. A cognitive intelligence (CI) datapoint identifier cache datastructure is generated, the CI datapoint identifier cache datastructure configured to comprise a category identifier and an entity segment identifier. A CI datapoint value cache datastructure is generated, the CI datapoint value cache datastructure configured to comprise a set of module datastructures, each module datastructure corresponding to a module identifier associated with the category identifier, each module datastructure comprising a set of metric datastructures, each metric datastructure corresponding to a set of calculated metrics. The generated CI datapoint identifier cache datastructure and the generated CI datapoint value cache datastructure are stored as a key-value pair.

公开(公告)号：US11663213B2

公开(公告)日：2023-05-30

申请号：US17105014

申请日：2020-11-25

申请人： Cloudera, Inc.

发明人： Alexander Behm ,  Mostafa Mokhtar

IPC分类号： G06F16/00 ,  G06F16/2453 ,  G06F16/2458 ,  G06F16/835

CPC分类号： G06F16/24545 ,  G06F16/2471 ,  G06F16/24547 ,  G06F16/8373 ,  G06F16/24549

摘要： The problem of distinct value estimation has many applications, but is particularly important in the field of database technology where such information is utilized by query planners to generate and optimize query plans. Introduced is a novel technique for estimating the number of distinct values in a given dataset without scanning all of the values in the dataset. In an example embodiment, the introduced technique includes gathering multiple intermediate probabilistic estimates based on varying samples of the dataset, 2) plotting the multiple intermediate probabilistic estimates against indications of sample size, 3) fitting a function to the plotted data points, and 4) determining an overall distinct value estimate by extrapolating the objective function to an estimated or known total number of values in the dataset.

公开(公告)号：US20230150687A1

公开(公告)日：2023-05-18

申请号：US17525690

申请日：2021-11-12

申请人： Rockwell Collins, Inc.

发明人： Jason L. Wong

IPC分类号： B64D43/00 ,  G06F3/14 ,  G06F3/0482 ,  G06F40/143 ,  G06T11/60 ,  G01C23/00 ,  G06F16/84 ,  G06F16/835

CPC分类号： B64D43/00 ,  G06F3/14 ,  G06F3/0482 ,  G06F40/143 ,  G06T11/60 ,  G01C23/00 ,  G06F16/84 ,  G06F16/8373

摘要： A system and method for flight chart element searching is disclosed. A host computing device is configured to convert PDF flight chart files to SVG flight chart files defined in XML, detect flight chart elements in the SVG flight chart files, convert the SVG flight chart files to flight charts defined in aircraft display hardware directives, and combine the flight charts and flight chart element data into a flight chart database. An aircraft computing device is configured to display a flight chart and a GUI using an aircraft display, and highlight a flight chart element in response to a user searching for and selecting a flight chart element name.

公开(公告)号：US11609911B2

公开(公告)日：2023-03-21

申请号：US16720481

申请日：2019-12-19

申请人： Ocient Holdings LLC

发明人： Jason Arnold

IPC分类号： G06F16/2453 ,  G06F16/835

摘要： A method for execution by a query processing module includes determining a query expression indicating a query for execution. An operator tree is generated based on a nested ordering of a plurality of operators indicated by the query expression. Conjunctive normal form (CNF) conversion cost data is generated based on the operator tree, and disjunctive normal form (DNF) conversion cost data is also generated based on the operator tree. Conversion selection data is generated based on the CNF conversion cost data and the DNF conversion cost data. The conversion selection data indicates a selection to perform either a CNF conversion or a DNF conversion. A normalized query expression is generated by performing either the CNF conversion or the DNF conversion upon the query expression based on the conversion selection data. Execution of the query is facilitated in accordance with the normalized query expression.

公开(公告)号：US11481439B2

公开(公告)日：2022-10-25

申请号：US17129085

申请日：2020-12-21

申请人： oracle international corporation

发明人： Zhen Hua Liu ,  Ying Lu ,  Hui Joe Chang

IPC分类号： G06F16/835

摘要： Techniques are described to improve query evaluation in computer systems. In an embodiment, a system receives a full text query for evaluation against a collection of hierarchically marked data object sets. The query specifies token(s) and context(s) which indicate hierarchical location(s) to match within a queried hierarchical data structure. To evaluate the query, the system determines a) data object set(s) that contain the query specified token(s) using token list(s), and/or b) data object set(s) that contain the query specified context(s) using label list(s).

公开(公告)号：US20220156796A1

公开(公告)日：2022-05-19

申请号：US17665840

申请日：2022-02-07

申请人： Persona IP Licensing, LLC

发明人： James Eustace ,  Kim Go ,  Michael Hovey

IPC分类号： G06Q30/02 ,  G06F16/835 ,  H04L67/02

摘要： A method of serving targeted content messages for display in a website accessed in a browser program of a networked computer communicatively connected to a network at a network address for communications, delivers uniquely targeted content messages displayed in websites viewed in web browsers. The method includes placing a script device in a website file, processing the website file, together with the script device by a particular web browser on download of the website file, including by determining the network address of the networked computer, determining an identifier of the website file, and sending an artifact representing the network address and the identifier over the network to a server computer. The method also includes detecting the network address and the identifier by the server computer, querying a database for a database article related to the network address and the identifier, constructing a script program stored in memory of the server computer for the particular browser and website file, and constructing an ad device stored in memory of the web browser device from the script program. The method further includes calling the server computer by the ad device by communication of an identifier representing an action of the web browser device, receiving the identifier by the server computer, querying the database for a select message artifact related to the script program, the identifier, the website file, and the web browser, and responding by the server computer to the web browser with the select message artifact. A message represented by the select message artifact is displayed in the website then viewed in a browser window of the web browser. Messages can be prioritized and are uniquely targeted in content, based on real-time activities of the web browser.

公开(公告)号：US20220156795A1

公开(公告)日：2022-05-19

申请号：US17529814

申请日：2021-11-18

申请人： Persona IP Licensing, LLC

发明人： James Eustace ,  Michael Hovey

IPC分类号： G06Q30/02 ,  H04W4/21 ,  G06F16/835

摘要： A method for identifying segments of a population of user devices communicating on a communications network. The segments correspond to user devices of the population exhibiting comparable behavioral patterns detectable by the communications network. A plurality of marketing systems are accessible on the communications network, and each of the plurality of marketing systems include respective use data corresponding to respective ones of the population for the marketing system. The method includes retrieving by a processor the respective use data for the population, from the plurality of marketing systems, determining by the processor if the respective use data exceeds a threshold for particular behavioral pattern of interest, for the respective use data, determining by the processor a unique identifier for each user device of the use data, grouping by the processor in a database, the respective use data in relation to the unique identifier, for each user device of the use data that exceeds the threshold, and mapping by the processor in the database, the behavioral pattern of the respective use data for each user device of the use data that exceeds the threshold. Behavioral patterns are determined for the respective segment, and related to the user devices of the segment. Content for delivery to the segment is sequenced, and placeholder in the sequence is stored in relation to each user device of the segment, to ensure that each next sequential content is delivered to the respective user device.

公开(公告)号：US11294874B2

公开(公告)日：2022-04-05

申请号：US16521934

申请日：2019-07-25

申请人： Ab Initio Technology LLC

发明人： Taro Ikai

IPC分类号： G06F16/22 ,  G06F16/81 ,  G06F16/835 ,  G06F16/245

摘要： An approach to structured record retrieval permits transmission and storage of records in a native concise format, without requiring that the records be interpreted and stored in a tabular form. Such storage of the records in a tabular form might double the space required, and more generally, requires substantially more space in applications in which there are many optional elements. In some embodiments, each message is parsed according to a specification of the message structure (e.g., according to a “grammar” for the message), and during parsing field values in predefined positions in the structure are extracted and added to an index structure that associates record identifiers with the (position, value) pairs.