Patents by Inventor Mihaela Ancuta BORNEA
Mihaela Ancuta BORNEA has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).
-
Patent number: 11765193Abstract: In a computer-implemented method for improving a static analyzer output, a processor receives a labeled data set with labeled true vulnerabilities and labeled false vulnerabilities. A processor receives pretrained contextual embeddings from a contextual embeddings model. A processor maps the true vulnerabilities and the false vulnerabilities to the pretrained contextual embeddings model. A processor generates a fine-tuned model with classifications for true vulnerabilities.Type: GrantFiled: December 30, 2020Date of Patent: September 19, 2023Assignee: International Business Machines CorporationInventors: Saurabh Pujar, Luca Buratti, Alessandro Morari, Jim Alain Laredo, Mihaela Ancuta Bornea, Jeffrey Scott McCarley, Yunhui Zheng
-
Patent number: 11755657Abstract: A method, computer program product, and/or computer system generate a first adversarial statement via: (1) receiving a question and an original context for the question; (2) converting the question into a statement with a placeholder answer; (3) picking randomly an answer entity from a training text corpus; (4) replacing the placeholder answer with the randomly picked answer entity; and (5) leaving a correct question entity in the statement. The first adversarial statement is inserted into the original context to form a first adversarial context. The question and the first adversarial context as a first pair and the question and the original context as a second pair are input into a question-answer dialog system to train the question-answer dialog system.Type: GrantFiled: September 19, 2022Date of Patent: September 12, 2023Assignee: International Business Machines CorporationInventors: Sara Rosenthal, Avirup Sil, Mihaela Ancuta Bornea, Radu Florian
-
Publication number: 20230148228Abstract: A method of using a computing device using natural language software to respond to a Boolean question is presented. The method includes receiving, by a computing device, a question and at least one passage. The computing device classifies the question as a Boolean type question. The computing device generates evidence from the at least one passage to the Boolean question. The computing device generates a response to the Boolean question using the generated evidence. The computing device provides the generated evidence that supports the response to the Boolean question.Type: ApplicationFiled: November 5, 2021Publication date: May 11, 2023Inventors: Sara Rosenthal, Jeffrey Scott McCarley, Mihaela Ancuta Bornea
-
Patent number: 11574130Abstract: A method includes receiving, by a question-answer system, a question in a first language and the question in a second language and predicting, by the question-answer system, a first answer to the question in the first language and a second answer to the question in the second language. The method also includes generating, by the question-answer system, a first vector representing the question in the first language and a second vector representing the question in the second language and adjusting the question-answer system based on the first and second answers and the first and second vectors such that when the question-answer system subsequently generates a third vector representing the question in the first language and a fourth vector representing the question in the second language, a distance between the third and fourth vectors is less than a distance between the first and second vectors.Type: GrantFiled: November 24, 2020Date of Patent: February 7, 2023Assignee: International Business Machines CorporationInventors: Mihaela Ancuta Bornea, Lin Pan, Sara Rosenthal, Avirup Sil, Radu Florian
-
Publication number: 20230009893Abstract: A method, computer program product, and/or computer system generate a first adversarial statement via: (1) receiving a question and an original context for the question; (2) converting the question into a statement with a placeholder answer; (3) picking randomly an answer entity from a training text corpus; (4) replacing the placeholder answer with the randomly picked answer entity; and (5) leaving a correct question entity in the statement. The first adversarial statement is inserted into the original context to form a first adversarial context. The question and the first adversarial context as a first pair and the question and the original context as a second pair are input into a question-answer dialog system to train the question-answer dialog system.Type: ApplicationFiled: September 19, 2022Publication date: January 12, 2023Inventors: Sara Rosenthal, Avirup Sil, Mihaela Ancuta Bornea, Radu Florian
-
Patent number: 11520829Abstract: A method, computer program product, and/or computer system protects a question-answer dialog system from being attacked by adversarial statements that incorrectly answer a question. A computing device accesses a plurality of adversarial statements that are capable of making an adversarial attack on a question-answer dialog system, which is trained to provide a correct answer to a specific type of question. The computing device utilizes the plurality of adversarial statements to train a machine learning model for the question-answer dialog system. The computing device then reinforces the trained machine learning model by bootstrapping adversarial policies that identify multiple types of adversarial statements onto the trained machine learning model. The computing device then utilizes the trained and bootstrapped machine learning model to avoid adversarial attacks when responding to questions submitted to the question-answer dialog system.Type: GrantFiled: October 21, 2020Date of Patent: December 6, 2022Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATIONInventors: Sara Rosenthal, Avirup Sil, Mihaela Ancuta Bornea, Radu Florian
-
Publication number: 20220210178Abstract: In a computer-implemented method for improving a static analyzer output, a processor receives a labeled data set with labeled true vulnerabilities and labeled false vulnerabilities. A processor receives pretrained contextual embeddings from a contextual embeddings model. A processor maps the true vulnerabilities and the false vulnerabilities to the pretrained contextual embeddings model. A processor generates a fine-tuned model with classifications for true vulnerabilities.Type: ApplicationFiled: December 30, 2020Publication date: June 30, 2022Inventors: Saurabh Pujar, Luca Buratti, Alessandro Morari, Jim Alain Laredo, Mihaela Ancuta Bornea, Jeffrey Scott McCarley, Yunhui Zheng
-
Publication number: 20220164538Abstract: A method includes receiving, by a question-answer system, a question in a first language and the question in a second language and predicting, by the question-answer system, a first answer to the question in the first language and a second answer to the question in the second language. The method also includes generating, by the question-answer system, a first vector representing the question in the first language and a second vector representing the question in the second language and adjusting the question-answer system based on the first and second answers and the first and second vectors such that when the question-answer system subsequently generates a third vector representing the question in the first language and a fourth vector representing the question in the second language, a distance between the third and fourth vectors is less than a distance between the first and second vectors.Type: ApplicationFiled: November 24, 2020Publication date: May 26, 2022Inventors: Mihaela Ancuta BORNEA, Lin PAN, Sara ROSENTHAL, Avirup SIL, Radu FLORIAN
-
Publication number: 20220121710Abstract: A method, computer program product, and/or computer system protects a question-answer dialog system from being attacked by adversarial statements that incorrectly answer a question. A computing device accesses a plurality of adversarial statements that are capable of making an adversarial attack on a question-answer dialog system, which is trained to provide a correct answer to a specific type of question. The computing device utilizes the plurality of adversarial statements to train a machine learning model for the question-answer dialog system. The computing device then reinforces the trained machine learning model by bootstrapping adversarial policies that identify multiple types of adversarial statements onto the trained machine learning model. The computing device then utilizes the trained and bootstrapped machine learning model to avoid adversarial attacks when responding to questions submitted to the question-answer dialog system.Type: ApplicationFiled: October 21, 2020Publication date: April 21, 2022Inventors: SARA ROSENTHAL, AVIRUP SIL, MIHAELA ANCUTA BORNEA, RADU FLORIAN
-
Patent number: 10949464Abstract: A system for identifying a schema for storing graph data includes a database containing a graph dataset of data and relationships between data pairs and a list of storage methods that each are a distinct structural arrangement of the data and relationships from the graph data set. An analyzer module collects statistics for the graph dataset, and a data classification module uses the collected statistics to calculate metrics describing the data and relationships in the graph dataset, uses the calculated metrics to group the data and relationships into a plurality of graph dataset subsets and associates each graph dataset subset with one of the plurality of storage methods. The resulting group of storage methods associated with the plurality of graph dataset subsets includes a unique storage method for each graph dataset subset. The data and relationships in each graph dataset subset are arranged in accordance with associated storage methods.Type: GrantFiled: March 23, 2016Date of Patent: March 16, 2021Assignee: International Business Machines CorporationInventors: Mihaela Ancuta Bornea, Julian Timothy Dolby, Achille Belly Fokoue-Nkoutche, Anastasios Kementsietsidis, Kavitha Srinivas
-
Patent number: 10509804Abstract: A system for storing graph data as a multi-dimensional cluster having a database with a graph dataset containing data and relationships between data pairs and a schema list of storage methods that use a table with columns and rows associated with data or relationships. An analyzer module to collect statistics of a graph dataset and a dimension identification module to identify a plurality of dimensions that each represent a column in the table. A schema creation and loading module creates a modified storage method and having a plurality of distinct table blocks and a plurality of table block indexes, one index for each table block and arranges the data and relationships in the given graph dataset in accordance with the modified storage method to create the multi-dimensional cluster.Type: GrantFiled: March 24, 2016Date of Patent: December 17, 2019Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATIONInventors: Mihaela Ancuta Bornea, Julian Timothy Dolby, Achille Belly Fokoue-Nkoutche, Anastasios Kementsietsidis, Kavitha Srinivas
-
Patent number: 10360262Abstract: Various embodiments of the invention relate to optimizing storage of schema-less data. At least one of a schema-less dataset including a plurality of resources one or more query workloads associated with the plurality of resources is received. Each resource is associated with at least a plurality of properties. At least one set of co-occurring properties from the plurality of properties is identified. A graph including a plurality of nodes is generated. Each of the nodes represents a unique property in the set of co-occurring properties. The graph further includes an edge connecting each node representing a pair of co-occurring properties. A schema is generated based on the graph that assigns a column identifier from a table to each unique property represented by one of the nodes in the graph.Type: GrantFiled: June 23, 2017Date of Patent: July 23, 2019Assignee: International Business Machines CorporationInventors: Mihaela Ancuta Bornea, Julian Dolby, Achille Belly Fokoue-Nkoutche, Anastasios Kementsietsidis, Kavitha Srinivas
-
Patent number: 10229200Abstract: Data elements from data sources and having a data value set are linked by using hash functions to determine a dimensionally reduced instance signature for each data element based on all data values associated with that data element to yield a plurality of dimensionally reduced instance signatures of equivalent fixed size such that similarities among the data values in the data value sets across all data elements is maintained among the plurality of instance signatures. Candidate pairs of data elements to link are identified using the plurality of instance signatures in locality sensitive hash functions, and a similarity index is generated for each candidate pair using a pre-determined measure of similarity. Candidate pairs of data elements having a similarity index above a given threshold are linked.Type: GrantFiled: June 8, 2012Date of Patent: March 12, 2019Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATIONInventors: Mihaela Ancuta Bornea, Songyun Duan, Achille Belly Fokoue-Nkoutche, Oktie Hassanzadeh, Anastasios Kementsietsidis, Kavitha Srinivas, Michael James Ward
-
Publication number: 20170286566Abstract: Various embodiments of the invention relate to optimizing storage of schema-less data. At least one of a schema-less dataset including a plurality of resources one or more query workloads associated with the plurality of resources is received. Each resource is associated with at least a plurality of properties. At least one set of co-occurring properties from the plurality of properties is identified. A graph including a plurality of nodes is generated. Each of the nodes represents a unique property in the set of co-occurring properties. The graph further includes an edge connecting each node representing a pair of co-occurring properties. A schema is generated based on the graph that assigns a column identifier from a table to each unique property represented by one of the nodes in the graph.Type: ApplicationFiled: June 23, 2017Publication date: October 5, 2017Applicant: International Business Machines CorporationInventors: Mihaela Ancuta BORNEA, Julian DOLBY, Achille Belly FOKOUE-NKOUTCHE, Anastasios KEMENTSIETSIDIS, Kavitha SRINIVAS
-
Patent number: 9715560Abstract: Various embodiments of the invention relate to optimizing storage of schema-less data. At least one of a schema-less dataset including a plurality of resources one or more query workloads associated with the plurality of resources is received. Each resource is associated with at least a plurality of properties. At least one set of co-occurring properties from the plurality of properties is identified. A graph including a plurality of nodes is generated. Each of the nodes represents a unique property in the set of co-occurring properties. The graph further includes an edge connecting each node representing a pair of co-occurring properties. A schema is generated based on the graph that assigns a column identifier from a table to each unique property represented by one of the nodes in the graph.Type: GrantFiled: June 27, 2013Date of Patent: July 25, 2017Assignee: International Business Machines CorporationInventors: Mihaela Ancuta Bornea, Julian Dolby, Achille Belly Fokoue-Nkoutche, Anastasios Kementsietsidis, Kavitha Srinivas
-
Patent number: 9535950Abstract: A semantic query over an RDF database is received with RDF database statistics and access methods for evaluating triple patterns in the query. The semantic query is expressed as a parse tree containing triple patterns and logical relationships among the triple patterns. The parse tree and access methods create a data flow graph containing a plurality of triple pattern and access method pair nodes connected by a plurality of edges, and an optimal flow tree through the data flow graph is determined such that costs are minimized and all triple patterns in the semantic query are contained in the optimal flow tree. A structure independent execution tree defining a sequence of evaluation through the optimal flow tree is created and is transformed into a database structure dependent query plan. This is used to create an SQL query that is used to evaluate the semantic query over the RDF database.Type: GrantFiled: January 27, 2016Date of Patent: January 3, 2017Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATIONInventors: Mihaela Ancuta Bornea, Julian Timothy Dolby, Anastasios Kementsietsidis, Kavitha Srinivas
-
Publication number: 20160203236Abstract: A system for identifying a schema for storing graph data includes a database containing a graph dataset of data and relationships between data pairs and a list of storage methods that each are a distinct structural arrangement of the data and relationships from the graph data set. An analyzer module collects statistics for the graph dataset, and a data classification module uses the collected statistics to calculate metrics describing the data and relationships in the graph dataset, uses the calculated metrics to group the data and relationships into a plurality of graph dataset subsets and associates each graph dataset subset with one of the plurality of storage methods. The resulting group of storage methods associated with the plurality of graph dataset subsets includes a unique storage method for each graph dataset subset. The data and relationships in each graph dataset subset are arranged in accordance with associated storage methods.Type: ApplicationFiled: March 23, 2016Publication date: July 14, 2016Applicant: International Business Machines CorporationInventors: Mihaela Ancuta Bornea, Julian Timothy Dolby, Achille Belly Fokoue-Nkoutche, Anastasios Kementsietsidis, Kavitha Srinivas
-
Publication number: 20160203206Abstract: A system for storing graph data as a multi-dimensional cluster having a database with a graph dataset containing data and relationships between data pairs and a schema list of storage methods that use a table with columns and rows associated with data or relationships. An analyzer module to collect statistics of a graph dataset and a dimension identification module to identify a plurality of dimensions that each represent a column in the table. A schema creation and loading module creates a modified storage method and having a plurality of distinct table blocks and a plurality of table block indexes, one index for each table block and arranges the data and relationships in the given graph dataset in accordance with the modified storage method to create the multi-dimensional cluster.Type: ApplicationFiled: March 24, 2016Publication date: July 14, 2016Applicant: International Business Machines CorporationInventors: Mihaela Ancuta Bornea, Julian Timothy Dolby, Achille Belly Fokoue-Nkoutche, Anastasios Kementsietsidis, Kavitha Srinivas
-
Publication number: 20160140172Abstract: A semantic query over an RDF database is received with RDF database statistics and access methods for evaluating triple patterns in the query. The semantic query is expressed as a parse tree containing triple patterns and logical relationships among the triple patterns. The parse tree and access methods create a data flow graph containing a plurality of triple pattern and access method pair nodes connected by a plurality of edges, and an optimal flow tree through the data flow graph is determined such that costs are minimized and all triple patterns in the semantic query are contained in the optimal flow tree. A structure independent execution tree defining a sequence of evaluation through the optimal flow tree is created and is transformed into a database structure dependent query plan. This is used to create an SQL query that is used to evaluate the semantic query over the RDF database.Type: ApplicationFiled: January 27, 2016Publication date: May 19, 2016Applicant: International Business Machines CorporationInventors: Mihaela Ancuta Bornea, Julian Timothy Dolby, Anastasios Kementsietsidis, Kavitha Srinivas
-
Patent number: 9323825Abstract: A system for storing graph data as a multi-dimensional cluster having a database with a graph dataset containing data and relationships between data pairs and a schema list of storage methods that use a table with columns and rows associated with data or relationships. An analyzer module to collect statistics of a graph dataset and a dimension identification module to identify a plurality of dimensions that each represent a column in the table. A schema creation and loading module creates a modified storage method and having a plurality of distinct table blocks and a plurality of table block indexes, one index for each table block and arranges the data and relationships in the given graph dataset in accordance with the modified storage method to create the multi-dimensional cluster.Type: GrantFiled: August 14, 2013Date of Patent: April 26, 2016Assignee: International Business Machines CorporationInventors: Mihaela Ancuta Bornea, Julian Timothy Dolby, Achille Belly Fokoue-Nkoutche, Anastasios Kementsietsidis, Kavitha Srinivas