International Business Machines Corporation

Ioana Roxana Stanoi

Christian Alexander Lang

George Andrei Mihaila

Ying Chen

Yuan-Chi Chang

Jun-Jang Jeng

John Richard Smith

Philip Shilung Yu

Jimeng Sun

Mo Yu

Feifei Li

Spyridon Papadimitriou

Themis Palpanas

Su Yan

Guy Maring Lohman

Richard Sefton Sidle

William Scott Spangler

Bishwaranjan Bhattacharjee

Sriram Kolinjivadi Padmanabhan

Yunyao Li

Milind R Naphade

Shivakumar Vaithyanathan

Timothy Ray Malkemus

Linda Ha Kato

Apostol Ivanov Natsev

Meenakshi Nagarajan

Jacques Labrie

Shahram Ebadollahi

Robbert Van Der Linden

Lexing Xie

Jakka Sairamesh

Sweefen Goh

Olivier Lichtarge

Jelena Tesic

Angela D Wilkins

Tyler Shore Baldwin

Uttam Jain

Lawrence A Donehower

Quanzhong Li

Benjamin J Bachman

Sam J Regenbogen

Curtis R Pickering

Bogdan Alexe

Thomas Fanghaenel

Anbu Karani Adikesavan

Maria E Terron-Diaz

A set of documents is parsed. Members of the set of documents include a set of text elements and a set of visual elements. A text content stream based on the set of text elements and a visual content stream based on the set of visual elements are produced. For respective documents, a set of respective visual element summarizations is built from the visual content stream. Each visual summarization includes a textual description of a respective visual element in the respective document. A holistic index is created by indexing the text content from the text content stream and the text descriptions of the visual elements for each document in a single search index. The indexing uses a set of semantic relationships between the text content from the text content stream and the textual descriptions of the visual elements. A user interface allows a user to selectively search text content and visual content.

Holistic document search

A set of documents is parsed. Members of the set of documents include a set of text elements and a set of visual elements. A text content stream based on the set of text elements and a visual content stream based on the set of visual elements are produced. For respective documents, a set of respective visual element summarizations is built from the visual content stream. Each visual summarization includes a text description of a respective visual element in the respective document. A holistic index is created by indexing the text content from the text content stream and the text descriptions of the visual elements in a single search index. The indexing uses a set of semantic relationships between the text content from the text content stream and the textual descriptions of the visual elements. A user interface allows a user to selectively search text content and visual content.

A method, system, and computer program product for event detection using roles and relationships of entities are provided in the illustrative embodiments. A training event and a set of entities participating in the training event are identified in a training data. For a first entity in the set of entities, a first role occupied by the entity in the event is determined. A behavior attribute is assigned to the first role. A relationship of the first role with a second role corresponding to a second entity in the set of entities is determined. An event rule is constructed to detect an event corresponding to the training event in new data and comprising a plurality of roles, behavior attributes, and the relationship. The plurality of roles includes the first role and the second role, and the plurality of behavior attributes includes the behavior attribute assigned to the first role.

Event detection using roles and relationships of entities

Embodiments of the invention relate to ambiguity detection. In one embodiment, an object and a topical domain associated with the object are obtained. In this embodiment, the object includes at least one term. At least one of a plurality of information sources is analyzed based on the at least one term and the topical domain. A determination is made that object is one of ambiguous and unambiguous based on analyzing at least one of the plurality of information sources.

Knowledge-rich automatic term disambiguation

Embodiments of the invention relate to a multi-level database compression technique to compress table data objects stored in pages. A compact dictionary structure is encoded that represents frequent values of data at any level of granularity. More than one level of compression is provided, wherein input to a finer level of granularity is an output of a coarser level of granularity. Based upon the encoded dictionary structure, a compression technique is applied to a stored page to compress each row on the page. Similarly, a de-compression technique may be applied to decompress the compressed data, utilizing the same dictionary structures at each level of granularity.

Multi-level database compression

Illustrative embodiments provide a computer-implemented method for hybrid push/pull of continuous structured query language queries. The computer-implemented method receives stream input, wherein the stream input comprises events of interest, builds and a state machine and stream plans, based on an original query, and replicates the stream input. Responsive to a push sub-query trigger, the computer-implemented method submits a pull sub-query to the database to produce a result, and sends the result to a requester.

Hybrid push/pull execution of continuous SQL queries

A system, method, and computer readable medium for reducing message flow on a message bus are disclosed. The method includes determining if at least one logical operator in a plurality of logical operators requires processing on a given physical processing node in a group of physical nodes. The logical operator is pinned to the given physical processing node. The pinning prevents any subsequent reassignment of the logical operator to another physical processing node. Each logical operator in the plurality of logical operators is assigned to an initial physical processing node in the group of physical processing nodes on a message bus. A determination is made as to whether at least one logical operating in the plurality of logical operators needs to be reassigned to a different physical processing node. The at least one logical operator is reassigned to the different physical processing node.

Reduction of message flow between bus-connected consumers and producers

A system, method, and computer readable medium for reducing message flow on a message bus are disclosed. The method includes determining if at least one logical operator in a plurality of logical operators requires processing on a given physical processing node in a group of physical nodes. In response to determining that the logical operator requires processing on the given physical processing node, the logical operator is pinned to the given physical processing node. Each logical operator in the plurality of logical operators is assigned to an initial physical processing node in the group of physical processing nodes on a message bus.

Providing consistency guarantees in a data stream processing engine is provided. Consistency tracking information is attached to data streams coming into the data stream processing engine. The consistency tracking information is propagated through a plurality of streaming operators that process the data streams within the data stream processing engine. Then, the propagated consistency tracking information is used to detect a consistent state in an output stream.

Providing consistency in processing data streams

Techniques are provided for enabling execution of a process employing a cache Method steps can include obtaining a first probability of accessing a given artifact in a state S, obtaining a second probability of using a predicate from a current state Sin the state S, determining a benefit of prefetching the given artifact using the predicate based on at least the first probability and the second probability, and whether and/or when a cache replacement should be conducted, based at least on the benefit determined.

Method and apparatus for execution of a process

A system, method, and computer readable medium for optimizing throughput of a stream processing system are disclosed. The method comprises analyzing a set of input streams and creating, based on the analyzing, an input profile for at least one input stream in the set of input streams. The input profile comprises at least a set of processing requirements associated with the input stream. The method also comprises generating a search space, based on an initial configuration, comprising a plurality of configurations associated with the input stream. A configuration in the plurality of configurations is identified that increases throughput more than the other configurations in the plurality of configurations based on at least one of the input profile and system resources.

Maximization of sustained throughput of distributed continuous queries

The present invention is directed to a method and apparatus for assisting in rating and filtering multimedia content, such as images, videos and sound recordings. One embodiment comprises a computer implemented method for rating the objectionability of specified digital content that comprises one or more discrete content items, wherein the method includes the step of moving the specified content to one or more filtering stages in a succession of filtering stages. After the specified content is moved to a given one of the filtering stages, a rating procedure is carried out to determine whether a rating can be applied to one or more of the content items, and if so, a selected rating is applied to each of the one or more content items. The method further comprises moving content items of the specified content to the next stage in the succession after the given stage, when at least one content item of the specified content remains without rating, after the rating procedure at the given stage. When none of the content items of the specified content remains without a rating after the rating procedure has been completed at the given stage, ratings that have been respectively applied to at least some of the content items are selectively processed, in order to determine an overall objectionability rating for the specified content.

Method and apparatus for semantic assisted rating of multimedia content

Disclosed is a method, information processing system, and computer readable medium for preserving privacy of nonstationary data streams. The method includes receiving at least one nonstationary data stream with time dependent data. Calculating, for a given instant of sub-space of time, A set of first-moment statistical values is calculated, for a given instant of sub-space of time, for the data. The first moment statistical values include a principal component for the sub-space of time. The data is perturbed with noise along the principal component in proportion to the first-moment of statistical values so that at least part of a set of second-moment statistical values for the data is perturbed by the noise only within a predetermined variance.

Preserving privacy of one-dimensional data streams using dynamic correlations

A method, information processing system, and computer readable medium are provided for preserving privacy of one-dimensional nonstationary data streams. The method includes receiving a one-dimensional nonstationary data stream. A set of first-moment statistical values are calculated, for a given instant of sub-space of time, for the data. The first moment statistical values include a principal component for the sub-space of time. The data is perturbed with noise along the principal component in proportion to the first-moment of statistical values so that at least part of a set of second-moment statistical values for the data is perturbed by the noise only within a predetermined variance.

Preserving privacy of one-dimensional data streams by perturbing data with noise and using dynamic autocorrelation

Growing community of inventors

San Jose, CA, United States of America

Ioana Roxana Stanoi