International Business Machines Corporation

Roberto Guarda

Antonio Castellucci

Vittorio Carullo

Alessandro Rea

Roberto Ragusa

Marzia E Castellani

Ciro Ragusa

Aniello Alessandro Rea

Emanuele Vercalli

Michele Iovino

Gianluca Pulcini

Glauco Cenciotti

An embodiment of the invention may include a method, computer program product and system for processing a natural language query. An embodiment may include receiving text of the natural language query. An embodiment may include extracting a set of features from the text through natural language processing. An embodiment may include generating a structured query based on the set of features. An embodiment may include normalizing the text to create a normalized natural language query. An embodiment may include executing a search of a corpus via the structured query and the normalized natural language query. An embodiment may include returning results of the search.

Increasing pertinence of search results within a complex knowledge base

Tracking position of at least one object in video frames. The tracking includes processing an initial frame of a set of frames, the processing the initial frame using feature extraction to identify locations of features of the at least one object. The tracking further includes using motion estimation to track locations of the features in subsequent frames of the set of frames, including iteratively performing: obtaining a next frame of the set of frames, and applying a motion estimation algorithm as between the next frame and a prior frame of the set of frames to identify updated locations of the features in the next frame, where locations of the features as identified based on the prior frame are used as input to the motion estimation algorithm to identify the updated locations of the features in the next frame based on searching less than an entirety of the next frame.

Object position tracking using motion estimation

In an approach for locating parking, a computer identifies driving directions to a destination. The computer receives a navigation input from a user to locate parking associated with the destination. The computer calculates one or more navigation paths based on the destination, wherein the one or more navigation paths include one or more street segments that connect identical starting and ending points with at least one different street segment and overall values indicating priority based on one or more weight factors associated with the one or more street segments. The computer selects a first navigation path from the calculated one or more navigation paths utilizing an ordered ranking of the one or more navigation paths. The computer provides driving directions associated with the first navigation path. The computer determines whether the user locates parking while utilizing the first navigation path.

Identifying a parking location with respect to a destination

A not crossable element (NCE) is marked using a shape on a map. A map data does not prevent a route from crossing the NCE. A closed area is selected on the map, which includes a polygon bound on one side by a portion of the shape. A set of mapping points (MP) that are included in the closed area is selected. A route is formed by connecting one MP to another MP. A subset of MPs is identified as a set of candidate MPs (CMP). Each CMP lies within a search distance of a not-mapping point (NMP). The CMPs are sorted according to CMP-NMP distances. A subset of CMPs is selected whose paths to the NMP do not intersect with any shape corresponding to any NCE in the closed area. The NMP is linked in the route with a selected CMP from the subset of CMPs.

Geographical constraints-aware map routing

A method includes tracking positions of object(s) in video frames, including: processing an initial frame of a set of frames of the video frames using feature extraction to identify locations of features of the object(s), obtaining a next frame of the set and applying a motion estimation algorithm as between the next frame and a prior frame to identify updated locations of the features in the next frame, where locations of the features as identified based on the prior frame are used as input to the motion estimation algorithm to identify the updated locations of the features in the next frame based one searching less than an entirety of the next frame. The tracking further includes recognizing occurrence of an event, halting the iteratively performing, and repeating, for at least one subsequent set of frames, the processing an initial frame and the using motion estimation.

A processor receives a media rating associated with a first media program and metadata associated with the first media program. The processor stores the metadata and the media rating associated with the first media program in a local database. The processor receives a second media program absent a media rating. The processor determines metadata associated with the second media program absent the media rating. The processor searches the local database based on the metadata associated with the second media program absent the parental control rating. The processor matches the metadata associated with the second media program absent the media rating to the metadata of the first media program including the media rating, and the processor applies the media rating of the first media program to the second media program.

Determining missing media control information based on previous media transmissions

The invention provides a system and method for retrieving documents from a collection of documents that match a word search query. A word index is generated for each document in which each entry is an enriched-term string built from the stemmed form of the word to be searched and a separator character followed by the original form of the word to be searched. During a retrieving operation, a search query is processed depending the original form or the stemmed form of a word to be searched. Cross-documents tables are addressed to find documents that match the enriched-term string of the word to be searched.

System, method and computer program product for documents retrieval

Embodiments of the invention may include a method for searching digital content in a data processing system. The method may include providing a set of sample digital resources. Each sample digital resource may be associated with metadata describing its content, including fields having associated metadata values. A user may select at least one sample digital resource from the set. One or more metadata values of the sample digital resource may be displayed to the user. The user may then select at least a portion of the metadata values. A digital resource having one or more metadata values substantially matching the selected metadata value of the sample digital resource may then be retrieved.

Digital content searching tool

A solution is provided for accessing a shared resource in a data processing system (such as a server of a DAM application) by a plurality of exploiter entities (such as clients). A corresponding method starts with the step of monitoring an activity of each enabled entity being granted an access to the shared resource. The access granted to each enabled entity is released in response to the reaching of a first threshold (such as a time-out period) by an inactivity indicator, which is indicative of the non-activity of the enabled entity (such as an idle time). The method continues by detecting a critical condition of the shared resource (such as after reaching a maximum number of concurrent accesses). The access granted to an eligible enabled entity (or more) is released in response to the critical condition; the inactivity indicator of each eligible enabled entity reaches a second threshold preceding the first threshold (such as a grant period lower than the time-out period).

Growing community of inventors

Pomezia, Italy

Roberto Guarda