International Business Machines Corporation

Amazon Technologies, Inc.

Audible, Inc.

Geetika Tewari Lakshmanan

Lisa Seacat DeLuca

Dana Lea Price

Shelbee D Smith-Eigenbrode

Martin Oberhofer

Michael Muller

Szabolcs Rozsnyai

Fei Wang

Rania Yousef Khalaf

Lydia Mai Do

Francisco Phelan Curbera

Paul Turquand Keyser

Robert J Schloss

Puneet Girdhar

Yuri G Rabinovich

Mitch Ray Kendall

Yurdaer Nezihi Doganata

Jianying Hu

Robert Evan Strom

Matthew Jason Duftler

Steven J Demuth

Philippe Bonnard

Davood Shamsi

Clifford A Pickover

Maja Vukovic

Wei Tan

Chitra Dorai

Rong Nickle Chang

Vinod Muthusamy

Aleksander A Slominski

Songyun Duan

James Thomas Klosowski

Merve Unuvar

Aman Sagar

Dikla Treves

Michael Angelo Macchione, Iii

Dan Stefanescu

Yale Li

A method for presenting search query results is provided. The method may include detecting an occurrence of the trigger event. The method may include determining a category of information based on data associated with the trigger event. The method may include identifying at least one constraint based on the determined category of information. The method may include appending to the identified at least one constraint to the determined category of information. The method may include generating at least one search query. The method may include selecting at least one candidate website based on the category of information. The method may include performing the at least one search query on the at least one candidate website. The method may include filtering each search query result within the search query results. The method may include sending each filtered search query result within the search query results to a user.

Presenting anticipated user search query results prompted by a trigger

Methods and systems for pruning process execution logs include learning a predictive model from a set of execution traces that characterize a process, where the predictive model determines a likelihood of a given instance reaching a specified outcome; identifying attributes in the predictive model that fall below a threshold measure of relevance to the specified outcome using a processor; and removing the identified attributes from the set of execution traces.

Pruning process execution logs

Described herein are techniques for providing a dictionary capabilities associated with foreign language learning content items. In some cases, a foreign language learning application operation on a device may cause definitions associated with a word to be displayed on a display of the device concurrently with the word and at least some text surrounding the word. In some cases, the definitions may be provided in the language of the content item and in a language native the user.

Foreign language learning dictionary system

Methods, systems and apparatus for identifying and utilizing content synchronization information are described. Methods for creating and utilizing a synchronization file that can be used to notify users when synchronization availability exists, and to identify ways to repair synchronization unavailability are described. The synchronization file includes a complete set of individual content elements that make up a content item, such as a book, where each content element has an indication whether play-back of an audio version of the content item can be synchronized with the presentation of a corresponding text version of the content item. Methods, systems and apparatus are also disclosed for providing the synchronization availability information to the user in a manner that enables the user to increase the benefits of synchronized presentation of the content item.

Identifying and utilizing synchronized content

A method of inferring intent in a hybrid network includes monitoring communications in the hybrid network between a plurality of members, triggering an estimation of an intent of one or more members of the hybrid network, estimating the intent, determining a confidence level of the intent, and triggering an action based on the confidence level.

Answer support systems for hybrid networks

Features are disclosed for determining the vocabulary of a user and identifying content items appropriate for the user based on the user's personal vocabulary. The user's vocabulary can be determined by analyzing user-generated textual items. Based on the analysis of such user-generated textual items, a list of words used frequently by the user in the user's own writings can be identified as being in the user's vocabulary. The list of words in the user's vocabulary can be compared to the words in various content items to determine a degree to which the words used in the content are in the user's vocabulary. Content can then be recommended or otherwise determined to be appropriate for the user's vocabulary, identified as challenging, too difficult, or too easy, etc.

Vocabulary determination and vocabulary-based content recommendations

Methods and computer program products relate to indicating search result relevancy that includes accepting a first set of query terms from a user, providing the user with a first set of search results corresponding to the first set of query terms, accepting from the user at least one indication of result relevancy from the first set of search results, accepting a second set of query terms from the user, the second set of query terms including the same or similar query terms to the first set of query terms, providing the user with a second set of search results corresponding to the second set of query terms, the second set of search results including one or more search results from the first set of search results, and including the indication of result relevancy in the second set of search results.

Search result optimization based on previous solutions

Methods and systems for refining a process model include determining whether the process model is too dense or too sparse. A predictive model is learned from execution traces to predict an outcome. The predictive model is modified responsive to the determination of whether the process model is too dense or too sparse. A refined process model is refined from updated traces based on attributes present in the modified predictive model.

Iterative refinement of pathways correlated with outcomes

Techniques for customizing a display of content related to dictionary entries are described herein. In one example, a user selects a word from a rendered electronic book. In response, the device outputs a dictionary user interface that includes display elements associated with a particular user. Preferences of a user may be utilized to determine what content (e.g., what display elements) to display within the dictionary UI as well as how the content is to be displayed. The preferences may be user-specified and/or learned from user behavior. For instance, a machine learning mechanism may be used to determine what content is the most interesting to a user.

Customizing display of content related to dictionary entries

Embodiments of the present invention provide methods, systems, and computer program products for re-launching contextually related application sets. In one embodiment, applications which are contextually-related are saved to a profile, along with the state of each application and data associated with each application. The set of applications, along with saved state and data, can be re-launched responsive to a trigger.

Re-launching contextually related application sets

Systems and methods for data analysis include constructing patient traces as a set of medical events for each patient of a patient population, the patient population being segmented based on patient outcomes. Medical events in one or more of the patient traces are reduced to provide processed patient traces. The processed patient traces are clustered to identify a cluster of patient traces. A process model is mined, using a processor, representing an aggregation of treatment pathways in the patient traces from the cluster. Patterns from patient traces are identified that are discriminative of patient outcomes. At least one of the patterns is represented with respect to the process model to identify treatment pathways correlated with the patient outcomes.

Extracting clinical care pathways correlated with outcomes

Embodiments of the present invention provide a method, system and computer program product for automated social networking for e-meetings. In an embodiment of the invention, a method for automated social networking for e-meetings is provided. The method includes monitoring content provided to an e-meeting managed by an e-meeting server executing in memory of a host computer. The method further includes detecting a name in the monitored content and comparing the detected name to names in a contact list for a social networking system executing externally to the e-meeting. Finally, the method includes triggering generation of a social networking introduction for the name to different participants to the e-meeting in response to matching the detected name to a name in the contact list.

Automated social network introductions for e-meetings

Methods, systems and apparatus for identifying and utilizing content synchronization information are described. Methods for creating and utilizing a synchronization file that can be used to notify users when synchronization availability exists, and to identify ways to repair synchronization unavailability are described. The synchronization file includes a complete set of individual content elements that make up a content item, such as a book, where each content element has an indication whether play-back of an audio version of the content item can be synchronized with the presentation of a corresponding text version of the content item. Methods, systems and apparatus are al'so disclosed for providing the synchronization availability information to the user in a manner that enables the user to increase the benefits of synchronized presentation of the content item.

Growing community of inventors

Winchester, MA, United States of America

Geetika Tewari Lakshmanan