Microsoft Technology Licensing, LLC

Lucretia Henrica Vanderwende

Oscar Gerardo Naim

Krist Wongsuphasawat

William B Dolan

Sumit Basu

Stephen D Richardson

Michael Gamon

Arul A Menezes

Charles Edward Jacobs

Michael Barnett

Junyi Li

Joshua T Goodman

Krysta Marie Svore

Joseph M Joy

Daniel Morris

Christopher John Brockett

Hisami Suzuki

Simon H Corston-Oliver

Wen-tau Yih

George E Heidorn

Sean Blagsvedt

Karen Jensen

Deqing Chen

Rajatish Mukherjee

Michele Banko

Arumugam Kumaran

Igor Labutov

Jamil Valliani

Chris Brockett

Aurl A Menezes

Julie A Medero

Questions about a passage of text that includes a sequence of two or more sentences are generated. Each question covers the content of a plurality of sentences in the passage, and includes a context portion of the passage and a question statement that is contextually related to the context portion of the passage. A user is also provided with questions about a passage of text they are reading. Each question is presented to the user, where this presentation includes displaying the context portion of the passage and the question statement that is contextually related to the context portion of the passage.

Generating high-level questions from sentences

Generating responses to input utilizing an ontology-crowd-relevance methodology is described. The techniques described herein access a plurality of data items and determine an ontology associated with the plurality of data items. The ontology includes one or more ontological elements. Furthermore, the techniques describe sending, to a plurality of devices, a request to generate response templates based on the one or more ontological elements and receiving, from the plurality of devices, the response templates directed to the one or more ontological elements.

Ontology-crowd-relevance deep response generation

A mixed-initiative annotation framework is described herein. A document that includes several sentences is presented to a user, and an extractor module analyzes at least one sentence and automatically extracts a subject, predicate, object triple therefrom. The subject, predicate, object triple is displayed in graphical relation with the sentence as a proposed annotation to the sentence. Graphical options are presented to the user that allow the user to accept the proposed annotation, reject the proposed annotation, or modify the proposed annotation.

Validation, rejection, and modification of automatically generated document annotations

A graphical user interface that facilitates annotating documents is described herein. The graphical user interface includes a first field that displays a document to a user. The graphical user interface includes a second field that displays an annotation that has been assigned to a portion of the document, wherein the annotation includes a defined structure. The annotation is in the form of a tuple, wherein the tuple includes a first word or phrase, a second word or phrase, and a third word or phrase that relates the first word or phrase to the second word or phrase. The graphical user interface also includes a third field that facilitates user-editing of annotations.

Graphical user interface that supports document annotation

Technologies pertaining to annotation aggregation are described herein. A user of a computing device assigns an annotation to a portion of a document, wherein the annotation comprises a tuple. The tuple comprises semantic relationships amongst words or phrases in the document. Relationship data is also generated, wherein the relationship data identifies the document, the author of the document, the author of the annotation, and other data. The annotations and relationship data are transmitted to a data store, where such information is aggregated with other annotations made by other annotators to various documents, thereby creating a network of knowledge

Aggregating document annotations

Generation of a graphical visualization that summarizes a research document is described herein. A data store includes first data that identifies a first research document, second data that identifies a second research document that includes a citation to the first research document, and at least a portion of a sentence in the second research document that includes the citation to the first research document. A graphical visualization is generated that includes a first node that is representative of the first research document, a graphical object that indicates that the second research document includes the citation to the first research document, and the at least the portion of the sentence that includes the citation.

Generating visual summaries of research documents

Described herein are technologies pertaining to generating co-citation graphs. A context-informative co-citation graph includes a first node that represents a first research document, a second node that represents a second research document, and a third node that represents a third research document that includes a citation to both the first research document and the second research document. The context-informative co-citation graph also includes a first edge that couples the first node and the third node, and a second edge that couples the second node and the third node. The two edges visually indicate that the first research document and the second research document are co-cited by the third research document. The context-informative co-citation graph further includes at least a portion of a sentence in the third research document that includes a citation to at least one of the first research document or the second research document.

Generation of context-informative co-citation graphs

Computing functionality converts an input linguistic item into a normalized linguistic item, representing a normalized counterpart of the input linguistic item. In one environment, the input linguistic item corresponds to a complaint by a person receiving medical care, and the normalized linguistic item corresponds to a definitive and error-free version of that complaint. In operation, the computing functionality uses plural reference resources to expand the input linguistic item, creating an expanded linguistic item. The computing functionality then forms a graph based on candidate tokens that appear in the expanded linguistic item, and then finds a shortest path through the graph; that path corresponds to the normalized linguistic item. The computing functionality may use a statistical language model to assign weights to edges in the graph, and to determine whether the normalized linguistic incorporates two or more component linguistic items.

Functionality for normalizing linguistic items

Conceptually related term identification technique embodiments are presented that involve identifying in the results of a search query, terms that are conceptually related to the search query terms. In one general embodiment, this is accomplished by first inputting the terms employed in a search query and the results of the search. Word tokens found in the search query terms are then identified, as are potential phrases that can be made from the identified word tokens. Conceptually related words and phrases are then identified in the search query results. These words and phrases correspond to the previously identified word tokens and potential phrases. The search query results are presented to a user on a display device in such a manner as to visually distinguish the conceptually related words and phrases from other words and phrases in the search query results.

Identifying conceptually related terms in search query results

Methods and computer-readable media are provided for summarizing the content of a plurality of documents and presenting the results of such multiple-document summarization to a user in such a way that the user is able to quickly and easily discern what, if any, unique information each document contains. Each sentence of each document is assigned a score based upon the perceived importance of the information contained therein. The sentences receiving the highest scores are then compared with one another to identify and remove any duplicate sentences. The remaining high-scoring sentences are extracted from the corresponding documents and presented to the user, for instance, in a bulleted list format. The user can then simply scan the list and can quickly and easily discern the unique information contained in each document.

Presenting multiple document summarization with search results

A summarization system and method. The summarization method includes utilizing a first body of information to obtain a second body of information, which is identified (by a hyperlink, an attachment identifier, a reference, etc.) in the first body of information. A summary of the obtained second body of information is then computed. The computed summary can be displayed to a user and/or stored for later use.

Summarization of attached, linked or related materials

A document translation system is described. In various embodiments, the document translation system improves translations of text or speech ('documents') that are performed by computing devices ('machine translations'). Upon translating a document into a target language, the system can provide a user interface containing both the original document and a translated document. The translated document can include portions that are translated automatically by a machine translation service, and other portions that have been amended or modified by users. Upon providing the translated document, the system may receive input from a user describing portions of the translated document that should be modified. These modifications may then be stored in corpora that the document translation system can employ during future requests to translate the document. When multiple modifications are stored, the system may select one of the stored modifications during translation, such as based on the document type, reputation of the user who provided the modification, and so forth. The stored corpora can then be employed to improve future machine translations.

Growing community of inventors

Sammamish, WA, United States of America

Lucretia Henrica Vanderwende