Microsoft Technology Licensing, LLC

Kuansan Wang

Michael D Malueg

David Milstein

Linda Criddle

Scott C Forbes

Hsiao-Wuen Hon

Lon-Chan Chu

Yu-ting Kuo

Bo-June Hsu

David Andrew Howell

Xuedong D Huang

Gursharan Singh Sidhu

Xiaolong Li

Philip Andrew Chou

Timothy Mark Moore

Emre Mehmet Kiciman

Frederic H Behr, Jr

Alejandro Acero

Heung-Yeung Shum

Jianfeng Gao

Milind V Mahajan

Arun Sacheti

Chao-Chia Liu

Nirav Girish Shah

Antonio Gulli

Chiyuan Huang

Jeremy Espenshade

Jiangbo Miao

Dong Yu

Yi-Min Wang

Johnson T Apacible

Jonathan Jay Cadiz

Ran Gilad-Bachrach

Daniel Marantz

David Gareth Ollason

Parthasarathy Govindarajen

Zijian Zheng

Siddharth Bhatia

Chun-Kai Wang

Aleksey Sinyagin

Christopher A Meyers

Kieran Richard McDonald

Aaron Chun Win Yuen

Lawrence William Colagiovanni

Peter K L Mau

Adwait Ratnaparkhi

Kumaresh Pattabiraman

Tapas Kanungo

Yu Fan Chen

Cornelia Carapcea

Yuzhe Jin

Robert C Wang

Neal Alan Bernstein

Toby H Walker

Ashok Ponnuswami

Yang Ye

Luke Evans Simon

Abulimiti Aji

Bo-June (Paul) Hsu

Yusuf Furkan Fidan

Nikolas Clemens Gloy

Guarang P Prajapati

Yin-Cheng Ting

Xuendong D Huang

Yu Dong

Described herein are technologies that are configured to assist a user in recollection information about people, places, and things. Computer-readable data is captured, and contextual data that temporally corresponds to the computer-readable data is also captured. In a database, the computer-readable data is indexed by the contextual data. Thus, when a query is received that references the contextual data, the computer-readable data is retrieved.

Extended memory system

Systems, methods, computer storage media, and user interfaces are provided for non-committal intent preview, disambiguation, and refinement of a search. A search prefix comprising one or more characters associated with an unexecuted search query is received. One or more intent suggestions are suggested to a user. For each of the one or more intent suggestions, one or more entity identifications associated with each of the one or more intent suggestions are received. Metadata corresponding to at least one entity associated with the one or more entity identifications is retrieved from an entity data store. Without retrieving search results for the unexecuted search query, an aggregated intent preview based on the retrieved metadata corresponding to the at least one entity is provided.

Search intent preview, disambiguation, and refinement

Systems, methods, and computer storage media having computer-executable instructions embodied thereon that provide contextual services are provided. Embodiments of the present invention allow an input mechanism to provide contextual services. Exemplary input mechanisms include a keyboard, a gesture interface, and a speech interface. These inputs may be used to provide user input into one or more applications running on a computer. The contextual services provided include composition assistance, grammatical assistance, communication-context assistance, and research assistance. In one embodiment, an input mechanism ('IME') program provides the contextual service. The IME program may work with a remote contextual-service provider. The IME program communicates user input and contextual information to the contextual-service provider. The contextual-service provider analyzes the input and contextual information to determine whether one or more contextual services should be provided.

Intent-aware keyboard

A method and system for collecting and verifying the location information of a calling party and a device of the calling party is provided. More specifically, a method and system is provided for determining whether the identity of the calling party can be confirmed, via evaluating location information, with an acceptable degree of certainty. The location information may be provided by the calling party or obtained from various sources over a digital communication channel. Some of the provided location information which can be accidentally or intentionally altered is identified and evaluated to determine its accuracy as part of the verification process of the caller's identity.

Secured communication via location awareness

Various technologies described herein pertain to disambiguation of a mention of an ambiguous entity in a document. A set of candidate entities can be retrieved from an entity knowledge base based upon the mention of the ambiguous entity, where each of the candidate entities has a respective entity feature representation. Moreover, a document feature representation can be generated based upon features of the document and the respective entity feature representations of the candidate entities. A processor can be caused to select a subset of features from the document feature representation based upon a measure of how discriminative the features from the document feature representation are for disambiguating the mention of the ambiguous entity. A disambiguated result for the mention of the ambiguous entity can be determined based upon the subset of the features. The disambiguated result can be an unknown entity or one of the candidate entities.

Posterior probability pursuit for entity disambiguation

Generally described, the present invention provides the ability to process digital voice conversations to identify data packets containing content of interest and to further process the identified data packets. More specifically, mining profiles may be developed identifying particular types of content that is to be mined and further identifying what is to be done when data packets containing such content is located. A system may search a digital voice conversation for the data packets containing the content and perform processing on the data packets once identified.

Mining data for services

A statistical language model (SLM) may be iteratively refined by considering N-gram counts in new data, and blending the information contained in the new data with the existing SLM. A first group of documents is evaluated to determine the probabilities associated with the different N-grams observed in the documents. An SLM is constructed based on these probabilities. A second group of documents is then evaluated to determine the probabilities associated with each N-gram in that second group. The existing SLM is then evaluated to determine how well it explains the probabilities in the second group of documents, and a weighting parameter is calculated from that evaluation. Using the weighting parameter, a new SLM is then constructed as a weighted average of the existing SLM and the new probabilities.

Adaptive construction of a statistical language model

The ranking quality of a ranked list may be evaluated. In an example embodiment, a method is implemented by a system to access log data, ascertain which entries of a ranked list are skipped, and determine a ranking quality metric from the skipped entries. More specifically, log data that reflects user interactions with a ranked list having multiple entries is accessed. The user interactions include at least indications of which of the multiple entries are selected entries. It is ascertained which entries of the multiple entries of the ranked list are skipped entries based on the selected entries. The ranking quality metric for the ranked list is determined responsive to the skipped entries.

Evaluating the ranking quality of a ranked list

Methods, systems, and computer-readable media for evaluating the quality of text within online advertisements using output from a language model are provided. The output from the language model may be used by a machine-learning algorithm to generate a quality score for an individual advertisement. The quality score may be used to filter out advertisements with poor text quality or to tax or penalize an advertisement within an online auction. The ad quality scores may also be used to rank or score advertisers that submit the ads. In one embodiment, the advertiser's quality score is combined with an individual ad's quality score to create a final score, which is used to evaluate the advertisement. The advertiser rank/score and ad quality score may be communicated to an advertiser as advertiser feedback.

Ad copy quality detection and scoring

A classifier that disambiguates among entities based on a dictionary, such as corpus of documents about those entities, is built by incorporating probabilities that an entity exists that is not in the dictionary. Given a document it is associated by the classifier with an entity. By incorporating out of collection probabilities into the classifier, a higher level of confidence in the match between an entity and a document is achieved.

Entity recognition using probabilities for out-of-collection data

An auto-answer feature is implemented in SIP by configuring a receiving device to automatically acknowledge and answer an incoming call or session from a specific trusted third party. The receiving device may skip to an OK response to an INVITE request when the call is routed through the trusted third party. When the device can automatically answer the incoming call, advanced features such as Push To Talk, Information Tone, Click to Call, and Remote Monitoring may be easily implemented.

Auto answer in voice over internet protocol

The claimed subject matter provides a method for entering technical formulas. The method includes receiving a sequence of inputs on a computing device. The sequence of inputs specifies the technical formula. The method also includes displaying the technical formula in response to receiving the sequence of inputs. Additionally, the method includes determining an alternate sequence of inputs that specify the technical formula. The alternate sequence of inputs may be input to the computing device to display the technical formula. The method further includes displaying the alternate sequence.

Entering technical formulas

Growing community of inventors

Bellevue, WA, United States of America

Kuansan Wang