Microsoft Technology Licensing, LLC

Motorola Corporation

Yahoo! Inc.

Google Technology Holdings LLC

Tasos Anastasakos

Ruhi Sarikaya

Larry Paul Heck

Daniel Boies

Senaka Balasuriya

Chi-Chao Chang

Michael Van Wie

Manish Tayal

Jianchang Mao

Jayavel Shanmugasundaram

Erik Nathan Vee

John Anthony Tomlin

Sergei Vassilvitskii

Alexandre Rochette

Jian Yang

Anurag Kumar Gupta

Oliver Kennedy

Tools are provided to allow developers to enable applications for Conversational Understanding (CU) using assets from a CU service. The tools may be used to select functionality from existing domains, extend the coverage of one or more domains, as well as to create new domains in the CU service. A developer may provide example Natural Language (NL) sentences that are analyzed by the tools to assist the developer in labeling data that is used to update the models in the CU service. For example, the tools may assist a developer in identifying domains, determining intent actions, determining intent objects and determining slots from example NL sentences. After the developer tags all or a portion of the example NL sentences, the models in the CU service are automatically updated and validated. For example, validation tools may be used to determine an accuracy of the model against test data.

Building conversational understanding systems using a toolset

Environmental conditions, along with other information, are used to adjust a response of a conversational dialog system. The environmental conditions may be used at different times within the conversational dialog system. For example, the environmental conditions can be used to adjust the dialog manager's output (e.g., the machine action). The dialog state information that is used by the dialog manager includes environmental conditions for the current turn in the dialog as well as environmental conditions for one or more past turns in the dialog. The environmental conditions can also be used after receiving the machine action to adjust the response that is provided to the user. For example, the environmental conditions may affect the machine action that is determined as well as how the action is provided to the user. The dialog manager and the response generation components in the conversational dialog system each use the available environmental conditions.

Environmentally aware dialog policies and response generation

User input expressed as text may be analyzed for determining a type of response, such as an application response, and/or determining a type of task that is requested by the user input. Entity representations may be identified, classified and/or or tagged based on a type of response, type of task and/or a set of entity types. A surface form of an entity, ambiguous entity representation and/or other type of expression within the user input may be resolved, normalized and/or mapped to a normalized value. Normalizing entities and/or entity attributes may involve using a set of normalization rules, a lookup table, one or more machined learned methods, and/or an entity normalization index that associates entities with alternate surface forms derived from web corpora. The normalized value may be used to construct a request to a structured knowledge source and/or an application.

Entity and attribute resolution in conversational applications

An advertisement impression distribution system includes a data processing system operable to generate an allocation plan for serving advertisement impressions. The allocation plan allocates a first portion of advertisement impressions to satisfy guaranteed demand and a second portion of advertisement impressions to satisfy nonguaranteed demand. The data processing system includes an optimizer configured to establish a relationship between the first portion of advertisement impressions and the second portion of advertisement impressions. The relationship defines a range of proportions of allocation of the first portion and the second portion. The optimizer generates a solution maximizing guaranteed demand fairness, non-guaranteed demand revenue, and click value. The solution identifies a determined proportion of the first portion of advertisement impressions to serve and a determined proportion of the second portion of advertisement impressions to serve. The data processing system outputs the allocation plan to control serving of the advertisement impressions.

System for display advertising optimization using click or conversion performance

A system is described for generating query suggestions using a network of users and advertisers. The system may include a memory, an interface, and a processor. The memory may store a data representing a network comprising query items linked to advertisement items via link items, wherein each link item comprises a weight representing the strength of the relationship between each query item and advertisement item, a search query item, and a relevance value for each query item. The processor may be operatively connected to the memory and the interface and may identify the data representing the network and receive a search query item. The processor may calculate a relevance value for each additional query item in the network based on its relationship to the received search query item. The processor may then suggest the query items with the highest relevance values to the user via the interface.

System for generating query suggestions using a network of users and advertisers

A system is described for generating query suggestions by integrating valuable query suggestions with experimental query suggestions using a network of users and advertisers. The system may include a memory, an interface, and a processor. The memory may store a historical dataset, a plurality of query suggestions, a plurality of query suggestion values, a query exploit set, a query explore set, and a data describing a network. The processor may identify the plurality of query suggestions in the historical dataset and generate data describing the network based on the historical dataset. The processor may calculate the query suggestion value for each query suggestion and may rank the query suggestions based on the query suggestion values. The processor may generate an exploit set comprising the top ranked query suggestions and an explore set comprising the remainder. The processor may suggest the query suggestions in the exploit set and the explore set.

System for generating query suggestions by integrating valuable query suggestions with experimental query suggestions using a network of users and advertisers

A method used in an electronic equipment () generates a set of joint multimodal interpretations () from a set of multimodal interpretations () generated by one or more modalities () during a turn, generates a set of integrated multimodal interpretations () including an integrated multimodal interpretation formed from each joint multimodal interpretation by unifying the type feature structure of each multimodal interpretation in the joint multimodal interpretation, and generates a multilevel confidence score for each integrated multimodal interpretation based on at least one of a context score, a content score, and a confidence score of the integrated multimodal interpretation. The method classifies the multimodal interpretations and generates a set of joint multimodal interpretations that comprises essentially all possible joint multimodal interpretations. The multilevel confidence scoring is based on up to eleven factors, which provides for an accurate ranking of the integrated multimodal interpretations.

Method and apparatus for classifying and ranking interpretations for multimodal input fusion

A method and apparatus for selective distributed speech recognition includes an embedded speech recognition engine () and a dialog manager (), such as a browser, coupled to the embedded speech recognition engine (). The method and apparatus further includes the dialog manager () being operably couple able to at least one external speech recognition engine (), such as a WLAN speech recognition engine () or a network speech recognition engine (). The method and apparatus further includes preference information (), environment information () and a speech input () all provided to the dialog manager (). The dialog manager (), in response to the preference information () and the environment information (), provides the speech input () to the embedded speech recognition engine (), the WLAN speech recognition engine () or the network speech recognition engine ().

Method and apparatus for selective distributed speech recognition

An apparatus and method for selective distributed speech recognition includes a dialog manager () that is capable of receiving a grammar type indicator (). The dialog manager () is capable of being coupled to an external speech recognition engine (), which may be disposed on a communication network (). The apparatus and method further includes an audio receiver () coupled to the dialog manager () wherein the audio receiver () receives a speech input () and provides an encoded audio input () to the dialog manager (). The method and apparatus also includes an embedded speech recognition engine () coupled to the dialog manager (), such that the dialog manager () selects to distribute the encoded audio input () to either the embedded speech recognition engine () or the external speech recognition engine () based on the corresponding grammar type indicator ().

Growing community of inventors

San Jose, CA, United States of America

Tasos Anastasakos