Amazon Technologies, Inc.

David Spike Palfrey

Petra Elisabeth Holmes

William Tunstall-Pedoe

Sina Samangooei

Mihai Valentin Tablan

Andrew Graham Fenn

Duncan Joseph Reynolds

Gregory Christopher Doran

Emilio Fabio Monti

Meng Li

Mark Edward Rafn

Anjishnu Kumar

Arpit Mittal

Giovanni Zappella

Samuel M Tucker

Rodolphe Jenatton

Samuel Theodore Sandler

Robert Peter Stacey

Dianhuan Lin

Janez Starc

Benjamin David Scott

Marcel Ackermann

Barry Dean Glennon

Sainyam Kapoor

Maxime Peyrard

Jianhan Zhu

Andrew Hopkinson

Pedro Ribeiro

Neil Copland

Graham John French

Techniques are described for determining the value of individual facts in a knowledge base, and various applications of such fact values. In one example, the knowledge base is part of a question answering system. A ranking of knowledge base facts based on the number of times each of the knowledge base facts is used in answering user questions (e.g., as determined from question answering logs) is used to derive a fact value function that may then be used to determine the value of other facts included in or subsequently added to the knowledge base.

Determining the value of facts in a knowledge base and related techniques

A method comprising sending query data representative of a query to a first and second query-answering subsystem of a query-answering system. It is determined that the query-answering system lacks a satisfactory answer to the query. Paraphrase data representative of a rephrased version of the query is received. The paraphrase data is sent to the first and second query-answering subsystems. Answer data representative of an answer to the rephrased version of the query is received from the first query-answering subsystem.

Query paraphrasing

A query parsing system uses a multi-stage process to parse the text of incoming queries before attempting to answer the queries. The multi-stage configuration involves a first trained classifier to determine the query type, or intent, of the text and a plurality of second trained classifiers, where each of the second trained classifiers is configured particularly for one specific respective query type. During query processing, the first trained classifier is used on the text to identify the query type. A second trained classifier for that specific identified query type is then found and used on the text to identify what strings in the text correspond to specific entities needed to resolve the query. The identified text strings and query type are then placed into a form understandable by a knowledge base and sent to the knowledge base for resolution. The classifiers may be trained using queries and answers previously processed by the knowledge base using a rules/template resolution process.

Multi-stage query processing

A system is disclosed that identifies early adopter users by creating a directed graph of item access information for an item category and performing a page rank type process on the item access information. This directed graph may be created in a reverse temporal order. The early adopter users can be identified as the users with nodes in the directed graph that have a threshold number or rate of incoming links directly or indirectly pointing towards the nodes. Using the early adopter users as a sample, systems herein can determine whether to recommend an item based on the popularity of the item with respect to the early adopter users. Further, systems herein can determine an inventory level to maintain for an item based on the popularity of the item with respect to the early adopter users.

Automated graph-based identification of early adopter users

Techniques are described for adding knowledge to a knowledge base.

Adding to a knowledge base using an ontological analysis of unstructured text

A system and method to recognize and resolve when a source data object is the same or similar to an existing data object in a database using structured information and facts about each object. The system and method compare relations of the source data object and relations of the existing data object in the database and determine how similar the source data object is to the data object in the database based on scores of the comparisons. The system and method may provide outputs, such as whether the data objects match, the data objects are distinct, the source data object is a strong match to multiple data objects in the database, and the data objects conflict one another. The system and method may then resolve the database entries based on the potential matching outputs.

Data object resolver

Techniques are described for adding knowledge to a knowledge base in which alternative information retrieval techniques are used to obtain missing information from an available corpus of documents (e.g., the World Wide Web). Existing relationships in a knowledge base are used to identify entities for which the relationships are not expressed in the knowledge base. The alternative information retrieval techniques are then used to find this information.

Using existing relationships in a knowledge base to identify types of knowledge for addition to the knowledge base

A system and method is disclosed for improving automated question-answering using real-world knowledge from a knowledge base. The system can be used to answer questions from various users. The system can generate answers to these questions using data stored in a knowledge base. In some embodiments, the system is provided with a list of questions separate from any user questions. The system can be trained using these questions in advance of a user question. The system can process the presented questions to determine that it can generate high-quality correct answers. The system can take various steps to determine a high-quality answer to a question. The system can utilize human intelligence providers to improve this process, such as through a human interactive task system. Human intelligence can be used to determine that a question is understood, that a question is answered, and that the answer is of high quality.

Scalable curation system

A system for resolving a user's question regarding media at the user's location where the question includes ambiguity as to the context of the question. Context information is acquired from a device or devices at the user's location, and if multiple possible contexts are identified, the system may attempt to resolve the question for more than one of the possible contexts. The system for answering questions and the source of the media may be different and independent.

System for resolving ambiguous queries based on user context

Growing community of inventors

Cambridge, United Kingdom

David Spike Palfrey