Google Inc.

Ryan T McDonald

Jeffrey C Reynar

Kerry L Hannan

Tyler J Neylon

Sasha Blair-Goldensohn

Slav Petrov

Keith Hall

Michael Lloyd Wells

Dipanjan Das

George Reis

Oscar Carl Tackstrom

Jason Katz-Brown

Joakim Nivre

A method, a system and a computer program product for ranking reviewable entities based on sentiment expressed about the entities. A plurality of review texts are identified wherein each review text references an entity. A plurality of sentiment scores associated with the plurality of review texts are generated, wherein each sentiment score for a review text indicates a sentiment directed to the entity referenced by the review text. A plurality of ranking scores for the plurality of entities are generated wherein each ranking score is based at least in part on one or more sentiment scores associated with one or more review texts referencing the entity. A plurality of search results associated with the plurality of entities are displayed based at least in part on the ranking scores.

Sentiment detection as a ranking signal for reviewable entities

A method and system are provided for a part-of-speech tagger that may be particularly useful for resource-poor languages. Use of manually constructed tag dictionaries from dictionaries via bitext can be used as type constraints to overcome the scarcity of annotated data in some instances. Additional token constraints can be projected from a resource-rich source language via word-aligned bitext. Several example models are provided to demonstrate this such as a partially observed conditional random field model, where coupled token and type constraints may provide a partial signal for training. The disclosed method achieves a significant relative error reduction over the prior state of the art.

Weakly supervised part-of-speech tagging with coupled token and type constraints

A source language sentence is tagged with non-lexical tags, such as part-of-speech tags and is parsed using a lexicalized parser trained in the source language. A target language sentence that is a translation of the source language sentence is tagged with non-lexical labels (e.g., part-of speech tags) and is parsed using a delexicalized parser that has been trained in the source language to produce k-best parses. The best parse is selected based on the parse's alignment with lexicalized parse of the source language sentence. The selected best parse can be used to update the parameter vector of a lexicalized parser for the target language.

Multi-source transfer of delexicalized dependency parsers

Phrases in the reviews that express sentiment about a particular aspect are identified. Reviewable aspects of the entity are also identified. The reviewable aspects include static aspects that are specific to particular types of entities and dynamic aspects that are extracted from the reviews of a specific entity instance. The sentiment phrases are associated with the reviewable aspects to which the phrases pertain. The sentiment expressed by the phrases associated with each aspect is summarized, thereby producing a summary of sentiment associated with each reviewable aspect of the entity. The summarized sentiment and associated phrases can be stored and displayed to a user as a summary description of the entity.

Aspect-based sentiment summarization

A plurality of highest intrinsically-ranked parser outputs are scored using an extrinsic metric that is relevant to one or more downstream applications. The parser can be trained by using an extrinsically highly-ranked parse selected from among the intrinsically highly-ranked parses. The selected parse need not be the highest intrinsically-ranked parse. In this way, a parser can be trained to produce better results for downstream applications.

Training dependency parsers by jointly optimizing multiple objectives

Disclosed herein is a method, a system and a computer product for generating a snippet for an entity, wherein each snippet comprises a plurality of sentiments about the entity. One or more textual reviews associated with the entity is selected. A plurality of sentiment phrases are identified based on the one or more textual reviews, wherein each sentiment phrase comprises a sentiment about the entity. One or more sentiment phrases from the plurality of sentiment phrases are selected to generate a snippet.

Phrase based snippet generation

A domain-specific sentiment classifier that can be used to score the polarity and magnitude of sentiment expressed by domain-specific documents is created. A domain-independent sentiment lexicon is established and a classifier uses the lexicon to score sentiment of domain-specific documents. Sets of high-sentiment documents having positive and negative polarities are identified. The n-grams within the high-sentiment documents are filtered to remove extremely common n-grams. The filtered n-grams are saved as a domain-specific sentiment lexicon and are used as features in a model. The model is trained using a set of training documents which may be manually or automatically labeled as to their overall sentiment to produce sentiment scores for the n-grams in the domain-specific sentiment lexicon. This lexicon is used by the domain-specific sentiment classifier.

Growing community of inventors

Mountain View, CA, United States of America

Ryan T McDonald