Medallia, Inc.

Palo Alto Research Center Inc.

Ji Fang

Tzu-Ting Kuo

Andrew J Yeager

Tracy Holloway King

Andrew Ko

Jessica Nicola Staddon

Kurt Edward Partridge

Nicolas B Ducheneaut

Richard Chow

Philippe J P Golle

Lauri J Karttunen

Jiacong He

Yibin Lin

Jason Kessler

Bi Chen

Po-Yu Li

Pablo Zivic

Valeria De Paiva

Embodiments of the present invention provide a system that can be used to classify a feedback image in a user review into a semantically meaningful class. During operation, the system analyzes the captions of feedback images in a set of user reviews and determines a set of training labels from the captions. The system then trains an image classifier with the set of training labels and the feedback images. Subsequently, the system generates a signature for a respective feedback image in a new set of user reviews using the image classifier. The signature indicates a likelihood of the image matching a respective label in the set of training labels. Based on the signature, the system can allocate the image to an image cluster.

System and method for cascading image clustering using distribution over auto-generated labels

Embodiments of the present invention provide a system that that can be used to determine whether a sentiment analysis model is portable between two data sets. During operation, the system analyzes the text of a respective review in a data set (e.g., a set of reviews) using the sentiment analysis model to determine a sentiment expressed in the review. The system then computes a confidence score, which indicates the accuracy of a respective sentiment. The system subsequently determines a confidence score distribution for various sentiments, as determined by the sentiment analysis model. The system determines the significance of changes between the confidence score distribution and a benchmark confidence score distribution, which is associated with a benchmark data set for which the sentiment analysis model yields a high accuracy. The system can then determine whether the sentiment analysis model is portable to the data set based on the significance of changes.

System and method for detecting portability of sentiment analysis system based on changes in a sentiment confidence score distribution

One embodiment provides a system that obtains, for a business entity, review and revenue information for a period of time from network packets, and extracts rankings and revenues from the review and revenue information. The system determines respective correspondence between the rankings and revenues for a plurality of points of time. The system determines first and second normalized rankings for first and second points of time, respectively, in the plurality of points of time. The system determines first and second normalized revenues for the first and second points of time, respectively. The system calculates a first correlation strength between the changes in the first and second normalized rankings, and the changes in the first and second normalized revenues. The system computes a correlation between the rankings and corresponding revenues based on a plurality of correlation strengths, which includes the first correlation strength, over the period of time.

System and method for facilitating non-parametric weighted correlation analysis

Text clustering includes: identifying, for a set of non-stop words in a text, a corresponding set of related topic clusters relating to the set of non-stop words, the identification being based at least in part on a plurality of topic clusters each comprising a corresponding plurality of topically related words and a corresponding cluster identifier; for non-stop words in the set of non-stop words that are identified to have corresponding related topic clusters, replacing the non-stop words with corresponding cluster identifiers of the corresponding related topic clusters to generate a clustered version of the text; and providing the clustered version of the text to be further analyzed.

Enhanced text clustering based on topic clusters

Context-based sentiment analysis includes: determining whether a piece of comment data included in a comment is context-sensitive, the determination being made with reference to a set of comment features; determining a context sentiment type of an associated context related to the piece of comment data, the associated context being distinct from the comment, in the event that the piece of comment data is determined to be context-sensitive, and the determination being made with reference to at least a set of context features; and classifying the comment data in the event that the piece of comment data is determined to be context-sensitive, the classification being based at least in part on the comment data and the context sentiment type.

Context-based sentiment analysis

An automatic paraphrase acquisition technique is provided. A common theme of the various embodiments described herein resides in careful design of simple tasks that can elicit the necessary information for the automated process. These tasks are performed quickly and inexpensively. By gathering the results produced, paraphrases can be generated automatically using the method and/or system.

System and method to acquire paraphrases

Some embodiments of the present invention provide a system that infers activity-related context information from a message. Upon receiving the message, the system looks for activity-related keywords in the message, wherein the activity-related keywords are from a content database. If one or more activity-related keywords are found in the message, the system infers message-related context information from the one or more activity-related keywords. Next, the system uses the message-related context information to facilitate recommending an activity to a user.

Using a content database to infer context information for activities from messages

Computer methods, apparatus and articles of manufacture therefor, are disclosed for text-characterization using a finite state transducer that along each path accepts on a first side an n-gram of text-characterization (e.g., a language or a topic) and outputs on a second side a sequence of symbols identifying one or more text-characterizations from a set of text-characterizations. The finite state transducer is applied to input data. For each n-gram accepted by the finite state transducer, a frequency counter associated with the n-gram of the one or more text-characterizations in the set of text-characterizations is incremented. The input data is classified as one or more text-characterizations from the set of text-characterizations using the frequency counters associated therewith.

Region-matching transducers for text-characterization

A sentiment classifier for sentiment classification of content. An aspect classifier is configured to classify content as being related to a particular aspect of information, the aspect classifier incorporating at least a portion of the domain specific sentiment lexicon. A polarity classifier is then configured to classify the content classified by the aspect classifier as having one of a positive sentiment of the particular aspect of information, a negative sentiment of the particular aspect of information or as having no sentiment as to the particular aspect of information. The polarity classifier also incorporating at least a portion of the domain specific sentiment lexicon.

Incorporating lexicon knowledge into SVM learning to improve sentiment classification

One embodiment of the present invention provides a system that detects sensitive content in a document. In doing so, the system receives a document, identifies a set of terms in the document that are candidate sensitive terms, and generates a combination of terms based on the identified terms that is associated with a semantic meaning. Next, the system performs searches through a corpus based on the combination of terms and determines hit counts returned for each term in the combination and for the combination. The system then determines whether the combination of terms is sensitive based on the hit count for the combination and the hit counts for the individual terms in the combination, and generates a result that indicates portions of the document which contain sensitive combinations.

Growing community of inventors

Mountain View, CA, United States of America

Ji Fang