Wal-mart Stores, Inc.

Walmart Apollo, LLC

Digvijay Singh Lamba

Wang Chee Lam

Xiaoyong Chai

Anand Rajaraman

Madhusudan Mathihalli

Fan Yang

Taraka Subrahmanya Prasad Siripurapu

Michel A Tourn

Nikesh Garera

Arvind Batra

Narasimhan Rampalli

Omkar Deshpande

Namrata Pramodkumar Tholiya

Chong Sun

Jun Xie

Tabassum Parveen

Abhishek Shrivastava

Gokul Kavaturi

Nicole Whisler

Gurmeet Singh

Omkar Deshoande

Indrani Chakravarty

Thomas E Chivers

Tracy ThuTrang Phung

A system, method, and computer-readable storage media, having computer-executable instructions embodied thereon, for organizing and describing product information is provided. The system includes a memory device, an IDML module, and a display module. The memory device is configured to store product data in a plurality of product records. Each product record is associated with an associated one of a plurality of products and has an associated product type. The product records provide may be used to the efficient storage, processing and exchange of product information. The IDML module is configured to receive a product request and receive, from the memory device, results data associated with the product request. The results data includes a plurality of product records. The display module receives the results data and responsively displays the results data associated with the product request.

System, method, and non-transitory computer-readable storage media for efficient storage, processing and exchange of product information

Systems and methods are provided for processing a data stream in real time to identify connections between one or more social elements social elements and aggregating and storing the identified connections over a rolling time window. Social elements may be stored as nodes and connection between social elements may be stored as labeled connections between nodes. Other embodiments are provided.

Social genome

A method of determining structured product information for a product from a product description using a product entity graph. The product graph can include a plurality of nodes. Each of the plurality of nodes can include an entity value key, one or more entity names, and an entity name count for each of the one or more entity names. The method can include determining k-grams of the product description. The method also can include, for each k-gram of the product description, determining a matching node of the plurality of nodes of the product entity graph that corresponds to the k-gram and determining a derived entity name for the product from the one or more entity names of the matching node based at least in part on the entity name counts corresponding to the one or more entity names. Other embodiments of related systems and methods are also disclosed.

Determination of product attributes and values using a product entity graph

Systems and methods are disclosed herein for performing classification of documents or performing other tasks based on rules. A rule generator receives a request for a rule that will receive as an input a document and output an outcome such as a classification of the document, addition of the document to a whitelist or blacklist, or occurrence of some other outcome. The rules are applied to a document and the document and outcome of the rules are presented to a rater. A rating of the accuracy of the outcome is received from the rater and the rating is propagated to quality metrics of rules that contributed to the outcome. Rules with a quality metric above a threshold may be added to a production rule set. Rules with a quality metric below a threshold may be removed.

Massive rule-based classification engine

Systems and methods are disclosed herein for rule-based item classification. The methods include receiving, by a computing device, an item record for analysis. The computing device may determine ranked lists of item types using rule-based classifiers and machine learning-based classifiers. Then, the computing device may aggregate the ranked lists of item types to generate a combined ranked list of item types.

Rule-based item classification

Systems and methods are disclosed herein for ranking the quality of documents, such as documents shared or referenced in postings by users. For a first set of documents quality attributes that are indicative of quality or lack of quality are identified. Ratings of the quality of the first set of documents are received. Classifiers are associated with each document and the ratings and quality attributes for each attribute used to train class-specific models corresponding to the classifiers. Subsequently received documents are then classified and corresponding quality attributes are evaluated using the corresponding class-specific model in order to rank the quality of the document.

Document quality measurement

Systems and methods are disclosed herein for extracting products referenced in a document. A document is analyzed to identify a product type that is referenced in the document. Attributes are extracted from the document. A set of candidate products are identified corresponding to the extracted attributes. A score is calculated for the candidate products and the products are further selected or filtered based on the score, whitelist rules, and blacklist rules in order to identify one or more inferred products referenced by the document. The whitelist and blacklist rules may take as inputs a domain, a user identifier, and keywords included in the document. A set of sufficient attributes may be identified for each product type. Selection of a candidate product may be based at least in part on the document including all of the attributes in the set of sufficient attributes.

Identifying product references in user-generated content

Systems and methods are provided for processing a data stream in real time to identify connections between one or more social elements social elements and aggregating and storing the identified connections over a rolling time window. Social elements may be stored as nodes and connection between social elements may be stored as labeled connections between nodes.

A method for merging two taxonomies is disclosed. Top levels of first and second taxonomies are merged. For the second taxonomy nodes are evaluated and selectively merged with nodes of the first taxonomy according to comparison of scores for these nodes with a threshold. The score for a node of the first taxonomy is a combination of one or more of a lineage quality score, Jaccard distance, string edit distance, and category depth score. After an iteration, mergings between nodes of the first and second taxonomies may be reversed if child nodes of the merged nodes were not likewise merged. Iterations may be repeated until no nodes are merged in an iteration.

Automatic taxonomy merge

Methods are disclosed for converting a directed graph to a taxonomy using guidelines from a user. An initial tree is output from a first pruning step in which subtree preferences (and other weights) are applied to preserve or remove paths from a node to one or more levels of descendent nodes. Subtree preferences (and infoboxes) may specify rules for automatically generating recommendations during application to nodes. In a second pruning step, the directed graph is again processed with additional weightings applied to edges in the graph in accordance with the recommendations. The recommendations may be human defined. Recommendations may specify a recommended ancestor for a particular node and may include a weighting to be applied to the recommendation itself, if there are multiple conflicting recommendations for the same node. Recommendations may also specify what standard weight to apply to the edge of the best parent.

Transforming a graph to a tree in accordance with analyst guidance

Disclosed are methods for returning to a user an answer to the question 'what is &#x3c;string&#x3e;.' Concepts and classes to which the concepts belong are determined from a corpus, such as taxonomy. The concepts are mapped to categories according to the structure of the taxonomy. Homonyms for words are collected and scored according to likeliness of use. Concept vectors are assembled for the identified concepts based on articles in the corpus and social media usage. Words are evaluated for generic-ness and a generic score is associated therewith. In responding to a query, the generic-ness of the terms of the query is evaluated and additional context solicited if the terms are generic. Candidate homonym concepts for a string in the query are selected according to context vectors for the homonym concepts. One or more homonym concepts are selected and the one or more categories corresponding to these concepts are returned.

Establishing 'is a' relationships for a taxonomy

Disclosed are methods for inferring interests of a user based on declared interests of the user. Text for which a user has expressed interest, e.g. 'liked' is evaluated to identify at least one principal concept. A principal article for the principal concept is located in a taxonomy and the link structure of the taxonomy analyzed to identify candidate articles related to the principal article. The candidate articles are scored according to a plurality of metrics and these scored are weighted and combined for a final score. Candidate articles are selected for the score and recommendations are generated and recommendations generated based on the concepts of the selected candidate articles.

Interest expansion using a taxonomy

Determining one or more concepts associated with a query is disclosed. A query is received. A list of concepts and associated scores is received. The concepts fit within a concept hierarchy. A density function is used to evaluate the received concepts. One or more concepts are associated with the query based at least in part on the results of the density function.

Growing community of inventors

San Bruno, CA, United States of America

Digvijay Singh Lamba