Salesforce Com, Inc.

Salesforce, Inc.

Microsoft Technology Licensing, LLC

Rajdeep Singh Dua

Joy Mustafi

Lakshya Kumar

Karthick Krishnamoorthy

Akash Singh

Machiraju Pakasasana Rama Rao

Raman Kumar

Piyush Singh

Shashank Tyagi

Debadri Basak

Mohan Krishna Kusuma

Nutana Sukumar Reddy Muramreddy

Gowri Shankar Raju Kurapati

Mani Kandar Madduri

A hierarchical neural network for predicting out of stock products comprises an input layer that receives data from data sources that store disparate datasets having different levels of attribute detail pertaining to products for sale in stores of a retailer. A first level of neural networks processes the data from the data sources into respective learned intermediate vector representations. A second level comprises a concatenate layer that concatenates the learned intermediate vector representations from the second level into a combined vector representation. A third level comprises a feed forward network that receives the combined vector representation and outputs to the retailer an out of stock probability indicating which store and product combinations are likely to have out of stock products over a predetermined timeframe.

Hierarchical neural network based implementation for predicting out of stock products

A method and system for synthetic data generation are provided that receive a schema configuration file in a synthetic data set request from a client application, create a set of worker processes to generate the synthetic data set based on the schema configuration file, upload the generated synthetic data to an analytics platform, and enable the client application to utilize the generated synthetic data in prediction models for the analytics platform.

Synthetic data generation for machine learning model simulation

A system is provided for object localization in image data. The system includes an object localization framework comprising a plurality of object localization processes. The system is configured to receive an image comprising unannotated image data having at least one object in the image, access a first object localization process of the plurality of object localization processes, determine first bounding box information for the image using the first object localization process, wherein the first bounding box information comprises at least one first bounding box annotating at least a first portion of the at least one object in the image, and receive first feedback regarding the first bounding box information determined by the first object localization process. The system is further configured to persist the image with the first bounding box information or access a second object localization process based on the first feedback.

Object localization framework for unannotated image data

Machine learning based models recognize objects in images. Specific features of the object are extracted from the image using machine learning based models. The specific features extracted from the image assist deep learning based models in identifying subtypes of a type of object. The system recognizes the objects and collections of objects and determines whether the arrangement of objects violates any predetermined policies. For example, a policy may specify relative positions of different types of objects, height above ground at which certain types of objects are placed, or an expected number of certain types of objects in a collection.

Machine learning based models for object recognition

Disclosed herein are system, method, and computer program product embodiments for compliance auditing using cloud based computer vision. In one aspect, a system is configured to receive, from a mobile device, a compliance audit request to at least recognize one or more products within the audit image. The system is further configured to select a first object recognition model having a first associated object recognition model identifier from a model selection list based at least on a required object recognition list, wherein the first object recognition model is configured to recognize a first set of object names within the required object recognition list. The system is further configured to request the computer vision system to perform object recognition using the first object recognition model to recognize the first set of object names within the audit image, and transmit audit result information to the mobile device.

Selection of object recognition models for computer vision

Embodiments described herein provide a method for obtaining information on product inventory and placement in a retail setting. An image including unannotated image data indicative of the retail setting is received. One or more shelves in the retail setting are determined from the unannotated image data, and the image is segmented into one or more sub-images corresponding to the one or more detected shelves. For each sub-image corresponding to a respective detected shelf, a product name is then and a number of appearances of the product name are detected using text recognition on the respective sub-image. Product inventory information and first placement information are derived based at least in part on the detected number of appearances and a shelf level corresponding to the sub-image.

System and method for live product report generation without annotated training data

The system and methods of the disclosed subject matter provide a hybrid machine learning approach for recommending items that a consumer should be shown as a next best offer. The recommendation may be based on the consumer's previous behavior, other consumers' previous behavior, and the consumer's profile. The system and methods may cluster an input dataset using an unsupervised clustering engine. The dataset output from the unsupervised clustering engine may be subsequently provided to the input of a supervised machine learning engine to generate a rules-based model. The system and methods may use the rules-based model to subsequently cluster new user data and generate recommendations based on the user's assigned cluster.

Automatic rule generation for recommendation engine using hybrid machine learning

RDF store database designs and efficient techniques for converting SPARQL queries to SQL queries are described that provide faster triplet access, and which can reduce the computational overhead and cost associated with storing large volumes of RDF metadata. In various embodiments RDF data can be stored in de-normalized tables tailored to provide efficient query and storage performance. The provided query conversion techniques provide reliable and efficient query performance.

Growing community of inventors

Hyderabad Telangana, India

Rajdeep Singh Dua