Microsoft Technology Licensing, LLC

Syed Farhan Raza

James Han Wang

Christopher Chinwen Yu

Rahul Anantha Padmanabha Udipi

Qian Liu

Justin James Wagle

Neal Alan Bernstein

Hemanshu Khimraj Chawda

Timothy Edward Bellay

Harlan Joy Frost

Amrita Chowdhury

Chad William Newbry

Nicholas Romano

Armani Ronaldo Rosiles

Nathaniel Gunther Roth

Zeyuan Guan

Pnina Eliyahu

A technique is described herein for automatically logging journeys taken by a user, and then automatically classifying the purposes of the journeys. In one implementation, the technique obtains journey data from one or more movement-sensing devices as a user travels from a starting location to an ending location in a vehicle. The technique generates a set of features based on the journey data, and then uses a machine-trainable model (such as a neural network) to make its classification based on the features. The machine-trainable model accepts at least one feature that is based on statistical information regarding at least one aspect of prior journeys that the user has taken. Overall, the technique provides a resource-efficient solution that rapidly provides personalized results to individual respective users. In some implementations, the technique performs its personalization without sharing journey data with a remote server.

Automated personalized classification of journey data captured by one or more movement-sensing devices

A computer-implemented technique is described for using an automation tool to facilitate the preparation of a summary document, such as an expense report. In a first data path, a receipt-processing component generates receipt items based on images of physical receipts that a user captures using a camera. In a second data path, one or more transaction-processing systems forward transaction items to the automation tool upon their capture; the transaction items correspond to electronic records of transactions made by the user. The technique then generates a set of receipt-transaction (R-T) pairings by matching pairs of receipt items and transaction items that describe the same respective transactions. The technique then uses context information to supplement the R-T pairings, and generates a summary document that summarizes transactions made by the user based on the supplemented R-T pairings. In one case, the context information corresponds to calendar information specified in a user's electronic calendar.

Automatically processing a stream of transaction items based on image and calendar information

Growing community of inventors

Issaquah, WA, United States of America

Syed Farhan Raza