Browse Prior Art Database

Cognitive Personal Shopping Assistant Optimized on Shopper’s Profile

IP.com Disclosure Number: IPCOM000241828D
Publication Date: 2015-Jun-02
Document File: 5 page(s) / 226K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a system for a Personal shopping assistant (PSA) that understands a user's shopping preferences and activities and dynamically generates and updates a shopping list. The main point of novelty is the ability of the system to learn while processing customer shopping experiences and then dynamically generate personalized recommendations for the next shopping list.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 35% of the total text.

Page 01 of 5

Cognitive Personal Shopping Assistant Optimized on Shopper's Profile

A shopper spends lot of time searching for personal choice items in stores, trying to keep track of previously purchased items, planning purchases for upcoming social events, etc.

A system or method is needed to generate a personal shopping list by processing

shopper preferences, social activities, shopping habits, etc. The shopper needs to have a personal shopping assistant that:

• Understands the user's shopping preferences • Reminds the user of required shopping • Provides purchasing suggestions based on the user's calendar events and previous associated purchases
• Provides purchasing recommendations based on the social event type, etc.

The novel contribution is a Personal shopping assistant (PSA). The proposed PSA understands the user's shopping preferences and activities and dynamically generates and updates a shopping list. The PSA processes and extracts information from various data sources such as e-mail with attachments, mail post cards (e.g., wedding

invitation), notes and list applications, personal calendar entries, and Global Positioning System (GPS) and time tracking features. The PSA receives updated market trend information from various social media and marketing data sources and applies a personal shopper profile to identify suggested items. The PSA reminds the user of required shopping and makes purchasing suggestions based on the event type. It tracks personal shopping habits and adjusts shopping list if observes changes in a shopping pattern.

To generate suggestions for a shopping list, the PSA can process information about the user's travel plans and evaluates shopping trends. The PSA is aware of the shopper's lifestyle and stores data about personal shopping activities, including preferred stores. To generate and update a personal shopping list, the PSA considers and evaluates the following shopping-related components:

• Driving habits, to suggest where to shop • Calendar events, to suggest gift purchases • Previously purchased items and feedback on those items • Daily shopping patterns • Personal travel, vacation plans, or busy times
• Context such as type and places of a user's calendar events

The main point of novelty of the proposed solution is the ability of the system to learn

while processing customer shopping experiences and then dynamically generate

personalized recommendations for the next shopping list. Cognitive abilities of this recommendation system utilize Machine Learning (ML) and Natural Language Processing (NLP) methods and techniques.

The Personal Shopping Assistant is comprised of the following components:

PSA Repository: located on the cloud, consists of following PSA components:
- PSA event models

1


Page 02 of 5


- PSA preferences


- PSA personal history

    - PSA Item Catalog data
PSA Item Catalog: stores details of shopping items and the associated mappings to the event types. The PSA Item Catalog...