Context-Based Routine Model for Personalized Recommendations

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing recommendation systems are limited in their ability to provide users with new options beyond their usual preferences, as they require extensive user behavior data and struggle to capture nuanced interests, leading to recommendations that are often repetitive and fail to explore new possibilities.

Innovation Solution

A context-based routine model is developed to analyze a user's historical data, identifying transitions between contexts to build a customized recommendation agent that selects recommendations based on the user's current or predicted context, incorporating both routine and personality metrics.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If existing recommendation systems use general genre and category preferences to generate user models, then the system complexity is low and ease of operation is high, but the recommendation precision and ability to capture nuanced user interests deteriorates

Engineering Contradiction:

Improverecommendation precisionVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments user preferences into multiple dimensions including general genre preferences, routine contexts (time, location, activity), and personality traits. This segmentation allows the system to capture nuanced user interests with higher precision while organizing complexity into manageable separate components that can be processed independently

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent adds new dimensions to the recommendation space by incorporating contextual dimensions (time, location, activity) and personality dimensions beyond traditional genre/category dimensions. This multi-dimensional approach enables more precise recommendations by considering user behavior across multiple axes simultaneously

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

2Adaptability or versatility

If existing recommendation systems recommend similar items based on observed user behavior, then the system requires less data and operates faster, but the ability to provide novel options and explore new possibilities deteriorates

Engineering Contradiction:

Improveability to provide new optionsVSAvoiddata collection time

Core Design Contradiction:

Adaptability or versatilityVSLoss of time

Solution Approach 1:

The patent performs preliminary action by collecting and analyzing mobile device data (location, activity, time) continuously in the background before recommendations are needed. This pre-collection of contextual data and pre-building of routine models enables the system to provide novel recommendations quickly when needed, without requiring extensive real-time data collection

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent introduces routine context models and personality profiles as intermediary representations that mediate between raw observed behavior and recommendation generation. These intermediaries capture nuanced user patterns and preferences, enabling the system to explore new possibilities and provide adaptable recommendations without requiring extensive direct observation of every user interaction

Inventive Principle:

Principle #24Intermediary (Mediator)

3Adaptability or versatility

If existing recommendation systems focus on specific user preferences in particular scenarios, then the recommendation precision for those scenarios is high, but the versatility to help users discover new interests and contexts deteriorates

Engineering Contradiction:

Improverecommendation versatilityVSAvoidrecommendation precision

Core Design Contradiction:

Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent creates a universal recommendation framework that handles multiple user needs simultaneously: maintaining precision for specific user preferences through routine context matching, while providing versatility through personality-based recommendations and contextual exploration. The system can adapt to different scenarios (specific preference matching, discovery, serendipity) using the same multi-dimensional model

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS9179250B2Recommendation agent using a routine model determined from mobile device data

Publication Date: 2015.11.03 VULCAN TECHNOLOGIES LLC

AI summary

A user's context history is analyzed to identify transitions between contexts therein. The identified transitions are used to build a routine model for the user. The routine model includes transition rules indicating a source context, a destination context, and, optionally, a probability that the user will transition from the source context to the destination context, based on the user's historical behavior. A customized recommendation agent for the user is built using the routine model. The customized recommendation agent selects recommendations from a corpus to present to the user, based on the routine model and the user's current or predicted future context.