Personalized Hot Topics Ranking via User Profile Similarity

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current search engines provide generic rankings of popular topics that may not be relevant to individual users, as they are based on collective user statistics, often including topics of little interest to a specific user.

Innovation Solution

A system and method that personalize popular topic rankings by calculating a personalization score based on similarities between a user profile and topic profiles, adjusting the generic ranking scores to provide a revised ranking that is more relevant to the individual user.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If generic popular topic rankings are provided to all users, then the system maintains simplicity and broad applicability, but the relevance to individual users deteriorates

Engineering Contradiction:
Improverelevance to individual userVSAvoidsystem complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent segments the generic popular topic rankings by creating personalized rankings for each user based on their profile characteristics. The system divides the single generic ranking into multiple user-specific rankings, allowing each user to see topics tailored to their interests while maintaining the underlying generic ranking structure as a base.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies local quality by customizing the ranking presentation for each individual user based on their specific profile attributes. Instead of uniform treatment for all users, the system adjusts the visibility and positioning of topics locally for each user based on their demonstrated interests and characteristics.

Inventive Principle:
Principle #3Local quality

2Measurement precision

If personalized topic rankings are created for each user, then the relevance to individual users improves, but the computational complexity and processing time worsen

Engineering Contradiction:
Improveranking accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary action by pre-computing user profiles and storing them in advance. The system analyzes user behavior, interests, and characteristics beforehand to create ready-to-use user profiles, so that when generating personalized rankings, the system can quickly retrieve and apply pre-existing profile data rather than computing from scratch.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent uses parameter changes by adjusting ranking parameters based on user profile attributes. The system modifies the weighting and scoring parameters of topic rankings according to individual user characteristics, allowing efficient computation by changing numerical parameters rather than fundamentally redesigning the ranking algorithm for each user.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS8762326B1Personalized hot topics
Publication Date: 2014.06.24 GOOGLE LLC
  • US8762326B1 patent drawing
  • US8762326B1 patent drawing
  • US8762326B1 patent drawing

AI summary

A server device receives a user request and retrieves, based on the user request, a list of popular topics, a generic ranking score for each topic in the list of popular topics, and a topic profile for a first topic in the list of popular topics. The server device identifies a user profile for the user and determines a personalization score for the first topic in the list of popular topics, where the personalization score for the first topic is based on one or more similarities between the user profile and the topic profile for the first topic. The server device determines a revised ranking score for the first topic in the list of popular topics based on the personalization score and the generic ranking score for the first topic; and ranks the topics in the list of popular topics, using the revised ranking score for the first topic.