Automated Digital Review Summarization via Aspect Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Customers face difficulties in efficiently evaluating digital item reviews due to the vast amount of information and time required to read through multiple reviews, especially when looking for specific features or experiences, which hinders their purchasing decisions.
Innovation Solution
A system that automatically generates and provides digital item review summaries by clustering relevant sentences from user reviews based on aspect data, using techniques like pointer-generator networks, to present more focused and relevant information to customers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If customers read through all reviews to evaluate an item, then they can make informed purchasing decisions, but it consumes excessive time and effort
Solution Approach 1:
The system extracts and highlights only the most relevant sentences from reviews that pertain to the customer's specific interests and the item's key features, rather than requiring customers to read through entire reviews. This extraction process isolates critical information while filtering out unnecessary content.
Solution Approach 2:
The system introduces an intermediary processing layer that automatically analyzes reviews, identifies relevant information based on customer profiles and item attributes, and presents synthesized insights. This intermediary acts as a bridge between the full review content and the customer's decision-making needs.
2Productivity
If customers filter reviews by specific features, then they can find relevant information faster, but the system complexity increases
Solution Approach 1:
The system performs preliminary analysis of reviews by pre-identifying and tagging sentences that relate to specific item features and customer interests before the customer even views them. This advance processing organizes information in a ready-to-present format, eliminating the need for complex real-time filtering interfaces.
Solution Approach 2:
The system automatically performs the filtering and selection of relevant review content based on customer profiles and item characteristics without requiring the customer to manually configure complex filter settings. The system serves itself by autonomously determining what information is most valuable to each customer.
3Loss of information
If the system presents all review information, then customers have complete data for analysis, but the information becomes difficult to evaluate and process
Solution Approach 1:
The system segments review information into distinct, manageable units organized by relevance to customer interests and item features. Instead of presenting a continuous wall of text, reviews are divided into highlighted sentences and grouped insights that can be easily scanned and evaluated.
Solution Approach 2:
Different portions of review information are presented with different levels of emphasis and detail based on their relevance to the customer's specific interests. Critical information receives prominent highlighting and prioritized positioning, while less relevant content is de-emphasized or omitted.
Data Source
AI summary
This application relates to apparatus and methods for automatically determining and providing item reviews to users. In some examples, a computing device obtains review data identifying one or more reviews for each of a plurality of items. The computing device determines keywords for each of the items based on parsing the review data corresponding to each of items. The computing device may obtain data identifying engagement of items for a user during a browsing session, such as items a user has clicked on. The computing device may also obtain data identifying previous purchase transactions, or previous review postings, for the user. The computing device then determines, based on the obtained data, which keywords may be of interest the user. In some examples, the keywords are used to identify reviews of an item for the user. In some examples, summaries of the reviews are generated and displayed to the user.