Browser Plugin Review Prompting via NLP Tagging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems lack an efficient method to guide users in authoring comprehensive product reviews by leveraging crowdsourced data and natural language processing to prompt users on relevant topics and sentiments, often resulting in incomplete or inaccurate reviews.
Innovation Solution
A method that examines user browsing data to classify current activities, performs web crawling for crowdsourced product review data, subjects it to natural language processing to extract data tags, and communicates prompting data to users to guide the authoring of product reviews, ensuring a more comprehensive and accurate review process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If web crawling is performed to obtain crowdsourced product review data, then the comprehensiveness of review data is improved, but the time consumption and system complexity increase
Solution Approach 1:
The system performs web crawling and data processing in advance, before the user actually needs to write a review. By proactively collecting crowdsourced review data, extracting tags, and generating prompting information beforehand, the system eliminates the need for time-consuming data collection at the moment of review writing, thus resolving the contradiction between data completeness and time consumption.
Solution Approach 2:
The system automatically performs web crawling, data extraction, tag generation, and prompt creation without requiring user intervention. This automated self-service approach handles the time-consuming data collection and processing tasks autonomously, allowing users to benefit from comprehensive review data without investing their own time in these processes.
2Measurement precision
If natural language processing is applied to extract data tags from crowdsourced reviews, then the quality and accuracy of review guidance is improved, but the computational complexity and processing time increase
Solution Approach 1:
The system extracts only the essential and relevant information from crowdsourced reviews through natural language processing - specifically focusing on extracting meaningful tags and keywords that represent key review topics. By selectively extracting only the necessary information rather than processing and storing all review data, the system achieves high accuracy in review guidance while managing computational complexity.
Solution Approach 2:
The system introduces an intermediary processing layer that translates unstructured crowdsourced review text into structured tags and prompts. This intermediary NLP layer acts as a mediator between raw review data and the user interface, converting complex unstructured data into simplified, actionable guidance elements that are easy for users to consume.
3Productivity
If the system proactively provides prompting data to users, then the productivity of review authoring is improved, but the system complexity and data processing requirements increase
Solution Approach 1:
The system monitors user browsing activities and provides dynamic feedback in the form of contextual prompts and suggestions during the review authoring process. By continuously observing user behavior and adjusting the prompting information accordingly, the system accelerates review writing by guiding users through relevant topics and sentiments without requiring complex pre-programming of all possible scenarios.
Solution Approach 2:
The system performs activity classification and prepares prompting data in advance based on detected user intentions. By proactively analyzing browsing patterns and pre-generating relevant prompts before the user actually starts writing, the system reduces the cognitive load and time required for review authoring while managing complexity through targeted, context-aware processing.
Data Source
AI summary
Methods, computer program products, and systems are presented. The method computer program products, and systems can include, for instance: examining user browsing data from a browser plugin of a client computer device, the client computing device being associated to the user; determining a current activity classification of the user in dependence on the examining user browsing data from the browser plugin of the client computer device; in response to the current activity classification satisfying a criterion, performing web crawling to obtain crowdsourced product review data of one or more product, the crowdsourced product review data comprising data from product reviews by multiple reviewers other than the user and including data from reviews from multiple product review sites; subjecting the crowdsourced product review data to natural language processing for extracting data tags associated to respective product reviews of the crowdsourced product review data; determining prompting data in dependence on the extracting data tags associated to respective product reviews of the crowdsourced product review data; and communicating the prompting data to the user, the prompting data guiding the user in the authoring of a product review of a certain product.