Social Platform Information Pre-processing with Blacklist Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current social platform systems lack efficient methods for monitoring and processing information from social platforms, leading to inaccurate sentiment analysis due to the presence of non-accurate information such as false posts, spam, and repeat postings, which wastes network resources and makes it difficult to identify actionable information.
Innovation Solution
A system that preprocesses information from social platforms using a manually created blacklist and watchlist, filtering out non-accurate information and categorizing sentiment, allowing users to generate tickets based on preprocessed data, with features like graphical user interfaces for blacklist management and sentiment analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If all information from social platforms is monitored and processed, then comprehensive sentiment analysis coverage is achieved, but processing accuracy deteriorates due to non-accurate information such as false posts, spam, and repeat postings
Solution Approach 1:
The system performs preliminary filtering by creating and applying blacklists and watchlists before sentiment analysis. User accounts are pre-identified as suspicious based on posting patterns, and their content is flagged or excluded before the main processing pipeline, preventing non-accurate information from contaminating the sentiment analysis results
Solution Approach 2:
The system extracts and removes content from blacklisted and suspicious user accounts from the overall information stream. By separating problematic content into distinct categories (blacklisted, watchlist, suspicious), the system processes only high-quality information for sentiment analysis, improving measurement precision while reducing the effective quantity of processed data
2Measurement precision
If manual monitoring and analysis of social platform information is performed, then accurate identification of problematic information is achieved, but time consumption and resource usage increase significantly
Solution Approach 1:
The system automatically monitors social platform information, identifies suspicious posting patterns, creates blacklists and watchlists, and generates tickets without requiring continuous manual intervention. The automated detection of repeat postings, spam patterns, and suspicious behavior enables the system to serve itself in maintaining data quality, significantly reducing time loss while preserving identification accuracy
Solution Approach 2:
The system implements feedback loops where sentiment analysis results and ticket creation outcomes inform ongoing blacklist and watchlist management. This continuous feedback mechanism allows the system to automatically adjust its filtering criteria based on actual performance, maintaining high identification accuracy while minimizing manual time investment
3Productivity
If non-accurate information is included in sentiment analysis, then comprehensive data processing is maintained, but network resources are wasted on unnecessary processing
Solution Approach 1:
The system extracts and isolates content from blacklisted and suspicious accounts before the main processing pipeline. By removing these non-accurate information sources upfront, the system processes only valuable content through the sentiment analysis and ticket creation workflows, maintaining productivity for meaningful information while eliminating wasted network resources on spam and false posts
Data Source
AI summary
A method for pre-processing information, performed by at least one processor, is provided. The method includes retrieving information from a social platform based on predetermined keywords defined by a system user, obtaining a blacklist of user accounts, the blacklist including user account information and a blacklist reason for adding the user account information to the blacklist; pre-processing the retrieved information, based on the blacklist, determining a sentiment value associated with the pre-processed information and assigning the pre-processed information to a sentiment category based on the sentiment value, and displaying a first graphical user interface (GUI) for receiving user input information to generate a ticket for the pre-processed information. A list of the retrieved information may be displayed, wherein the user accounts included in the blacklist are distinguishably displayed on the GUI.