Semantic Sweeping of Metadata-Enriched Service Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The complexity of large-scale software services makes it difficult to collect and analyze user feedback, which often comes from multiple sources and is disjointed, making it challenging for developers to quickly identify and address service issues.
Innovation Solution
A system and method that processes service data by enriching it with metadata, pre-cleaning, determining semantically similar data points, generating similarity scores, and analyzing clusters to identify significant issues, thereby generating service alerts and trend data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If service data is collected from multiple sources (service calls, on-line support, social media), then the comprehensiveness of service feedback is improved, but the complexity and disjointed nature of the data increases
Solution Approach 1:
The patent segments the service data into structured fields including service_id, service_name, service_version, service_platform, service_channel, service_status, and service_content. This segmentation allows data from multiple sources to be organized into consistent categories, reducing complexity while maintaining comprehensiveness.
Solution Approach 2:
The patent creates a universal data structure that can accommodate service data from multiple sources (service calls, on-line support, social media) through standardized fields. This multi-functional schema allows a single system to process diverse data types without requiring separate processing mechanisms for each source.
2Quantity of substance
If large quantities of disjointed service data are collected, then the coverage of service issues is improved, but the difficulty of analyzing and identifying actual service issues increases
Solution Approach 1:
The patent performs preliminary actions by pre-defining service categories, issue types, and analysis dimensions before data collection. This preparation enables systematic processing of large datasets, transforming the analysis of disjointed data into a structured process that identifies service issues more efficiently.
Solution Approach 2:
The patent introduces an intermediary analysis layer that processes raw service data through standardized fields and categories before presenting service issues. This intermediary structure mediates between the volume of raw data and the need for clear issue identification, making detection easier despite data quantity.
3Speed
If service data is processed in near real time, then the responsiveness to service issues is improved, but the computational resources and processing complexity increase
Solution Approach 1:
The patent segments the data processing into distinct stages: data collection, data enrichment with metadata, preliminary cleaning, similarity analysis, and issue detection. This segmentation enables near real-time processing by handling operations in manageable batches, reducing the complexity burden on any single processing step.
Solution Approach 2:
The patent changes parameters such as data granularity, metadata fields, and analysis thresholds to optimize processing speed. By adjusting these parameters, the system can process data in near real-time while managing computational resources effectively, balancing responsiveness with processing complexity.
4Measurement precision
If metadata parameters are enriched to service data, then the accuracy of service issue identification is improved, but the data processing time and complexity increase
Solution Approach 1:
The patent performs metadata enrichment as a preliminary action during data collection, rather than as a subsequent processing step. By preparing and attaching metadata parameters (service_id, service_name, service_version, service_platform, serviceChannel, service_status) at the time of data intake, the system reduces later processing time while maintaining high identification accuracy.
Data Source
AI summary
A system for detecting service issues within multi-sourced service data. The system includes a memory and one or more electronic processors coupled to the memory. The electronic processors are configured to receive one or more data sets in near real time, and to enrich the dataset with one or more metadata parameters, pre-clean the data within the dataset, and determine one or more data points within the dataset that are semantically similar to each other. The electronic processors are also configured to generate a similarity score for each of the semantically similar data points, and determine one or more significant clusters within the dataset within a predefined lookback window. The electronic processors are also configured to analyze the determined significant clusters to determine the existence of one or more service issues, and generate a service alert based on the analysis determining that one or more service issues are present.