ML Event Tag Prediction via Self-Service Data Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Developing high-quality machine-learning models for predicting event tags, such as IT incident categories, is challenging due to the need for large amounts of manually labeled training data, which is time-consuming and costly to compile.
Innovation Solution
A system that generates training examples from unlabeled event data by leveraging subject-matter expert knowledge, using n-gram features and a machine-learning model to predict event tags, allowing for training without relying on extensive manual labeling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manually labeled training data is used to train machine-learning models, then model prediction accuracy is improved, but time consumption and cost increase significantly
Solution Approach 1:
The system automatically generates training data by having the machine-learning model predict tags for unlabeled events, then using these predictions as training labels. This self-service approach eliminates the need for manual labeling while still providing sufficient training data for model improvement.
Solution Approach 2:
The system performs preliminary tagging by having the machine-learning model generate predicted tags for unlabeled events before these predictions are used as training data. This preliminary action creates a pool of training examples that can be used to retrain and improve the model without waiting for manual labeling.
2Measurement precision
If manually labeled training data is used to train machine-learning models, then model prediction accuracy is improved, but cost increases significantly
Solution Approach 1:
The system generates its own training data by having the machine-learning model predict tags for unlabeled events and using these predictions as training labels. This eliminates the need to purchase or pay for manually labeled datasets, significantly reducing costs while maintaining model improvement.
Solution Approach 2:
The system creates copies of unlabeled events by generating predicted tags for them, then uses these copied examples as training data. This allows the model to learn from synthesized training examples without incurring the costs associated with obtaining manually labeled data.
3Manufacturing precision
If extensive manual labeling is performed to create training data, then training data quality is improved, but productivity of event processing decreases
Solution Approach 1:
The system automatically generates training data by having the machine-learning model predict tags for unlabeled events. This self-service data generation process maintains training data quality while eliminating the manual labeling bottleneck that reduces event processing productivity.
Solution Approach 2:
The system enables continuous event processing by automatically generating training data in the background without interrupting the main event processing workflow. This maintains productivity while still improving model quality through continuous training data generation.
Data Source
AI summary
Methods, computer systems, and apparatus, including computer programs encoded on computer storage media, for training a machine-learning model for predicting event tags. The system obtains event data that specifies, for each of a plurality of events, a respective set of text fields characterizing the respective event. The system generates, from the event data, encoded language features for the plurality of events. The system also obtains knowledge data that specifies information of the event data. The system generates, from the event data and the knowledge data, tag data specifying a respective tag for each of the plurality of events. The system generates, from the tag data and the encoded language features, a respective encoded feature vector for each of the plurality of events. The system combines the tag data with the encoded feature vectors to generate a plurality of training examples.