# Using Graphs of Classifiers to Impose Declarative Constraints on Semi-supervised Learning

## TL;DR

This paper introduces a declarative language for modeling semi-supervised learning algorithms, enabling the combination of multiple heuristics and achieving improved results in text classification and relation extraction tasks.

## Contribution

It presents a novel declarative framework for representing and combining SSL heuristics, including both traditional and domain-specific methods, optimized via Bayesian methods.

## Key findings

- Modest improvements on link-based classification benchmarks.
- State-of-the-art results on relation extraction in two domains.
- Demonstrates effectiveness of combining multiple SSL heuristics.

## Abstract

We propose a general approach to modeling semi-supervised learning (SSL) algorithms. Specifically, we present a declarative language for modeling both traditional supervised classification tasks and many SSL heuristics, including both well-known heuristics such as co-training and novel domain-specific heuristics. In addition to representing individual SSL heuristics, we show that multiple heuristics can be automatically combined using Bayesian optimization methods. We experiment with two classes of tasks, link-based text classification and relation extraction. We show modest improvements on well-studied link-based classification benchmarks, and state-of-the-art results on relation-extraction tasks for two realistic domains.