Iterative Data Programming for Expanding Text Classification Corpora
This work addresses the need for cost-effective labeling in text classification, particularly for conversational agents, but it appears incremental as it builds on existing data programming techniques.
The paper tackles the problem of expensive labeled data for text classification by introducing an iterative data programming method that generates weak models with minimal supervision and identifies sparse examples from unlabeled data, showing empirical improvements on sentence classification tasks such as intent recognition in conversational agents.
Real-world text classification tasks often require many labeled training examples that are expensive to obtain. Recent advancements in machine teaching, specifically the data programming paradigm, facilitate the creation of training data sets quickly via a general framework for building weak models, also known as labeling functions, and denoising them through ensemble learning techniques. We present a fast, simple data programming method for augmenting text data sets by generating neighborhood-based weak models with minimal supervision. Furthermore, our method employs an iterative procedure to identify sparsely distributed examples from large volumes of unlabeled data. The iterative data programming techniques improve newer weak models as more labeled data is confirmed with human-in-loop. We show empirical results on sentence classification tasks, including those from a task of improving intent recognition in conversational agents.