Do Not Have Enough Data? Deep Learning to the Rescue!

Based on recent advances in natural language modeling and those in text generation capabilities, we propose a novel data augmentation method for text classification tasks. We use a powerful pre-trained neural network model to artificially synthesize new labeled data for supervised learning. We mainly focus on cases with scarce labeled data. Our method, referred to as language-model-based data augmentation (LAM- BADA), involves fine-tuning a state-of-the-art language generator to a specific task through an initial training phase on the existing (usually small) labeled data. Using the fine-tuned model and given a class label, new sentences for the class are generated. Our process then filters these new sentences by using a classifier trained on the original data. In a series of experiments, we show that LAMBADA improves classi- fiers’ performance on a variety of datasets. Moreover, LAM- BADA significantly improves upon the state-of-the-art techniques for data augmentation, specifically those applicable to text classification tasks with little data.

Authors

• Ateret Anaby-Tavor
• Boaz Carmeli Technion - Israel Institute of Technology
• Esther Goldbraich
• Amir Kantor
• George Kour
• Segev Shlomov
• Naama Tepper
• Naama Zwerdling

Keywords

No keywords are indexed for this paper.