Text Understanding from Scratch
This work addresses the problem of text understanding for AI researchers by proposing a novel deep learning approach that bypasses traditional linguistic preprocessing, though it is incremental in building on existing ConvNet methods.
The paper tackles text understanding by applying temporal convolutional networks directly to character-level inputs, achieving strong performance on tasks like ontology classification, sentiment analysis, and text categorization without relying on linguistic structures.
This article demontrates that we can apply deep learning to text understanding from character-level inputs all the way up to abstract text concepts, using temporal convolutional networks (ConvNets). We apply ConvNets to various large-scale datasets, including ontology classification, sentiment analysis, and text categorization. We show that temporal ConvNets can achieve astonishing performance without the knowledge of words, phrases, sentences and any other syntactic or semantic structures with regards to a human language. Evidence shows that our models can work for both English and Chinese.