CUTIE: Learning to Understand Documents with Convolutional Universal Text Information Extractor
This addresses the need for automated document processing in areas such as accounting and finance, offering a more efficient alternative to rule-based or NER methods, though it appears incremental as it builds on existing CNN approaches.
The paper tackles the problem of extracting key information from documents like receipts by proposing CUTIE, a model that uses convolutional neural networks on gridded text to leverage both semantic and spatial information, achieving state-of-the-art performance with 4,484 labeled receipts and better speed and accuracy than NER-based methods.
Extracting key information from documents, such as receipts or invoices, and preserving the interested texts to structured data is crucial in the document-intensive streamline processes of office automation in areas that includes but not limited to accounting, financial, and taxation areas. To avoid designing expert rules for each specific type of document, some published works attempt to tackle the problem by learning a model to explore the semantic context in text sequences based on the Named Entity Recognition (NER) method in the NLP field. In this paper, we propose to harness the effective information from both semantic meaning and spatial distribution of texts in documents. Specifically, our proposed model, Convolutional Universal Text Information Extractor (CUTIE), applies convolutional neural networks on gridded texts where texts are embedded as features with semantical connotations. We further explore the effect of employing different structures of convolutional neural network and propose a fast and portable structure. We demonstrate the effectiveness of the proposed method on a dataset with up to $4,484$ labelled receipts, without any pre-training or post-processing, achieving state of the art performance that is much better than the NER based methods in terms of either speed and accuracy. Experimental results also demonstrate that the proposed CUTIE model being able to achieve good performance with a much smaller amount of training data.