Ben Hachey

Yidong Gan, Maciej Rybinski, Ben Hachey et al.

Clinical coding is crucial for healthcare billing and data analysis. Manual clinical coding is labour-intensive and error-prone, which has motivated research towards full automation of the process. However, our analysis, based on US English electronic health records and automated coding research using these records, shows that widely used evaluation methods are not aligned with real clinical contexts. For example, evaluations that focus on the top 50 most common codes are an oversimplification, as there are thousands of codes used in practice. This position paper aims to align AI coding research more closely with practical challenges of clinical coding. Based on our analysis, we offer eight specific recommendations, suggesting ways to improve current evaluation methods. Additionally, we propose new AI-based methods beyond automated coding, suggesting alternative approaches to assist clinical coders in their workflows.

Xiang Dai, Sarvnaz Karimi, Ben Hachey et al.

Recent studies on domain-specific BERT models show that effectiveness on downstream tasks can be improved when models are pretrained on in-domain data. Often, the pretraining data used in these models are selected based on their subject matter, e.g., biology or computer science. Given the range of applications using social media text, and its unique language variety, we pretrain two models on tweets and forum text respectively, and empirically demonstrate the effectiveness of these two resources. In addition, we investigate how similarity measures can be used to nominate in-domain pretraining data. We publicly release our pretrained models at https://bit.ly/35RpTf0.

Xiang Dai, Sarvnaz Karimi, Ben Hachey et al.

Unlike widely used Named Entity Recognition (NER) data sets in generic domains, biomedical NER data sets often contain mentions consisting of discontinuous spans. Conventional sequence tagging techniques encode Markov assumptions that are efficient but preclude recovery of these mentions. We propose a simple, effective transition-based model with generic neural encoding for discontinuous NER. Through extensive experiments on three biomedical data sets, we show that our model can effectively recognize discontinuous mentions without sacrificing the accuracy on continuous mentions.

Nicky Ringland, Xiang Dai, Ben Hachey et al.

Named entity recognition (NER) is widely used in natural language processing applications and downstream tasks. However, most NER tools target flat annotation from popular datasets, eschewing the semantic information available in nested entity mentions. We describe NNE---a fine-grained, nested named entity dataset over the full Wall Street Journal portion of the Penn Treebank (PTB). Our annotation comprises 279,795 mentions of 114 entity types with up to 6 layers of nesting. We hope the public release of this large dataset for English newswire will encourage development of new techniques for nested NER.

Xiang Dai, Sarvnaz Karimi, Ben Hachey et al.

Word vectors and Language Models (LMs) pretrained on a large amount of unlabelled data can dramatically improve various Natural Language Processing (NLP) tasks. However, the measure and impact of similarity between pretraining data and target task data are left to intuition. We propose three cost-effective measures to quantify different aspects of similarity between source pretraining and target task data. We demonstrate that these measures are good predictors of the usefulness of pretrained models for Named Entity Recognition (NER) over 30 data pairs. Results also suggest that pretrained LMs are more effective and more predictable than pretrained word vectors, but pretrained word vectors are better when pretraining data is dissimilar.

Andrew Chisholm, Will Radford, Ben Hachey

We investigate the generation of one-sentence Wikipedia biographies from facts derived from Wikidata slot-value pairs. We train a recurrent neural network sequence-to-sequence model with attention to select facts and generate textual summaries. Our model incorporates a novel secondary objective that helps ensure it generates sentences that contain the input facts. The model achieves a BLEU score of 41, improving significantly upon the vanilla sequence-to-sequence model and scoring roughly twice that of a simple template baseline. Human preference evaluation suggests the model is nearly as good as the Wikipedia reference. Manual analysis explores content selection, suggesting the model can trade the ability to infer knowledge against the risk of hallucinating incorrect information.

Bo Han, Will Radford, Anaïs Cadilhac et al.

Text generation is increasingly common but often requires manual post-editing where high precision is critical to end users. However, manual editing is expensive so we want to ensure this effort is focused on high-value tasks. And we want to maintain stylistic consistency, a particular challenge in crowd settings. We present a case study, analysing human post-editing in the context of a template-based biography generation system. An edit flow visualisation combined with manual characterisation of edits helps identify and prioritise work for improving end-to-end efficiency and accuracy.

Will Radford, Andrew Chisholm, Ben Hachey et al.

We report on an exploratory analysis of Emoji Dick, a project that leverages crowdsourcing to translate Melville's Moby Dick into emoji. This distinctive use of emoji removes textual context, and leads to a varying translation quality. In this paper, we use statistical word alignment and part-of-speech tagging to explore how people use emoji. Despite these simple methods, we observed differences in token and part-of-speech distributions. Experiments also suggest that semantics are preserved in the translation, and repetition is more common in emoji.

Xavier Holt, Will Radford, Ben Hachey

The timeline generation task summarises an entity's biography by selecting stories representing key events from a large pool of relevant documents. This paper addresses the lack of a standard dataset and evaluative methodology for the problem. We present and make publicly available a new dataset of 18,793 news articles covering 39 entities. For each entity, we provide a gold standard timeline and a set of entity-related articles. We propose ROUGE as an evaluation metric and validate our dataset by showing that top Google results outperform straw-man baselines.