skip to main content

Explore Research Products in the NSF-PAR

Title: A New Approach to Animacy Detection
Animacy is a necessary property for a referent to be an agent, and thus animacy detection is useful for a variety of natural language processing tasks, including word sense disambiguation, co-reference resolution, semantic role labeling, and others. Prior work treated animacy as a word-level property, and has developed statistical classifiers to classify words as either animate or inanimate. We discuss why this approach to the problem is ill-posed, and present a new approach based on classifying the animacy of co-reference chains. We show that simple voting approaches to inferring the animacy of a chain from its constituent words perform relatively poorly, and then present a hybrid system merging supervised machine learning (ML) and a small number of hand-built rules to compute the animacy of referring expressions and co-reference chains. This method achieves state of the art performance. The supervised ML component leverages features such as word embeddings over referring expressions, parts of speech, and grammatical and semantic roles. The rules take into consideration parts of speech and the hypernymy structure encoded in WordNet. The system achieves an F1 of 0.88 for classifying the animacy of referring expressions, which is comparable to state of the art results for classifying the animacy of words, and achieves an F1 of 0.75 for classifying the animacy of coreference chains themselves. We release our training and test dataset, which includes 142 texts (all narratives) comprising 156,154 words, 34,698 referring expressions, and 10,941 co-reference chains. We test the method on a subset of the OntoNotes dataset, showing using manual sampling that animacy classification is 90% +/- 2% accurate for coreference chains, and 92% +/- 1% for referring expressions. The data also contains 46 folktales, which present an interesting challenge because they often involve characters who are members of traditionally inanimate classes (e.g., stoves that walk, trees that talk). We show that our system is able to detect the animacy of these unusual referents with an F1 of 0.95.  more » « less
Award ID(s):
1749917
NSF-PAR ID:
10140307
Author(s) / Creator(s):
; ;
Date Published:
Journal Name:
Proceedings of the 27th International Conference on Computational Linguistics
Page Range / eLocation ID:
1 - 12
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ; ; ( , 1st Workshop on Artificial Intelligence for Narratives (AI4N 2020))
    null (Ed.)
    Animacy is the characteristic of a referent beingable to independently carry out actions in a storyworld (e.g., movement, communication). It is anecessary property of characters in stories, and sodetecting animacy is an important step in automaticstory understanding; it is also potentially useful formany other natural language processing tasks suchas word sense disambiguation, coreference resolu-tion, character identification, and semantic role la-beling. Recent work by Jahanet al.[2018]demon-strated a new approach to detecting animacy whereanimacy is considered a direct property of corefer-ence chains (and referring expressions) rather thanwords. In Jahanet al., they combined hand-builtrules and machine learning (ML) to identify the an-imacy of referring expressions and used majorityvoting to assign the animacy of coreference chains,and reported high performance of up to 0.90F1. Inthis short report we verify that the approach gener-alizes to two different corpora (OntoNotes and theCorpus of English Novels) and we confirmed thatthe hybrid model performs best, with the rule-basedmodel in second place. Our tests apply the animacyclassifier to almost twice as much data as Jahanetal.’s initial study. Our results also strongly suggest,as would be expected, the dependence of the mod-els on coreference chain quality. We release ourdata and code to enable reproducibility. 
    more » « less
  2. ; ; ; ; ; ; ; ; ( , ACM Transactions on the Web)
    Event detection in power systems aims to identify triggers and event types, which helps relevant personnel respond to emergencies promptly and facilitates the optimization of power supply strategies. However, the limited length of short electrical record texts causes severe information sparsity, and numerous domain-specific terminologies of power systems makes it difficult to transfer knowledge from language models pre-trained on general-domain texts. Traditional event detection approaches primarily focus on the general domain and ignore these two problems in the power system domain. To address the above issues, we propose a Multi-Channel graph neural network utilizing Type information for Event Detection in power systems, named MC-TED , leveraging a semantic channel and a topological channel to enrich information interaction from short texts. Concretely, the semantic channel refines textual representations with semantic similarity, building the semantic information interaction among potential event-related words. The topological channel generates a relation-type-aware graph modeling word dependencies, and a word-type-aware graph integrating part-of-speech tags. To further reduce errors worsened by professional terminologies in type analysis, a type learning mechanism is designed for updating the representations of both the word type and relation type in the topological channel. In this way, the information sparsity and professional term occurrence problems can be alleviated by enabling interaction between topological and semantic information. Furthermore, to address the lack of labeled data in power systems, we built a Chinese event detection dataset based on electrical Power Event texts, named PoE . In experiments, our model achieves compelling results not only on the PoE dataset, but on general-domain event detection datasets including ACE 2005 and MAVEN. 
    more » « less
  3. ; ; ; ; ( , Interspeech)
    Current leading mispronunciation detection and diagnosis (MDD) systems achieve promising performance via end-to-end phoneme recognition. One challenge of such end-to-end solutions is the scarcity of human-annotated phonemes on natural L2 speech. In this work, we leverage unlabeled L2 speech via a pseudo-labeling (PL) procedure and extend the fine-tuning approach based on pre-trained self-supervised learning (SSL) models. Specifically, we use Wav2vec 2.0 as our SSL model, and fine-tune it using original labeled L2 speech samples plus the created pseudo-labeled L2 speech samples. Our pseudo labels are dynamic and are produced by an ensemble of the online model on-the-fly, which ensures that our model is robust to pseudo label noise. We show that fine-tuning with pseudo labels achieves a 5.35% phoneme error rate reduction and 2.48% MDD F1 score improvement over a labeled-samples-only finetuning baseline. The proposed PL method is also shown to outperform conventional offline PL methods. Compared to the state-of-the-art MDD systems, our MDD solution produces a more accurate and consistent phonetic error diagnosis. In addition, we conduct an open test on a separate UTD-4Accents dataset, where our system recognition outputs show a strong correlation with human perception, based on accentedness and intelligibility. 
    more » « less
  4. ; ( , Journal of Intelligent Manufacturing)
    null (Ed.)
    The number of published manufacturing science digital articles available from scientifc journals and the broader web have exponentially increased every year since the 1990s. To assimilate all of this knowledge by a novice engineer or an experienced researcher, requires signifcant synthesis of the existing knowledge space contained within published material, to fnd answers to basic and complex queries. Algorithmic approaches through machine learning and specifcally Natural Language Processing (NLP) on a domain specifc area such as manufacturing, is lacking. One of the signifcant challenges to analyzing manufacturing vocabulary is the lack of a named entity recognition model that enables algorithms to classify the manufacturing corpus of words under various manufacturing semantic categories. This work presents a supervised machine learning approach to categorize unstructured text from 500K+manufacturing science related scientifc abstracts and labelling them under various manufacturing topic categories. A neural network model using a bidirectional long-short term memory, plus a conditional random feld (BiLSTM+CRF) is trained to extract information from manufacturing science abstracts. Our classifer achieves an overall accuracy (f1-score) of 88%, which is quite near to the state-of-the-art performance. Two use case examples are presented that demonstrate the value of the developed NER model as a Technical Language Processing (TLP) workfow on manufacturing science documents. The long term goal is to extract valuable knowledge regarding the connections and relationships between key manufacturing concepts/entities available within millions of manufacturing documents into a structured labeled-property graph data structure that allow for programmatic query and retrieval. 
    more » « less
  5. ; ; ; ( , Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP))
    null (Ed.)
    We propose AutoQA, a methodology and toolkit to generate semantic parsers that answer questions on databases, with no manual effort. Given a database schema and its data, AutoQA automatically generates a large set of high-quality questions for training that covers different database operations. It uses automatic paraphrasing combined with template-based parsing to find alternative expressions of an attribute in different parts of speech. It also uses a novel filtered auto-paraphraser to generate correct paraphrases of entire sentences. We apply AutoQA to the Schema2QA dataset and obtain an average logical form accuracy of 62.9% when tested on natural questions, which is only 6.4% lower than a model trained with expert natural language annotations and paraphrase data collected from crowdworkers. To demonstrate the generality of AutoQA, we also apply it to the Overnight dataset. AutoQA achieves 69.8% answer accuracy, 16.4% higher than the state-of-the-art zero-shot models and only 5.2% lower than the same model trained with human data. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email