SP-10K: A Large-scale Evaluation Set for Selectional Preference Acquisition

Zhang, Hongming; Ding, Hantian; Song, Yangqiu

doi:10.18653/v1/p19-1071

Cited by 29 publications

(45 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We create two training sets based on separate corpora: first, we parse English Wikipedia using the StanfordNLP neural pipeline (Qi et al, 2018) and extract attested s-v-o triples. Wikipedia has led to relatively good results for selectional preference (Zhang et al, 2019), and in total we extract 6 million unique triples with a cumulative 10 million occurrences. Second, we use the NELL (Carlson et al, 2010) dataset of 604 million s-v-o triples extracted from the dependency parsed ClueWeb09 dataset.…”

Section: Learning From Textmentioning

confidence: 99%

Can a Gorilla Ride a Camel? Learning Semantic Plausibility from Text

Porada¹,

Suleman

Cheung

2019

Proceedings of the First Workshop on Commonsense Inference in Natural Language Processing

View full text Add to dashboard Cite

Modeling semantic plausibility requires commonsense knowledge about the world and has been used as a testbed for exploring various knowledge representations. Previous work has focused specifically on modeling physical plausibility and shown that distributional methods fail when tested in a supervised setting. At the same time, distributional models, namely large pretrained language models, have led to improved results for many natural language understanding tasks. In this work, we show that these pretrained language models are in fact effective at modeling physical plausibility in the supervised setting. We therefore present the more difficult problem of learning to model physical plausibility directly from text. We create a training set by extracting attested events from a large corpus, and we provide a baseline for training on these attested events in a self-supervised manner and testing on a physical plausibility task. We believe results could be further improved by injecting explicit commonsense knowledge into a distributional model.

show abstract

Section: Learning From Textmentioning

confidence: 99%

Can a Gorilla Ride a Camel? Learning Semantic Plausibility from Text

Porada¹,

Suleman

Cheung

2019

Proceedings of the First Workshop on Commonsense Inference in Natural Language Processing

View full text Add to dashboard Cite

show abstract

“…Besides that, we also compare with the selectional preference (SP) based method [46]. Following the original setting, two resources (human annotation and Posterior Probability) of SP knowledge are considered and we denote them as SP (human) and SP (PP) respectively 15 .…”

Section: Baseline Methodsmentioning

confidence: 99%

“…Given one sentence 'The dog is chasing the cat, it barks loudly', we can correctly resolve 'it' to 'dog' rather than 'cat' because 'dog barks' appears 12,247 times in ASER, while 'cat barks' never appears. This is usually called selectional preference [43], which has recently been evaluated in a larger scale in [46]. ASER naturally reflects human's selectional preference for many kinds of syntactic patterns.…”

Section: # Eventuality # Relation # R Typesmentioning

confidence: 99%

ASER: A Large-scale Eventuality Knowledge Graph

Zhang

Liu

Pan

et al. 2020

Proceedings of the Web Conference 2020

Self Cite

View full text Add to dashboard Cite

Understanding human's language requires complex world knowledge. However, existing large-scale knowledge graphs mainly focus on knowledge about entities while ignoring knowledge about activities, states, or events, which are used to describe how entities or things act in the real world. To fill this gap, we develop ASER (activities, states, events, and their relations), a large-scale eventuality knowledge graph extracted from more than 11-billion-token unstructured textual data. ASER contains 15 relation types belonging to five categories, 194-million unique eventualities, and 64-million unique edges among them. Both intrinsic and extrinsic evaluations demonstrate the quality and effectiveness of ASER.

show abstract

“…• Distributional Similarity (DS) (Erk et al, 2010), a method that uses the similarity of the SP Evaluation Set #W #P Keller (Keller and Lapata, 2003) 571 360 SP-10K (Zhang et al, 2019a) 2,500 6,000 embedding of the target argument and average embedding of observed golden arguments in the corpus to predict the preference strength.…”

Section: Baselinesmentioning

confidence: 99%

“…Thus, in this section, we evaluate different SP acquisition methods with ground truth (human labeled datasets). Two representative datasets, Keller (Keller and Lapata, 2003) and SP-10K (Zhang et al, 2019a) word pairs are provided. For each of the word pairs, the datasets also provide the annotated plausibility score of how likely a preference exists between that word pair under the corresponding SP relation.…”

Section: Selectional Preference Acquisitionmentioning

confidence: 99%

Multiplex Word Embeddings for Selectional Preference Acquisition

Zhang

Bai

Yan

et al. 2019

Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conferen

Self Cite

View full text Add to dashboard Cite

Conventional word embeddings represent words with fixed vectors, which are usually trained based on co-occurrence patterns among words. In doing so, however, the power of such representations is limited, where the same word might be functionalized separately under different syntactic relations. To address this limitation, one solution is to incorporate relational dependencies of different words into their embeddings. Therefore, in this paper, we propose a multiplex word embedding model, which can be easily extended according to various relations among words. As a result, each word has a center embedding to represent its overall semantics, and several relational embeddings to represent its relational dependencies. Compared to existing models, our model can effectively distinguish words with respect to different relations without introducing unnecessary sparseness. Moreover, to accommodate various relations, we use a small dimension for relational embeddings and our model is able to keep their effectiveness. Experiments on selectional preference acquisition and word similarity demonstrate the effectiveness of the proposed model, and a further study of scalability also proves that our embeddings only need 1/20 of the original embedding size to achieve better performance.

show abstract

SP-10K: A Large-scale Evaluation Set for Selectional Preference Acquisition

Cited by 29 publications

References 19 publications

Can a Gorilla Ride a Camel? Learning Semantic Plausibility from Text

Can a Gorilla Ride a Camel? Learning Semantic Plausibility from Text

ASER: A Large-scale Eventuality Knowledge Graph

Multiplex Word Embeddings for Selectional Preference Acquisition

Contact Info

Product

Resources

About