Zero-shot Entity Extraction from Web Pages

Pasupat, Panupong; Liang, Percy

doi:10.3115/v1/p14-1037

Cited by 43 publications

(30 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Ritter et al (2011) looked into recognizing entities from social media data that involves informal and potentially noisy texts. Pasupat and Liang (2014) looked into the issue of zero-shot entity extraction from Web pages with natural language queries where minimal supervision was used. Neelakantan and Collins (2014) looked into the problem of automatically constructing dictionaries with minimal supervision for improved named entity extraction.…”

Section: Related Workmentioning

confidence: 99%

Joint Mention Extraction and Classification with Mention Hypergraphs

Lu¹,

Roth²

2015

Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing

184

274

View full text Add to dashboard Cite

We present a novel model for the task of joint mention extraction and classification. Unlike existing approaches, our model is able to effectively capture overlapping mentions with unbounded lengths. The model is highly scalable, with a time complexity that is linear in the number of words in the input sentence and linear in the number of possible mention classes. Our model can be extended to additionally capture mention heads explicitly in a joint manner under the same time complexity. We demonstrate the effectiveness of our model through extensive experiments on standard datasets.

show abstract

Section: Related Workmentioning

confidence: 99%

Joint Mention Extraction and Classification with Mention Hypergraphs

Lu¹,

Roth²

2015

Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing

184

274

View full text Add to dashboard Cite

show abstract

“…We extract a set of features from the web page for the event page classifier, including term frequency, web entity [27], anchors, URL segments and page title. we remove features that occur on less than 100 different domains to make sure the features are general enough and not specific to our training set.…”

Section: Event Extraction 41 Event Page Classifiermentioning

confidence: 99%

Constructing a Comprehensive Events Database from the Web

Wang

Kanagal

Garg

et al. 2019

Proceedings of the 28th ACM International Conference on Information and Knowledge Management

View full text Add to dashboard Cite

In this paper, we consider the problem of constructing a comprehensive database of events taking place around the world. Events include small hyper-local events like farmer's markets, neighborhood garage sales, as well as larger concerts and festivals. Designing a high-precision and high-recall event extractor from unstructured pages across the whole web is a challenging problem. We cannot resort overly to domain-specific strategies since it needs to work on all web pages, including on new domains; we need to account for variations in page layouts and structure across websites. Further, we need to deal with low-quality pages on the web with limited structure. We have built an ML-powered extraction system to solve this problem, using schema.org annotations as training data. Our extraction system operates in two phases. In the first phase, we generate raw event information from individual web pages. To do this, an event page classifier predicts if a web page contains any event information; this is then followed by a single/multiple classifier that decides if the page contains a single event or multiple events; the first phase concludes by applying event extractors that extract the key fields of a public event (the title, the date/time information, and the location information). In the second phase, we further improve the extraction quality via three novel algorithms, repeated patterns, event consolidation and wrapper induction, which are designed to use the raw event extractions as input and generate events whose quality is significantly higher. We evaluate our extraction models on two large scale publicly available web corpus, Common Crawl and ClueWeb12. Experimental analysis shows that our methodology achieves over 95% extraction precision and recall on both datasets.

show abstract

“…In this paradigm, facts from existing knowledge bases are paired with unlabeled documents to create noisy or “weakly” labeled training examples [1, 25, 26, 28]. In addition to existing knowledge bases, crowdsourcing [12] and heuristics from domain experts [29] have also proven to be effective weak supervision sources. In our work, we show that by incorporating all kinds of supervision in one framework in a noise-aware way, we are able to achieve high quality in knowledge base construction.…”

Section: Related Workmentioning

confidence: 99%

Fonduer

Hsiao

Cheng

et al. 2018

Proceedings of the 2018 International Conference on Management of Data

View full text Add to dashboard Cite

We focus on knowledge base construction (KBC) from richly formatted data. In contrast to KBC from text or tabular data, KBC from richly formatted data aims to extract relations conveyed jointly via textual, structural, tabular, and visual expressions. We introduce , a machine-learning-based KBC system for richly formatted data. presents a new data model that accounts for three challenging characteristics of richly formatted data: (1) prevalent document-level relations, (2) multimodality, and (3) data variety. uses a new deep-learning model to automatically capture the representation (i.e., features) needed to learn how to extract relations from richly formatted data. Finally, provides a new programming model that enables users to convert domain expertise, based on multiple modalities of information, to meaningful signals of supervision for training a KBC system. -based KBC systems are in production for a range of use cases, including at a major online retailer. We compare against state-of-the-art KBC approaches in four different domains. We show that achieves an average improvement of 41 F1 points on the quality of the output knowledge base—and in some cases produces up to 1.87× the number of correct entries—compared to expert-curated public knowledge bases. We also conduct a user study to assess the usability of ’s new programming model. We show that after using for only 30 minutes, non-domain experts are able to design KBC systems that achieve on average 23 F1 points higher quality than traditional machine-learning-based KBC approaches.

show abstract

Zero-shot Entity Extraction from Web Pages

Cited by 43 publications

References 23 publications

Joint Mention Extraction and Classification with Mention Hypergraphs

Joint Mention Extraction and Classification with Mention Hypergraphs

Constructing a Comprehensive Events Database from the Web

Fonduer

Contact Info

Product

Resources

About