Open question answering over curated and extracted knowledge bases

Fader, Anthony; Zettlemoyer, Luke; Etzioni, Oren

doi:10.1145/2623330.2623677

Cited by 367 publications

(290 citation statements)

References 13 publications

Supporting

Mentioning

285

Contrasting

Unclassified

Order By: Relevance

“…In the final phase, the type alignment component performs semantic integration based on domain ontology and DUL ontology by combining linked data discovering (LDD), terminological similarity computation (TSS) and semantic similarity computation (SSC). Aligned DUL classes will be associated with identified type by rdfs:subClassOf 8 . The rationale of each component implementation is discussed in detail below.…”

Section: Methodsmentioning

confidence: 99%

“…FreeBase 1 and DBpedia 2 are famous examples of an effort to produce large scale world knowledge in a structured form. The structured facts are quite useful in tasks like question answering [20,8], facilitating both understanding the question and finding the answer. For example, in order to answer the question "Which personification in Marvel Comics Universe was created by Bill Mantlo and Mike Mignola?…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Exploiting Linked Open Data to Uncover Entity Types

Gao

Mazumdar

2015

Semantic Web Evaluation Challenges

View full text Add to dashboard Cite

Abstract. Extracting structured information from text plays a crucial role in automatic knowledge acquisition and is at the core of any knowledge representation and reasoning system. Traditional methods rely on hand-crafted rules and are restricted by the performance of various linguistic pre-processing tools. More recent approaches rely on supervised learning of relations trained on labelled examples, which can be manually created or sometimes automatically generated (referred as distant supervision). We propose a supervised method for entity typing and alignment. We argue that a rich feature space can improve extraction accuracy and we propose to exploit Linked Open Data (LOD) for feature enrichment. Our approach is tested on task-2 of the Open Knowledge Extraction challenge, including automatic entity typing and alignment. Our approach demonstrate that by combining evidences derived from LOD (e.g. DBpedia) and conventional lexical resources (e.g. WordNet) (i) improves the accuracy of the supervised induction method and (ii) enables easy matching with the Dolce+DnS Ultra Lite ontology classes.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Exploiting Linked Open Data to Uncover Entity Types

Gao

Mazumdar

2015

Semantic Web Evaluation Challenges

View full text Add to dashboard Cite

show abstract

“…However, because predefined relations are required, they are not effective when no relations are extracted in advance. Open IE systems (Mausam, 2016) overcome this issue and use raw textual phrases as relations, which has been applied to several NLP tasks such as question answering (Fader et al, 2014). To the best of our knowledge, this paper is the first to use Open IE to understand large amounts of literature in a specific medical domain in combination with topic modeling methods.…”

Section: Journal Of Data and Information Sciencementioning

confidence: 99%

“…Open IE is an information extraction technique applied in natural language processing (Fader, Zettlemoyer, & Etzioni, 2014;Mausam, 2016) that gathers facts in the form of triples <subject, predicate, object>. For example, given the sentence, "Alzheimer is strongly correlated with the apoe genotype," it extracts <Alzheimer, is strongly correlated with, the apoe genotype>.…”

Section: Introductionmentioning

confidence: 99%

Using Machine Reading to Understand Alzheimer’s and Related Diseases from the Literature

Tsutsui

Ding

2017

Journal of Data and Information Science

View full text Add to dashboard Cite

Purpose: This paper aims to better understand a large number of papers in the medical domain of Alzheimer's disease (AD) and related diseases using the machine reading approach.Design/methodology/approach: The study uses the topic modeling method to obtain an overview of the field, and employs open information extraction to further comprehend the field at a specific fact level. Practical implications:This study uses the literature to answer specific questions on a scientific domain, which can help domain experts find interesting and meaningful relations among entities in a similar manner, such as to discover relations between AD and AIDS/HIV. Originality/value: Both the overview and specific information from the literature are obtained using two distinct methods in a complementary manner. This combination is novel because previous work has only focused on one of them, and thus provides a better way to understand an important scientific field using data-driven methods.

show abstract

“…SQA, which contains 6,066 question sequences with 17,553 total question-answer pairs, is to the best of our knowledge the first semantic parsing dataset for sequential question answering. Section 3 describes our novel dynamic neural semantic parsing framework (DynSP), a weakly su-1 For instance, there are only 3.75% questions with more than 15 words in WikiAnswers (Fader et al, 2014). 2 Studies have shown increased sentence complexity links to longer reading times (Hale, 2006;Levy, 2008;Frank, 2013 pervised structured-output learning approach based on reward-guided search that is designed for solving sequential QA.…”

Section: Introductionmentioning

confidence: 99%

Search-based Neural Structured Learning for Sequential Question Answering

Iyyer¹,

Yih²,

Chang³

2017

Proceedings of the 55th Annual Meeting of the Association For Computational Linguistics (Volume 1: Long Papers)

175

217

View full text Add to dashboard Cite

Recent work in semantic parsing for question answering has focused on long and complicated questions, many of which would seem unnatural if asked in a normal conversation between two humans. In an effort to explore a conversational QA setting, we present a more realistic task: answering sequences of simple but inter-related questions. We collect a dataset of 6,066 question sequences that inquire about semistructured tables from Wikipedia, with 17,553 question-answer pairs in total. To solve this sequential question answering task, we propose a novel dynamic neural semantic parsing framework trained using a weakly supervised reward-guided search. Our model effectively leverages the sequential context to outperform state-of-the-art QA systems that are designed to answer highly complex questions.

show abstract

Open question answering over curated and extracted knowledge bases

Cited by 367 publications

References 13 publications

Exploiting Linked Open Data to Uncover Entity Types

Exploiting Linked Open Data to Uncover Entity Types

Using Machine Reading to Understand Alzheimer’s and Related Diseases from the Literature

Search-based Neural Structured Learning for Sequential Question Answering

Contact Info

Product

Resources

About