A Seq2seq-Based Approach to Question Answering over Knowledge Bases

Wu, Linjuan; Wu, Peiyun; Zhang, Xiaowang

doi:10.1007/978-981-15-3412-6_17

Cited by 8 publications

(5 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…For each NL query, we perform document retrieval in 2 steps: first, get a list of relevant documents and then choose a top-k scored list of passages from them. (1) Entities of the question are extracted using BLINK (Wu et al, 2019) and Wikipedia pages of all the entities are collected allowing minimal lexical variations 3 . We also use NL text query generated from λ-expression to search on MediaWiki API 4 .…”

Section: Extraction Pipelinementioning

confidence: 99%

Best of Both Worlds: Towards Improving Temporal Knowledge Base Question Answering via Targeted Fact Extraction

Kannen,

Sharma,

Neelam

et al. 2023

Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing

View full text Add to dashboard Cite

Temporal question answering (QA) is a special category of complex question answering task that requires reasoning over facts asserting time intervals of events. Previous works have predominately relied on Knowledge Base Question Answering (KBQA) for temporal QA. One of the major challenges faced by these systems is their inability to retrieve all relevant facts due to factors such as incomplete KB and entity/relation linking errors (Patidar et al., 2022). A failure to fetch even a single fact will block KBQA from computing the answer. Such cases of KB incompleteness are even more profound in the temporal context. To address this issue, we explore an interesting direction where a targeted temporal fact extraction technique is used to assist KBQA whenever it fails to retrieve temporal facts from the KB. We model the extraction problem as an open-domain question answering task using offthe-shelf language models. This way, we target to extract from textual resources those facts that failed to get retrieved from the KB. Experimental results on two temporal QA benchmarks show promising ~30% & ~10% relative improvements in answer accuracies without any additional training cost.Abdalghani Abujabal, Mohamed Yahya, Mirek Riedewald, and Gerhard Weikum. 2017. Automated template generation for question answering over knowledge graphs. In

show abstract

Section: Extraction Pipelinementioning

confidence: 99%

Best of Both Worlds: Towards Improving Temporal Knowledge Base Question Answering via Targeted Fact Extraction

Kannen,

Sharma,

Neelam

et al. 2023

Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing

View full text Add to dashboard Cite

show abstract

“…Apart from classical NER methods, Probabilistic Graphical methods like the Maximum Entropy Markov Model (MEMM) and Conditional Random Field (CRF) are also very popular as NER. Chen et al [112] uses two-stage MEMM as NER, whereas Bach et al [113], Hu et al [114], Wu et al [115], Sui [116] uses the CRF method and various RNN models.Hu et al [114] uses BiLSTM to capture the context while the CRF layer generates probability distribution for tag sequence. Bach et al [113] proposes the three-stage NER model.…”

Section: Initial Data Transformationsmentioning

confidence: 99%

“…Ever since this type of solution was proposed in Unger et al [89], researchers have come up with many template-based QA systems. Most of the approaches use handcrafted rules to extract the query structure of question [120,134] and hand-crafted templates [115,134].…”

mentioning

confidence: 99%

See 1 more Smart Citation

Question Answer System: A State-of-Art Representation of Quantitative and Qualitative Analysis

Zope

Mishra

Shaw

et al. 2022

BDCC

View full text Add to dashboard Cite

Question Answer System (QAS) automatically answers the question asked in natural language. Due to the varying dimensions and approaches that are available, QAS has a very diverse solution space, and a proper bibliometric study is required to paint the entire domain space. This work presents a bibliometric and literature analysis of QAS. Scopus and Web of Science are two well-known research databases used for the study. A systematic analytical study comprising performance analysis and science mapping is performed. Recent research trends, seminal work, and influential authors are identified in performance analysis using statistical tools on research constituents. On the other hand, science mapping is performed using network analysis on a citation and co-citation network graph. Through this analysis, the domain’s conceptual evolution and intellectual structure are shown. We have divided the literature into four important architecture types and have provided the literature analysis of Knowledge Base (KB)-based and GNN-based approaches for QAS.

show abstract

“…A NLQ é transformada em uma representac ¸ão intermediária que pode ser posteriormente representada como uma forma lógica [Trivedi et al 2017]. As abordagens nesta categoria criam uma lista de regras predefinidas como expressão lógica, tempalte questions, aproximac ¸ão de sub-grafos, regras gramaticais ou outra estrutura semântica equivalente que representa a semântica da NLQ [Wu et al 2019a]. Quando a NLQ corresponde a esses padrões, é mais fácil consultá-lo em uma KB.…”

Section: Representac ¸ãO Das Perguntas E Gerac ¸ãO De Candidatosunclassified

Uma revisão breve sobre perguntas complexas em bases de conhecimento para sistemas de perguntas e respostas

Gomes

Mello

Reis

et al. 2021

Anais Do XIII Simpósio Brasileiro De Tecnologia Da Informação E Da Linguagem Humana (STIL 2021)

View full text Add to dashboard Cite

O avanço nos sistemas de Question Answering alcançou resultados importantes e novos problemas relacionados, como Complex Question Answering e Knowledge Base Question Answering, surgiram. No entanto, faltam estudos que analisam o problema e abordagens para Complex Knowledge Base Question Answering (C-KBQA). Este trabalho preenche essa lacuna apresentando uma visão geral do C-KBQA. Uma coleção de 54 artigos foi selecionada e um mapa dos métodos, abordagens, tendências e lacunas sobre C-KBQA foi realizado. É mostrado que as questões de múltiplos saltos e restritivas são os dois tipos de questões abordadas na literatura. Três etapas foram identificadas para criar um sistema C-KBQA e duas abordagens são geralmente usadas.

show abstract

A Seq2seq-Based Approach to Question Answering over Knowledge Bases

Cited by 8 publications

References 14 publications

Best of Both Worlds: Towards Improving Temporal Knowledge Base Question Answering via Targeted Fact Extraction

Best of Both Worlds: Towards Improving Temporal Knowledge Base Question Answering via Targeted Fact Extraction

Question Answer System: A State-of-Art Representation of Quantitative and Qualitative Analysis

Uma revisão breve sobre perguntas complexas em bases de conhecimento para sistemas de perguntas e respostas

Contact Info

Product

Resources

About