Question Similarity Calculation for FAQ Answering

Song, Wanpeng; Min, Feng; Gu, Naijie; Wenyin, Liu

doi:10.1109/skg.2007.247

Cited by 31 publications

(17 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Query-Question Relevance: Query-question relevance estimation is important for community question answering (QA) forums in identifying existing QA pairs which match a new query. Previous work on answering a question given FAQ pages or a database of QA pairs (FAQ-Finder [9], Auto-FAQ [28], [15,24,25]) was based on traditional feature engineering for surfacing statistical/semantic similarities between query and questions. Recently, Transformer based models have been shown to be quite successful for the task, especially in chatbot settings where the query is being matched with a question-answer pair rather than just a question [6].…”

Section: Related Workmentioning

confidence: 99%

Transformer Models for Recommending Related Questions in Web Search

Mitra

Gupta

Dandapat

2020

Proceedings of the 29th ACM International Conference on Information &Amp; Knowledge Management

View full text Add to dashboard Cite

People Also Ask (PAA) is an exciting feature in most of the leading search engines which recommends related questions for a given user query, thereby attempting to reduce the gap between user's information need. This helps users in diving deep into the topic of interest, and reduces task completion time. However, showing unrelated or irrelevant questions is highly detrimental to the user experience. While there has been significant work on query reformulation and related searches, there is hardly any published work on recommending related questions for a query. Question suggestion is challenging because the question needs to be interesting, structurally correct, not be a duplicate of other visible information, and must be reasonably related to the original query. In this paper, we present our system which is based on a Transformer-based neural representation, BERT (Bidirectional Encoder Representations from Transformers), for query, question and corresponding search result snippets. Our best model provides an accuracy of ∼81%. CCS CONCEPTS • Computing methodologies → Neural networks; Machine learning; Natural language processing; • Information systems → Content ranking.

show abstract

Section: Related Workmentioning

confidence: 99%

Transformer Models for Recommending Related Questions in Web Search

Mitra

Gupta

Dandapat

2020

Proceedings of the 29th ACM International Conference on Information &Amp; Knowledge Management

View full text Add to dashboard Cite

show abstract

“…No entanto,é possível encontrar na literatura alguns trabalhos relacionados aos sistemas CQA [Lo et al 2016, Chen et al 2015, Hovy et al 2000, Pôssas et al 2005, Song et al 2007, Grappy et al 2011, Aslay et al 2013, Dalip et al 2013, que possuem características que podem ser utilizadas na busca e ordenação de questionários e, portanto, podem ser analisados a partir dessa perspectiva. Por exemplo, trabalhos que analisam a qualidade da resposta a uma determinada pergunta pode ser visto de forma análogo quando a busca por questionáriosé realizada a partir de apenas uma palavra-chave ou de frases.…”

Section: Trabalhos Relacionadosunclassified

Comparando a eficácia na recuperação de questionários: QSMatching vs Vector model vs Fuzzy

Souza

Dorneles

2019

iSys

View full text Add to dashboard Cite

Elaborar um questionário útil representa uma tarefa importante para a pesquisa descritiva. Perguntas mal elaboradas podem levar a respostas com interpretações sem sentido, sutis ou ingênuas. Portanto, pode ser interessante reutilizar, parcial ou totalmente, questionários já criados com o mesmo propósito. Neste trabalho, comparamos o QSMatching com os modelos vetorial e fuzzy para calcular a similaridade entre questionários e, consequentemente, obter uma ordenação de questionários de acordo com a consulta do usuário. Para verificar a efetividade, foi realizado um experimento comparando as abordagens QSMatching, modelo vetorial e fuzzy. O resultado da análise do experimento mostra que o QSMatching é mais efetivo que outros modelos para recuperação de questionários.

show abstract

“…For example, Frequently Asked Question (FAQ) Finder (Burke et al, 1997) heuristically combined statistical similarities computed using conventional vector space models with semantic similarities between questions estimated using WordNet (Fellbaum, 1998) to rank FAQs. Song et al (2007) presented an approach which is a linear combination of statistic similarity, calculated based on word co-occurrence, and semantic similarity, calculated using WordNet and a bipartite mapping. Auto-FAQ (Whitehead, 1995) applied shallow language understanding into automatic FAQ answering, where the matching of a question to FAQs is based on keyword comparison enhanced by limited language processing techniques.…”

Section: Question Retrievalmentioning

confidence: 99%

Learning Semantic Relatedness in Community Question Answering Using Neural Models

Nassif¹,

Mohtarami

Glass

2016

Proceedings of the 1st Workshop on Representation Learning for NLP

View full text Add to dashboard Cite

Community Question Answering forums, such as Quora and Stackoverflow contain millions of questions and answers. Automatically finding the relevant questions from the existing questions and finding the relevant answers to a new question are Natural Language Processing tasks. In this paper, we aim to address these tasks, which we refer to as similar-Question Retrieval and Answer Selection. We present a neural-based model with stacked bidirectional LSTMs and MLP to address these tasks. The model generates the vector representations of the question-question or question-answer pairs and computes their semantic similarity scores, which are then employed to rank and predict relevancies. Extensive experiments demonstrate our results outperform the baselines.

show abstract

Question Similarity Calculation for FAQ Answering

Cited by 31 publications

References 9 publications

Transformer Models for Recommending Related Questions in Web Search

Transformer Models for Recommending Related Questions in Web Search

Comparando a eficácia na recuperação de questionários: QSMatching vs Vector model vs Fuzzy

Learning Semantic Relatedness in Community Question Answering Using Neural Models

Contact Info

Product

Resources

About