Proceedings of the Eleventh ACM International Conference on Web Search and Data Mining 2018
DOI: 10.1145/3159652.3159685
|View full text |Cite
|
Sign up to set email alerts
|

Modelling Domain Relationships for Transfer Learning on Retrieval-based Question Answering Systems in E-commerce

Abstract: Nowadays, it is a heated topic for many industries to build automatic question-answering (QA) systems. A key solution to these QA systems is to retrieve from a QA knowledge base the most similar question of a given question, which can be reformulated as a paraphrase identification (PI) or a natural language inference (NLI) problem. However, most existing models for PI and NLI have at least two problems: They rely on a large amount of labeled data, which is not always available in real scenarios, and they may n… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
78
1

Year Published

2019
2019
2023
2023

Publication Types

Select...
5
4

Relationship

2
7

Authors

Journals

citations
Cited by 86 publications
(88 citation statements)
references
References 34 publications
0
78
1
Order By: Relevance
“…In the following subsections, we will introduce our proposed Query-bag Matching (QBM) model which output is the matching probability indicating whether the query and bag are asking the same questions. The basic Q-Q (query-question) matching model hybrid CNN (hCNN) [8] is presented as the background. Then we will show the base model and its two components designed to promote the performance: Mutual Coverage and Bag Representation.…”
Section: Methodsmentioning
confidence: 99%
See 1 more Smart Citation
“…In the following subsections, we will introduce our proposed Query-bag Matching (QBM) model which output is the matching probability indicating whether the query and bag are asking the same questions. The basic Q-Q (query-question) matching model hybrid CNN (hCNN) [8] is presented as the background. Then we will show the base model and its two components designed to promote the performance: Mutual Coverage and Bag Representation.…”
Section: Methodsmentioning
confidence: 99%
“…, r n } to predict the query-bag matching score. Due to the page limitation, please refer to Yu et al [8] for more details on hCNN.…”
Section: Background: Hcnn For Q-q Matchingmentioning
confidence: 99%
“…Neural Ranking Models. A number of neural ranking models have been proposed for information retrieval, question answering and conversation response ranking [9,13,14,24,26,39,40,44,47]. These models could be classified into three categories [9,10].…”
Section: Related Workmentioning
confidence: 99%
“…For instance, (Liu et al, 2017) successfully uses this technique to improve generalization in a document classification task. It has also been used recently for varied tasks such as transfer learning on Q&A systems (Yu et al, 2018) or duplicate question detection (Shah et al, 2018) and removal of protected attributes from social media textual data (Elazar and Goldberg, 2018).…”
Section: Domain-adversarial Trainingmentioning
confidence: 99%