Weakly hierarchical lasso based learning to rank in best answer prediction

Tian, Qiguo; Li, Baoxin

doi:10.1109/asonam.2016.7752250

Cited by 4 publications

(4 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…That is, for each vertex v ∈ V , there are no edges (v, u) to any other vertices u v ∈ V . Much of the prior work on feature driven answer selection [2,15,30,29] adopts this view.…”

Section: Reflexivementioning

confidence: 99%

“…Feature-Driven Models [2,15,30,29] in CQA identify and incorporate user features, content features, and thread features, e.g., in treebased models [2,15,30] to identify the best answer. Tian et al [30] found that the best answer tend to be early and novel, with more details and comments.…”

Section: Related Workmentioning

confidence: 99%

“…One approach to identify relevant answers is to identify salient features for each question-answer tuple (q, a) and treat it as a supervised classification problem [2,15,30,29]. Deep Text Models further develop this approach [38,35,32,27].…”

Section: Introductionmentioning

confidence: 99%

“…Modular, Induced Relational Framework: We introduce a modular framework that separates the construction of the graph with the label selection mechanism. In contrast, prior work in answer selection (e.g., [2,15,30,29].) looked at individual tuples, and work on GCNs (e.g., [17,39]) work with the given graph (i.e., no induced graphs) and with similarity as a mechanism for label propagation.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

An Induced Multi-Relational Framework for Answer Selection in Community Question Answer Platforms

Narang,

Yang,

Krishnan

et al. 2019

Preprint

View full text Add to dashboard Cite

This paper addresses the question of identifying the best candidate answer to a question on Community Question Answer (CQA) forums. The problem is important because Individuals often visit CQA forums to seek answers to nuanced questions. We develop a novel induced relational graph convolutional network (IR-GCN) framework to address the question. We make three contributions. First, we introduce a modular framework that separates the construction of the graph with the label selection mechanism. We use equivalence relations to induce a graph comprising cliques and identify two label assignment mechanisms-label contrast, label sharing. Then, we show how to encode these assignment mechanisms in GCNs. Second, we show that encoding contrast creates discriminative magnification-enhancing the separation between nodes in the embedding space. Third, we show a surprising result-boosting techniques improve learning over familiar stacking, fusion, or aggregation approaches for neural architectures. We show strong results over the state-of-the-art neural baselines in extensive experiments on 50 StackExchange communities.

show abstract

“…That is, for each vertex v ∈ V , there are no edges (v, u) to any other vertices u v ∈ V . Much of the prior work on feature driven answer selection [2,15,30,29] adopts this view.…”

Section: Reflexivementioning

confidence: 99%