# Multi-Perspective Relevance Matching with Hierarchical ConvNets for Social Media Search

Source: [Multi-Perspective Relevance Matching with Hierarchical ConvNets for Social Media Search](<https://www.sota2.com/research/paper/68fa65f71e209131cea0d879>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-25T21:39:29.972Z

Authors: Jinfeng Rao, Wei Yang, Yuhao Zhang, Ferhan Ture, Jimmy Lin

Published: 2018-05-21T16:25:15.000Z

## Abstract

Despite substantial interest in applications of neural networks to information retrieval, neural ranking models have only been applied to standard ad hoc retrieval tasks over web pages and newswire documents. This paper proposes MP-HCNN (Multi-Perspective Hierarchical Convolutional Neural Network) a novel neural ranking model specifically designed for ranking short social media posts. We identify document length, informal language, and heterogeneous relevance signals as features that distinguish documents in our domain, and present a model specifically designed with these characteristics in mind. Our model uses hierarchical convolutional layers to learn latent semantic soft-match relevance signals at the character, word, and phrase levels. A pooling-based similarity measurement layer integrates evidence from multiple types of matches between the query, the social media post, as well as URLs contained in the post. Extensive experiments using Twitter data from the TREC Microblog Tracks 2011--2014 show that our model significantly outperforms prior feature-based as well and existing neural ranking models. To our best knowledge, this paper presents the first substantial work tackling search over social media posts using neural ranking models.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Ad hoc document retrieval on TREC Microblog 2011 (test)](<https://www.sota2.com/research/sota/ad-hoc-document-retrieval-on-trec-microblog-2011-test.md>): Ad hoc document retrieval; TREC Microblog 2011 (test); AP: 40.43.

- [Ad hoc document retrieval on TREC Microblog 2012 (test)](<https://www.sota2.com/research/sota/ad-hoc-document-retrieval-on-trec-microblog-2012-test.md>): Ad hoc document retrieval; TREC Microblog 2012 (test); AP: 24.6.

- [Ad hoc document retrieval on TREC Microblog 2013 (test)](<https://www.sota2.com/research/sota/ad-hoc-document-retrieval-on-trec-microblog-2013-test.md>): Ad hoc document retrieval; TREC Microblog 2013 (test); AP: 28.96.

- [Ad hoc document retrieval on TREC Microblog 2014 (test)](<https://www.sota2.com/research/sota/ad-hoc-document-retrieval-on-trec-microblog-2014-test.md>): Ad hoc document retrieval; TREC Microblog 2014 (test); AP: 0.442.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/1805.08159>)

[PDF](<https://arxiv.org/pdf/1805.08159.pdf>)
