Learning Continuous Word Embedding with Metadata for Question Retrieval in Community Question Answering

Abstract

Community question answering (cQA) has become an important issue due to the popularity of cQA archives on the web. This paper is concerned with the problem of question retrieval. Question retrieval in cQA archives aims to find the existing questions that are semantically equivalent or relevant to the queried questions. However, the lexical gap problem brings about new challenge for question retrieval in cQA. In this paper, we propose to learn continuous word embeddings with metadata of category information within cQA pages for question retrieval. To deal with the variable size of word embedding vectors, we employ the framework of fisher kernel to aggregated them into the fixedlength vectors. Experimental results on large-scale real world cQA data set show that our approach can significantly outperform state-of-the-art translation models and topic-based models for question retrieval in cQA.

Neural information retrieval: at the end of the early years

“…Explicit Learning of word embeddings coupled with category metadata for CQA is proposed by Zhou et al (2015). They adopt word2vec's Skip-gram model augmented with category metadata from online questions, with category information encoding the attributes of words in the question (see Zhang et al 2016a for another example of integrating categorical data with word embeddings).…”

“…Other than the given word vectors, no further deep learning is used. Like Clinchant and Perronnin (2013) and Zhou et al (2015), they adopt the Fisher Kernel framework to convert variable-size concatenations of word embeddings to fixed length. However, this resulting fixed-length Fisher vector is very high-dimensional and dense, so they test various state-of-the-art hashing methods (e.g., Spectral Hashing Weiss et al 2009;Charikar 2002) for reducing the Fisher vector to a lower-dimensional binary vector.…”

Classification Performance of Bio-Marker and Disease Word using Word Representation Models

“…Word representation processing using vectors to indicate lexical similarity between words has recently attracted considerable attention by improving the performance of machine learning models of natural language processing [2][3][4][5][6][7][8][9][10][11].…”

A Review on Neural Network Question Answering Systems

“…They share the common concept of using neural networks to obtain the semantic similarity by learning a similarity metric. The solutions include capturing the semantic similarity be-tween the current and archived questions and incorporating similarity scores to rank and select semantically equivalent questions [12], [17], [22], also incorporating metadata information such as question category to learn the similarity metric [14]. Another proposed solution was to use a neural network to first learn semantic representation of question and answer pairs from a collection of question and answer pairs.…”