FAQ-based Question Answering via Word Alignment

By Zhiguo Wang, Abraham Ittycheriah. 2015

44 citations recorded by the source corpus.

In this paper, we propose a novel word-alignment-based method to solve the FAQ-based question answering task. First, we employ a neural network model to calculate question similarity, where the word alignment between two questions is used for extracting features. Second, we design a bootstrap-based feature extraction method to extract a small set of effective lexical features. Third, we propose a learning-to-rank algorithm to train parameters more suitable for the ranking tasks. Experimental results, conducted on three languages (English, Spanish and Japanese), demonstrate that the question similarity model is more effective than baseline systems, the sparse features bring 5% improvements on top-1 accuracy, and the learning-to-rank algorithm works significantly better than the traditional method. We further evaluate our method on the answer sentence selection task. Our method outperforms all the previous systems on the standard TREC data set.

Subfields: Artificial Intelligence, Information Systems

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Zhiguo Wang
2. Steven Bethard
3. Wael Hamza
4. Mihai Surdeanu
5. Mohit Bansal
6. Radu Florian
7. Vikas Yadav
8. Kang Liu

In Information Systems

1. Steven Bethard
2. Mihai Surdeanu
3. Vikas Yadav
4. Phil Blunsom
5. Jun Zhao
6. Kang Liu
7. Haitao Mi
8. Lei Yu

Citations in the corpus

• Neural Variational Inference for Text Processing (2015, methodology): “We alternately optimise the generative model and the inference network by fixing the parameters of one while updating the parameters of the other.”
• Bilateral Multi-Perspective Matching for Natural Language Sentences (2017, background): “Early approaches focused on designing hand-craft features to capture n-gram overlapping, word reordering and syntactic alignments phenomena [Heilman and Smith, 2010; Wang and Ittycheriah, 2015].”
• Convolutional Neural Networks vs. Convolution Kernels: Feature Engineering for Answer Sentence Reranking (2016, unknown): “…al. (2007) 68.52 60.29 Heilman and Smith (2010) 69.17 60.91 Wang and Manning (2010) 69.51 59.51 Yao et al. (2013) 74.77 63.07 Severyn and Moschitti (2013) 73.58 67.81 Yih et al. (2013a) 77.00 70.92 Yu et al. (2014) 78.64 71.13 Wang and Ittycheriah (2015) 77.40 70.63 Tymoshenko and Moschitti (2015)”
• Modeling Relational Information in Question-Answer Pairs with Convolutional Neural Networks (2016, unknown): “…(2010) .5951 .6951 Yao et al. (2013) .6307 .7477 Severyn & Moschitti (2013) .6781 .7358 Yih et al. (2013) .7092 .7700 Yu et al. (2014) .7113 .7846 Wang and Ittycheriah (2015) .7063 .7740 Yin et al. (2015) .6951 .7633 Miao et al. (2015) .7339 .8117 CNN R on (TRAIN) .6857 .7660 CNN R on…”
• Sentence Similarity Learning by Lexical Decomposition and Composition (2016, methodology): “For question answering and information retrieval tasks, sentence similarities between query-answer pairs are used for assessing the relevance and ranking all the candidate answers (Severyn and Moschitti, 2015; Wang and Ittycheriah, 2015).”
• Inner Attention based Recurrent Neural Networks for Answer Selection (2016, methodology): “Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics, pages 1288–1297, Berlin, Germany, August 7-12, 2016. c©2016 Association for Computational Linguistics”
• Alignment over Heterogeneous Embeddings for Question Answering (2019, methodology): “For example, Wang and Ittycheriah (2015) utilized the alignment of words between two questions as a feature in a feedforward neural network that matches similar FAQ questions.”