Abstract
With the rapid increase of text information on the Internet, intelligent question answering has attracted considerable attention in many different domains. Many studies use intelligent question answering as an answer selection task and use local or global features to encode sentence representation, which has been proven to be very effective. However, due to the different characteristics of the corpora in different domains or languages, these methods usually suffer a great reduction in performance and cannot produce a sentence representation with good qualities. In this paper, we introduce Hybrid Model combining Local Composite features and Global features into a Siamese network (HM-LCGS) to alleviate the issue. The framework consists of a novel convolution neural network like architecture called local composite features convolution neural network to extract sufficient semantic information of the text from different granularity, shortcut connection to combine local composite features into pre-trained embeddings, alignment layer to mine the correlation between question and answer and bidirectional Long Short-Term Memory to encode the final sentence representation. The experimental results on InsuranceQA and cMedQA datasets show that with suitable granularity selection and embedding method, our proposed model can achieve competitive performance compared with other state-of-the-art models.
Highlights
W ITH the rapid development of society and economy, the amount of information, especially text data, is growing exponentially on the Internet
We propose a Hybrid Model combining Local Composite features and Global features into a Siamese bidirectional Long Short-Term Memory network (HM-LCGS) for Answer Selection
To compensate for this distortion or loss, we propose a novel Convolutional Neural Network (CNN) like architecture called Local Composite Features extraction Convolutional Neural Network (LCF-CNN) to extract multi-type local composite features of the sentence derived from a large number of different scale feature maps, inspired by the works from [12] and [13]
Summary
W ITH the rapid development of society and economy, the amount of information, especially text data, is growing exponentially on the Internet. Y.Lan et al.: Answer Selection Based on Aligned Local Composite Features and Global Features years, deep learning technologies have achieved significant results in a variety of tasks of Natural Language Processing, such as machine translation [7], semantic analysis [8], reading comprehension [9], question answering [10], and text summarization [11]. Siamese biLSTM network is effective in mapping the semantic features of the entire sentence, in Answer Selection, the length of the answer is generally much larger than the length of the question, which leads to a certain loss or distortion of the global features of the answer sentence To compensate for this distortion or loss, we propose a novel Convolutional Neural Network (CNN) like architecture called Local Composite Features extraction Convolutional Neural Network (LCF-CNN) to extract multi-type local composite features of the sentence derived from a large number of different scale feature maps, inspired by the works from [12] and [13].
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
