Name
Abstract | ||
|---|---|---|
With the development of the Internet, more and more people express their ideas on the internet in the form of short text. And a question with the same intention can be expressed in different ways. So it is necessary and important to understand the meaning of short text when we want to provide intelligent service to customer. Many studies have focused on the short texts based on public data sets. However, little studies have been carried out or can be effectively used in some specific fields. Taking logistics Intelligent Customer Service (ICS) as an example, the short texts has the above characteristics. To solve this issue about intention multiple-representation in logistics, a self-attention-based model, that is, One question to Many question (O2M) is proposed. On the basis of classification task, the model can learn the mapping relation from customer questions to standard questions. And it consists of three parts: standard questions domain, customer questions domain and selector. For the two domains, they learn semantic patterns of their own questions. And the selector becomes the bridge between them. Extensive experiments were carried out on logistics corpus. And the results show that the model is effective and the accuracy of the model is higher than that of traditional neural network models. |
Year | DOI | Venue |
|---|---|---|
2020 | 10.1007/978-3-030-55130-8_16 | KSEM (1) |
DocType | Citations | PageRank |
Conference | 0 | 0.34 |
References | Authors | |
0 | 6 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Jingxiang Hu | 1 | 0 | 0.34 |
| Junjie Peng | 2 | 40 | 14.41 |
| Wenqiang Zhang | 3 | 5 | 6.50 |
| Lizhe Qi | 4 | 0 | 0.68 |
| Miao Hu | 5 | 0 | 0.34 |
| Huanxiang Zhang | 6 | 0 | 0.34 |