Abstract
Structured tabular data are widely used in various information systems, especially with the development of big data technology, making it more difficult to query on these complex data. SQL facilitates the query on structured tabular, however, the mastery of SQL has a certain threshold for most non-expert users. Therefore, in order to facilitate ordinary users to quickly obtain the required information in complex structured data, we design and implement a Q&A system for structured knowledge. First, we make a detailed distinction between Q&A scenarios for structured data and design different approaches, respectively. Then, we introduce deep learning models in system algorithm layer to enhance the generalization ability. Finally, the TorchServe framework is used to optimize system deployment and improve system performance using batch inference. The experimental results show that the prototype system has certain generalization ability and also has some advantages in performance compared with traditional methods.
Export citation and abstract BibTeX RIS
Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.