基于卷积神经网络的旅游信息关系抽取研究 |
| |
引用本文: | 鲍玉来,耿雪来,飞龙.基于卷积神经网络的旅游信息关系抽取研究[J].现代情报,2019,39(8):132-136. |
| |
作者姓名: | 鲍玉来 耿雪来 飞龙 |
| |
作者单位: | 1. 福建工程学院图书馆, 福建 福州 350118;2. 内蒙古大学计算机学院, 内蒙古 呼和浩特 010021 |
| |
摘 要: | 目的/意义]在非结构化语料集中抽取知识要素,是实现知识图谱的重要环节,本文探索了应用深度学习中的卷积神经网络(CNN)模型进行旅游领域知识关系抽取方法。方法/过程]抓取专业旅游网站的相关数据建立语料库,对部分语料进行人工标注作为训练集和测试集,通过Python语言编程实现分词、向量化及CNN模型,进行关系抽取实验。结果/结论]实验结果表明,应用卷积神经网络对非结构化的旅游文本进行关系抽取时能够取得满意的效果(Precision 0.77,Recall 0.76,F1-measure 0.76)。抽取结果通过人工校对进行优化后,可以为旅游知识图谱构建、领域本体构建等工作奠定基础。
|
关 键 词: | 卷积神经网络 关系抽取 旅游信息 词向量 |
Research on Tourism Information Relations Extraction Based on Convolutional Neural Network |
| |
Authors: | Bao Yulai Geng Xuelai Fei Long |
| |
Institution: | 1. Library, Fujian University of Technology, Fuzhou 350118, China;2. School of Computer Science, Inner Mongolia University, Hohhot 010021, China |
| |
Abstract: | Purpose/Significance]Extracting knowledge elements in unstructured corpus is an important part of realizing knowledge map.This paper explored the application of convolutional neural network(CNN)model relations extration in tourism field.Method/Process]We built a tourism corpus by grasping the relevant data of professional travel websites.Some corpora were manually labeled as a training dataset and test dataset.Then we implemented word segmentation,vectorization and CNN model through Python language programming,and conducted relationship extraction experiments.Results/Conclusions]The experimental results showed that the application of convolutional neural networks could achieve satisfactory results when extracting relations in unstructured tourism texts(Precision 0.77,Recall 0.76,F1-measure 0.76).After manual optimization,the results could lay the foundation for the construction of tourism knowledge maps and domain ontology construction. |
| |
Keywords: | convolutional neural networks relations extraction tourism information word embedding |
|
| 点击此处可从《现代情报》浏览原始摘要信息 |
| 点击此处可从《现代情报》下载免费的PDF全文 |
|