游客情感计算的文本大数据挖掘方法比较研究

旅游文本大数据以其方便、快捷和低门槛的特点为游客情感计算提供了极大便利，已经成为旅游大数据的主要来源之一。基于大数据理论和情感理论，以文本大数据为数据源，在全面梳理国内外情感计算相关成果的基础上，利用人工智能中的逻辑/算法编程方法、机器学习方法、深度学习方法对旅游文本大数据进行挖掘，探索最佳的基于文本大数据的游客情感计算方法。研究发现：（1）基于情感词典的游客情感计算模型，其核心是构建情感词典和设计情感计算规则，方法简单，容易实现，适用语料范围广。（2）机器学习，用统计学方法抽取文本中的特征项，具有非线性特征，可靠性较线性特征的情感词典方法高。（3）基于深度学习技术的游客情感计算，效果良好，准确率在85%以上。训练多领域的文本语料易于移植，实用性强，且泛化能力好，较适合大数据时代游客情感计算研究。

A comparative study of big text data mining methods on tourist emotion computing

The big tourism text data has greatly facilitated the emotional calculation of tourists with its convenience, rapidity and low threshold, and has become a main sources of tourism big data. Under the guidance of big data theory and emotional theory, we adopt the logic / algorithm programming method, machine learning method to mine the big tourism text data, and explore the best tourist emotional computing method. The research findings are as follows: (1) The core of the emotional calculation model for tourists based on sentiment dictionary is to construct emotional dictionary and design emotional calculation rules. The idea is easy to implement with a wide range of applicable corpus.(2)The machine learning method uses statistical methods to extract feature items in the text, and its non-linear feature improves the reliability of emotion calculation comparing with the linear feature of sentiment dictionary method. (3)The emotional calculation of tourists based on the deep learning has a good effect, and the accuracy rate has reached to 85% or higher. The training of multi-domain text corpus is easy to transplant, practical and general, and suitable for the emotional computing research of tourists in the era of big data.