Data collection and language understanding of food descriptions

Data collection and language understanding of food descriptions
复制标题

食物描述的数据收集和语言理解

DOI:
10.1109/slt.2014.7078635
复制
发表时间:
2014
期刊:
2014 IEEE Spoken Language Technology Workshop (SLT)
影响因子:
--
通讯作者:
James R. Glass
James R. Glass
中科院分区:
--
文献类型:
--
作者:
M. Korpusik;Nicole Schmidt;Jennifer Drexler;D. S. Cyphers;James R. Glass

文献摘要

被引文献

相似文献

本文介绍了初始的数据收集和语言理解实验的一部分,更大的努力,以创建一个营养对话系统,自动提取食物的概念,从用户的口头膳食描述。我们首先总结了通过Amazon Mechanical Turk进行的食品描述的数据收集和注释,然后使用半马尔可夫条件随机场(CRF)进行语义标记实验,获得了85.1的F1测试分数。最后,我们报告了食物分割实验,探索了三种方法将食物与其相应的属性相关联:生成马尔可夫模型,基于转换的学习和CRF分类器。CRF表现最好,F1测试得分为87.1。
This paper presents initial data collection and language understanding experiments conducted as part of a larger effort to create a nutrition dialogue system that automatically extracts food concepts from a user's spoken meal description. We first summarize the data collection and annotation of food descriptions performed via Amazon Mechanical Turk. We then present semantic labeling experiments using a semi-Markov conditional random field (CRF) that obtains an F1 test score of 85.1. Finally, we report food segmentation experiments that explored three methods for associating foods with their corresponding attributes: a generative Markov model, transformation-based learning, and a CRF classifier. The CRF performed best, achieving an F1 test score of 87.1.