基于深度学习的跨语言情感分析模型需要借助预训练的双语词嵌入(Bilingual Word Embedding, BWE)词典获得源语言和目标语言的文本向量表示。为了解决BWE词典较难获得的问题,该文提出一种基于词向量情感特征表示的跨语言文本情感分析方法...
详细信息
基于深度学习的跨语言情感分析模型需要借助预训练的双语词嵌入(Bilingual Word Embedding, BWE)词典获得源语言和目标语言的文本向量表示。为了解决BWE词典较难获得的问题,该文提出一种基于词向量情感特征表示的跨语言文本情感分析方法,引入源语言的情感监督信息以获得源语言情感感知的词向量表示,使得词向量的表示兼顾语义信息和情感特征信息,用于跨语言文本的情感预测。实验以英语为源语言,分别以汉语、法语、德语、日语、韩语和泰语6种语言为目标语言进行跨语言情感分析。实验结果表明,该文所提模型与机器翻译方法、不采用情感特征表示的跨语言情感分析方法比较,能够分别提高约9.3%和8.7%预测准确率。该模型在德语上的跨语言情感分析效果最好,英语与德语同属日耳曼语族,在语法和语义上更为接近,符合实验预期。实验部分对影响跨语言情感分析模型的相关因素进行了分析。
To reduce network access latency, network traffic volume and server load, caching capacity has been proposed as a component of evolved Node B(e Node B) in the ratio access network(RAN). These e Node B caches reduce tr...
详细信息
To reduce network access latency, network traffic volume and server load, caching capacity has been proposed as a component of evolved Node B(e Node B) in the ratio access network(RAN). These e Node B caches reduce transport energy consumption but lead to additional energy cost by equipping every e Node B with caching capacity. Existing researches focus on how to minimize total energy consumption, but often ignore the trade-off between energy efficiency and end user quality of experience, which may lead to undesired network performance degradation. In this paper, for the first time, we build an energy model to formulate the problem of minimizing total energy consumption at e Node B caches by taking a trade-off between energy efficiency and end user quality of experience. Through coordinating all the e Node B caches in the same RAN, the proposed model can take a good balance between caching energy and transport energy consumption while also guarantee end user quality of experience. The experimental results demonstrate the effectiveness of the proposed model. Compared with the existing works, our proposal significantly reduces the energy consumption by approximately 17% while keeps superior end user quality of experience performance.
暂无评论