人工智能培训

搜索

论文代码开源:利用语境检测社交媒体上的帮派升级(Detecting Gang-Involved Escalation on Social Media Using

[复制链接]
admin 发表于 2018-9-15 10:20:17 | 显示全部楼层 |阅读模式
admin 2018-9-15 10:20:17 1089 0 显示全部楼层
人工智能论文代码开源:利用语境检测社交媒体上的帮派升级(Detecting Gang-Involved Escalation on Social Media Using Context)请注意该人工智能论文代码开源在github,大部分是python写的,框架可能是tensorflow或者pytorch。作为一种精致而简洁的文学形式,诗歌是人类文化的瑰宝。自动诗歌生成是计算机创造力的重要一步。近年来,为此任务设计了几种神经模型。然而,在整首诗中,意义和主题的连贯性仍然是一个巨大的挑战。在本文的启发下,理论概念为无意识心理学,我们提出了一种新的工作记忆模型用于诗歌创作。与以前的方法不同,我们的模型在神经记​​忆中明确地保持了历史和信息有限的历史。在生成过程中,我们的模型从内存槽中读取最相关的部分以生成当前行。生成每一行后,它会将前一行的最多部分写入内存插槽。通过对记忆的动态操纵,我们的模型保持了连贯的信息流,并学会灵活自然地表达每个主题。我们试验了三种不同的中国诗歌:quatrain,iambic和chinoiserie lyric。 Bothautomatic和人类评估结果表明,我们的模型优于当前最先进的方法。
As an exquisite and concise literary form, poetry is a gem of human culture.Automatic poetry generation is an essential step towards computer creativity.In recent years, several neural models have been designed for this task.However, among lines of a whole poem,the coherence in meaning and topics stillremains a big challenge.In this paper, inspired by the theoretical concept incognitive psychology, we propose a novel Working Memory model for poetrygeneration.Different from previous methods, our model explicitly maintainstopics and informative limited history in a neural memory.During thegeneration process, our model reads the most relevant parts from memory slotsto generate the current line.After each line is generated, it writes the mostsalient parts of the previous line into memory slots.By dynamic manipulationof the memory, our model keeps a coherent information flow and learns toexpress each topic flexibly and naturally.We experiment on three differentgenres of Chinese poetry: quatrain, iambic and chinoiserie lyric.Bothautomatic and human evaluation results show that our model outperforms currentstate-of-the-art methods.论文代码开源:利用语境检测社交媒体上的帮派升级(Detecting Gang-Involved Escalation on Social Media Using Context) pXO7iTQLG9GYZ4Qt.jpg
URL地址:https://arxiv.org/abs/1809.04306v1     ----pdf下载地址:https://arxiv.org/pdf/1809.04306v1    ----         ----github下载地址:https://github.com/xiaoyuanYi/WMPoetry    ----    论文代码开源:利用语境检测社交媒体上的帮派升级(Detecting Gang-Involved Escalation on Social Media Using Context)请注意该人工智能论文代码开源在github,大部分是python写的,框架可能是tensorflow或者pytorch,keras,至于具体是哪一个没有完全测试。
回复

使用道具 举报

您需要登录后才可以回帖 登录 | 立即注册

本版积分规则 返回列表 发新帖

admin当前离线
管理员

查看:1089 | 回复:0

快速回复 返回顶部 返回列表