复旦java培训_Deep Learning

Deep neural network-based NLP toolkit

Deep neural network-based system (FudanDNN-NLP4.2) for Chinese language processing (word segmentation, named identity recognition, POS tagging, semantic analysis, text classification, sentiment analysis, key word and phrase extraction, summarization, question answering, triple extraction and dialog systems) can be downloaded from (Online disk)

Fudan deep neural networks (FudanDNN-NLP4.1) for Chinese language processing (word segmentation, named identity recognition, POS tagging, semantic analysis, text classification, sentiment analysis, summarization and dialog systems) can be downloaded from (Online disk) (Manual)

Fudan deep neural networks (FudanDNN-NLP4.0) for Chinese language processing (word segmentation, named identity recognition, POS tagging, semantic analysis, and dialog systems) can be downloaded from (Online disk).

Fudan deep neural networks (FudanDNN-NLP3.0) for Chinese language processing (word segmentation, named identity recognition, POS tagging, semantic analysis, and dialog systems) can be downloaded from (Online disk).

Fudan deep neural networks (FudanDNN-NLP2.2) for Chinese language processing (word segmentation, named identity recognition, POS tagging, semantic analysis, and dialog) can be downloaded from (Online disk).

Fudan deep neural networks (FudanDNN-NLP2.1) for Chinese language processing (word segmentation, named identity recognition, POS tagging and semantic analysis) can be downloaded from (Online disk).

Fudan deep neural networks (FudanDNN-NLP2.0) for Chinese language processing (word segmentation, named identity recognition, POS tagging and semantic analysis) can be downloaded from (Online disk).

Fudan deep neural networks (FudanDNN-NLP1.0) for Chinese language processing (word segmentation, named identity recognition and POS tagging) can be downloaded from (Online disk).

If you have any problem or comment about the tools, please feel free to contact Xiaoqing Zheng by sending a mail to zhengxq@fudan.edu.cn.

基于深度学习的中文自然语言处理工具

复旦深度网络中文自然语言处理工具FudanDNN-NLP4.2(在FudanDNN-NLP4.1的基础上新增加自定义多种词类及词库，并可指定词类之间的优先级；常见问题检索(FAQ)可返回Top-k个回答；三元组抽取(实体属性和关系)功能；序列注标模型性能优化；修复发现错误)下载：(链接) (介绍)

复旦深度网络中文自然语言处理工具FudanDNN-NLP4.1(在4.0的基础上新增词典动态加载和加密； 依存句法分析； 文本分类、聚类、摘要、情感分析以及关键短语抽取；文本信息抽取和热点新闻发现；篇章级中文分词和词性标注的功能，一定程度上避免在一篇文章中相同实体在不同出现处识别不一致情况；优化了对话管理模块等功能；修复了已知的一些错误)下载：(网盘)

复旦深度网络中文自然语言处理工具FudanDNN-NLP4.0(在3.0的基础上新增上下文相关问答。分为两种情况：第一种情况处理类似上一句问“今天北京天气如何？”，然后追问“上海呢？”的情况；另一种情况是根据对话主题展开、转换和递进给出合适的回答；多轮对话。处理类似订购机票的场景。不同场景可以根据对话进展自由切换，并且期间可插入其他问答；海量自定义问答对的高效检索。检索匹配时考虑同义词替换，可根据发音相似性纠正可能的错误，并且支持一次提问包括多个问题的情况；可为每一位用户定义各自的上下文信息；图形客户端用于系统演示和调试，支持本地或服务器快速部署；问答过程中检测禁用词功能)下载：(网盘)

复旦深度网络中文自然语言处理工具FudanDNN-NLP3.0(在2.2的基础上新增融合任务导向问答和聊天机器人的对话系统；支持访问以RDF表示的知识库(类似知识图谱)；可配置自定义问答的优先处理机制；针对语音识别不准确的发音相似性匹配功能(用于知识库检索和自定义问题的匹配)；简单算术、诗词朗读、术语解读、菜谱查询、家电控制等场景；增加金融、疾病、药物、动物、植物、化学等专业词汇的识别。)下载：(网盘) (介绍)

复旦深度网络中文自然语言处理工具FudanDNN-NLP2.2(在2.1的基础上新增语音对话问答功能，可以用于智能助理、智能客服、问答机器人等应用开发。工具自带包括天气预报、互联网信息检索、出行路线查询等十个应用场景，并可在所提供的框架下方便地增加应用场景)下载：(网盘)

复旦深度网络中文自然语言处理工具FudanDNN-NLP2.1(在2.0的基础上新增双向LSTM的语义分析模型和带动态k-max池化卷积神经网络的句子分类模型、优化多事件语义分析性能、增加自定义词汇类型和时期描述短语识别、修复2.0版已知的错误)下载：(网盘)

复旦深度网络中文自然语言处理工具FudanDNN-NLP2.0(中文分词、自定义词典、命名识别、词性分析、语义分析、文本规范化)下载：(网盘)

复旦深度网络中文自然语言处理工具FudanDNN-NLP1.0(包括：中文分词、命名识别和词性分析)下载：(网盘)

注意：由于进行了整合优化，使用工具包FudanDNN-NLP中的相同功能较仅能够完成单独任务的程序包的准确度高。使用过程中如有任何问题或建议，请邮件zhengxq@fudan.edu.cn(郑骁庆)。

Reference / 参考文献

Xiaoqing Zheng, Hanyang Chen, Tianyu Xu. Deep learning for Chinese word Segmentation and POS tagging. In Proc. Conference on Empirical Methods on Natural Language Processing (EMNLP’13), October 18-21, 2013, pp. 647-657. (pdf)

Xiaoqing Zheng, Haoyuan Peng, Yi Chen, Pengjing Zhang, Wenqiang Zhang. Character-based parsing with convolutional neural network. In Proc. The Twenty-Fourth International Joint Conference on Artificial Intelligence (IJCAI’15), 2015. (pdf)

Xiaoqing Zheng, Jiangtao Feng, Mengxiao Lin, Wenqiang Zhang. Context-specific and multi-prototype character representations. In Proc. The Twenty-Fifth International Joint Conference on Artificial Intelligence (IJCAI’16), 2016. (pdf)

Xiaoqing Zheng, Jiangtao Feng, Yi Chen, Haoyuan Peng, Wenqiang Zhang. Learning context-specific word/character embeddings. In Proc. AAAI Conference on Artificial Intelligence (AAAI’17), 2017. (pdf)

Xiaoqing Zheng. Incremental graph-based neural dependency parsing. In Proc. Conference on Empirical Methods on Natural Language Processing (EMNLP’17), 2017. (pdf)

Haoyuan Peng, Lu Liu, Yi Zhou, Junying Zhou, Xiaoqing Zheng*. Attention-based belief or disbelief feature extraction for dependency parsing. In Proc. AAAI Conference on Artificial Intelligence (AAAI’18), 2018. (Corresponding author) (to appear)

Jiangtao Feng, Xiaoqing Zheng*. Geometric relationship between word and context representations. In Proc. AAAI Conference on Artificial Intelligence (AAAI’18), 2018. (Corresponding author) (to appear)

Yi Zhou, Junying Zhou, Lu Liu, Jiangtao Feng, Xiaoqing Zheng*. RNN-based sequence-preserved attention for dependency parsing. In Proc. AAAI Conference on Artificial Intelligence (AAAI’18), 2018. (Corresponding author) (to appear)