汉语分词算法，包含最大匹配和基于概率的分词算法-Chinese word segmentation algorithm, and contains the largest matching probability - based segmentation algorithm 下载
|
最大概率分词法,这种分词算法能够较好的解决汉语分词中的歧义问题,但分词效率比最大匹配分词算法要低-greatest probability points accidence, Segmentation algorithm can be used to solve the Chinese word segmentation of Ambiguity, but Word efficient than the largest matching segmentation algorithm lower 下载
|
利用最大匹配法进行汉语句子的分词 最大匹配算法是最常用的分词算法，简单实用正确率可达到80%以上-the maximum matching method for the Chinese Sentence Word maximum matching algorithm is the most commonly used word segmentation algorithm, simple and practical accuracy rate can reach more than 80% 下载
|
计算所汉语词法分析系统ICTCLAS介绍 词是最小的能够独立活动的有意义的语言成分。 但汉语是以字为基本的书写单位，词语之间没有明显的区分标记，因此，中文词语分析是中文信息处理的基础与关键。为此,我们中国科学院计算技术研究所在多年研究基础上，耗时一年研制出了汉语词法分析系统ICTCLAS(Institute of Computing Technology, Chinese Lexical Analysis System)，该系统的功能有：中文分词；词性标注；未登录词识别。分词正确率高达97%以上，未登录词识别召回率均高于90%，其中中国人名的识别召回率接近98%处理速度为31.5Kbytes/s。ICTCLAS的特色还在于：可以根据需要输出多个高概率结果，有多种输出格式，支持北大词性标注集，973专家组给出的词性标注集合。该系统得到了专家的好评，并有多篇论文在国内外发表。 计算所汉语词法分析系统ICTCLAS同时还提供一套完整的动态连接库ICTCLAS.dll和相应的概率词典,开发者可以完全忽略汉语词法分析，直接在自己的系统中调用ICTCLAS，ICTCLAS可以根据需要输出多个高概率的结果，输出格式也可以定制，开发者在分词和词性标注的基础上继续上层开发。-calculation Chinese lexical analysis system ICTCLAS introduced the term is the smallest independent of meaningful activities language components. It is Chinese characters written for the basic unit, the word no clear distinction between markers, therefore, the Chinese term analysis of the Chinese information processing infrastructure and key. To this end, we CAS Institute of Computing Technology based on years of research, 976,000 developed the Chinese lexical analysis system ICTCLAS (Institute of Compu Hosiery Technology, Chinese Lexical Analysis System), the system functions : the Chinese word; tagging; Unknown word recognition. Word accuracy rate of as high as 97%, unknown word recognition recall rate is higher than 90%. these names identify the recall rate of nearly 98% for the process 下载
|
计算所汉语词法分析系统ICTCLAS.分词正确率高达97.58%(973专家组评测)，未登录词识别召回率均高于90%，其中中国人名的识别召回率接近98%处理速度为31.5Kbytes/s。ICTCLAS的特色还在于：可以根据需要输出多个高概率结果，有多种输出格式，支持北大词性标注集，973专家组给出的词性标注集合。 下载
|
I3S文档关键词提取系统 1.0 关键词抽取（Keywords Extraction）指的是如何从一篇文档（或多篇相关文档）中自动抽取出能很好地代表文档主题的若干个词或短语。关键词抽取技术广泛应用于信息检索、文本分类/聚类、信息过滤、文档摘要等各种智能文本信息处理领域，具有很好的应用价值。 I3S KeyExtrator采用适合于汉语特点的关键词抽取技术。该系统基于中科院计算所多年积累的背景知识库，将汉语智能分词算法与基于AV邻接的新词/短语识别方法有机结合，并利用信息论方法对抽取出的词和短语进行重要性评估，同时借助模糊匹配技术对中间抽取结果进行冗余过滤，使抽取出的关键词具有很好的主题代表性和区分度。 -directional document extraction system from 1.0 Keywords (Keywords Extract Ion) refers to a document from the one (or more articles related files) automatically extracted will be good representatives of the document theme several words or phrases. Keywords extraction technology widely used in information retrieval, text classification / clustering, information filtering, Document Summary, and other intelligent text information processing field and has good value. Using directional KeyExtrator suited to the characteristics of Chinese Keyword extraction technology. The system is based on calculations by the Chinese Academy of years of accumulation of background knowledge base, Intelligent Chinese will segmentation algorithm based on the AV adjacent to a new word / phrase recognition organic 下载
|
本文完成了对唇动身份识别技术几个基本问题的理论研究,并对整个系统加以实现.作为本文研究的实验基础,我们建立了唇动方式身份识别数据库(HITLUDB), 该库目前包含30个说话人每人20个汉语词的音视频语料.数据库的扩充与完善工作仍在不断的进行之中.在嘴唇检测方面, 我们对自适应色度过滤模型进行改进,提高了算法的鲁棒性,完成了对嘴唇的精确定位.结合DCT变换与K-L变换的各自特点, 我们提出了特征提取算法,使用较少维数的特征完成了对嘴唇区域主要信息的刻画.由于唇动信息同时包含了生理特征与行为特征, 我们使用静念动念混合建模的方式,完成了对说话人唇动个性特点的精确描述.在HMM训练时,我们提出了特征的归一化处理方法,提高了HMM在实际应用中的性能. 最后,我们分别对身份辨认系统与身份确认系统的基本理论进行了叙述,并完成了系统的实践工作. 关 键 词：身份识别 唇动 特征提取 隐马尔可夫模型 K-L变换 -paper completed the lip movements identification technology several basic issues of theoretical study, system as a whole be achieved. As this paper, the experimental basis, We have established a dynamic manner lip identification database (HITLUDB) The library currently contains 30 words each of 20 Chinese words, sound and video corpus. and the expansion of the database is still perfect keep going on. Detection of the lips, we adaptive color filter model improvements, improve the robustness of the algorithm, completed a pair of lips the precise positioning. DCT combined with the K-L transform their own characteristics, We have proposed a feature extraction algorithm, use less dimension of the lips completed the main message of regional characterization. As the lip movements of information, 下载
|