基于查询的新闻多文档自动摘要技术研究

针对新闻文本领域,该文提出一种基于查询的自动文本摘要技术,更加有针对性地满足用户信息需求。根据句子的TF-IDF、与查询句的相似度等要素,计算句子权重,并根据句子指示的时间给定不同的时序权重系数,使得最近发生的新闻内容具有更高的权重,最后使用最大边界相关的方法选择摘要句。通过与基于TF-IDF、Text-Rank、LDA等六种方法的对比,该摘要方法ROUGE评测指标上优于其他方法。从结合评测结果及摘要示例可以看出,该文提出的方法可以有效地从新闻文档集中摘取核心信息,满足用户查询内容的信息需求。     

文本摘要常用数据集和方法研究综述

文本摘要成为人们从互联网上海量文本信息中便捷获取知识的重要手段。现有方法都是在特定数据集上进行训练和效果评价,包括一些公用数据集和作者自建数据集。已有综述文献对现有方法进行全面细致的总结,但大多都是对方法进行总结,而缺少对数据集的详细描述。该文从调研数据集的角度出发,对文本摘要常用数据集及在该数据集上的经典和最新方法进行综述。对公用数据集的综述包括数据来源、语言及获取方式等,对自建数据集的总结包括数据规模、获取和标注方式等。对于每一种公用数据集,给出了文本摘要问题的形式化定义。同时,对经典和最新方法在特定数据集上的实验效果进行了分析。最后,总结了已有常用数据集和方法的现状,并指出存在的一些问题。     

基于深度学习的文本自动摘要方案

针对自然语言处理（NLP）生成式自动摘要领域的语义理解不充分、摘要语句不通顺和摘要准确度不够高的问题，提出了一种新的生成式自动摘要解决方案，包括一种改进的词向量生成技术和一个生成式自动摘要模型。改进的词向量生成技术以Skip-Gram方法生成的词向量为基础，结合摘要的特点，引入词性、词频和逆文本频率三个词特征，有效地提高了词语的理解；而提出的Bi-MulRnn+生成式自动摘要模型以序列映射（seq2seq）与自编码器结构为基础，引入注意力机制、门控循环单元（GRU）结构、双向循环神经网络（BiRnn）、多层循环神经网络（MultiRnn）和集束搜索，提高了生成式摘要准确性与语句流畅度。基于大规模中文短文本摘要（LCSTS）数据集的实验结果表明，该方案能够有效地解决短文本生成式摘要问题，并在Rouge标准评价体系中表现良好，提高了摘要准确性与语句流畅度。     

Weighted archetypal analysis of the multi-element graph for query-focused multi-document summarization

Most existing research on applying the matrix factorization approaches to query-focused multi-document summarization (Q-MDS) explores either soft/hard clustering or low rank approximation methods. We employ a different kind of matrix factorization method, namely weighted archetypal analysis (wAA) to Q-MDS. In query-focused summarization, given a graph representation of a set of sentences weighted by similarity to the given query, positively and/or negatively salient sentences are values on the weighted data set boundary. We choose to use wAA to compute these extreme values, archetypes, and hence to estimate the importance of sentences in target documents set. We investigate the impact of using the multi-element graph model for query focused summarization via wAA. We conducted experiments on the data of document understanding conference (DUC) 2005 and 2006. Experimental results evidence the improvement of the proposed approach over other closely related methods and many of state-of-the-art systems.     

28000m~3/h空分装置开车总结

内蒙古鄂尔多斯市乌审旗世林化工有限责任公司30万t/a甲醇项目配置了28 000 m3/h空分装置。介绍了空分装置的工艺流程;指出了装置设计与安装阶段不尽合理之处;采取了相关的完善措施;提出了装置试车的操作要点。结果表明:空分装置一次试车成功,产出了纯度大于99.6%的合成氧气和纯度大于99.999%的合格氮气。     

Improving player experience in Computer Games by using players' behavior analysis and linguistic descriptions

Understanding player behavior has an interest to computer games researchers and developers since it allows them to improve the design and implementation of computer games and also to ensure that players have the expected experiences. Currently this knowledge is not usually reported to players as feedback, although sometimes it is already used as an analysis tool. This paper presents a novel technology for automatically generating linguistic reports and immediate feedback from actions performed by players during play sessions. These reports allow developers to provide players with a more complete and personalized feedback about their behaviors, abilities, attitudes, skills or movements. In order to show and explore the possibilities of this new technology, we have incorporated it in the core of a computer game. We have evaluated positively that the incorporation of this kind of feedback into the core of YADY computer game allows us to improve the overall player experience.     

纳米铝粉表面包覆改性研究的最新进展

综述了固体推进剂用纳米铝粉的几种包覆改性方法,包括:碳包覆、金属包覆、氧化物包覆、聚合物包覆、有机酸包覆及推进剂组分包覆,评述了各种方法的优缺点,分析认为若在固体推进剂中应用,可重点发展碳包覆、金属包覆和推进剂组分包覆等方法。     

三峡工程坝址区主要工程地质问题研究

简要介绍了工程坝址区最为关键的坝基深层抗滑稳定、永久船闸高坡稳定、地下电站主厂房围岩块体稳定、断裂构造等几个工程地质问题研究过程、方法、工作内容及基本结论,并结合工程施工、监测及运行有关情况对各专题进行全面总结与评价。