Modeling under-resourced languages for speech recognition

One particular problem in large vocabulary continuous speech recognition for low-resourced languages is finding relevant training data for the statistical language models. Large amount of data is required, because models should estimate the probability for all possible word sequences. For Finnish, Estonian and the other fenno-ugric languages a special problem with the data is the huge amount of different word forms that are common in normal speech. The same problem exists also in other language technology applications such as machine translation, information retrieval, and in some extent also in other morphologically rich languages. In this paper we present methods and evaluations in four recent language modeling topics: selecting conversational data from the Internet, adapting models for foreign words, multi-domain and adapted neural network language modeling, and decoding with subword units. Our evaluations show that the same methods work in more than one language and that they scale down to smaller data resources.     

一种面向维文网站的动态天气图文信息服务支持系统

维吾尔文地理信息服务的研究与应用的匮乏,使维吾尔用户很难享受基于本民族语言的地理信息服务。该文以新疆动态天气图文信息服务为例,提出了一种面向维文网站的WebGIS服务解决方案。方案首先通过集成免费Google Maps API地图服务和实时免费Yahoo Weather RSS天气信息,制作新疆县域维吾尔文天气信息服务模块;其次,将该模块作为引用对象,根据用户对动态天气图文信息服务框架大小和位置的设定,个性化地嵌入到任意维吾尔文大众网站中,使更广大的维吾尔用户享受WebGIS带来的信息服务。该研究推动了WebGIS在维吾尔语语境的新应用,为新疆各族群众提供了便捷动态天气图文信息服务。     

维吾尔文智能输入法研究

开发智能化的输入法是维吾尔文输入技术的一个发展方向,也是维吾尔文自然语言信息处理的基础性研究课题。该文根据维吾尔文自身特点,对用户输入过程中所犯各种错误进行分析,设计并实现了词语搭配知识库,提出了基于二元语法模型的一种新型维吾尔文输入算法,并给出了实现其关键模块功能的程序流程图。它具有自动预测、自动联想输入和自动校对等功能。     

Lossless convexification of a class of optimal control problems with non-convex control constraints

We consider a class of finite time horizon optimal control problems for continuous time linear systems with a convex cost, convex state constraints and non-convex control constraints. We propose a convex relaxation of the non-convex control constraints, and prove that the optimal solution of the relaxed problem is also an optimal solution for the original problem, which is referred to as the lossless convexification of the optimal control problem. The lossless convexification enables the use of interior point methods of convex optimization to obtain globally optimal solutions of the original non-convex optimal control problem. The solution approach is demonstrated on a number of planetary soft landing optimal control problems.     

基于CG-MSNWF的空时自适应抗干扰算法研究

针对多级维纳滤波MSNWF(Multi-Stage Nested Winner Filter)算法的计算量大,在处理高维数据时不能满足实时性的需要问题,采用基于共轭梯度CG(Conjugate Gradient)的多级维纳滤波算法—CG-MSNWF。在相同的干扰抑制性能条件下,该算法相比于MSNWF不需要后向迭代过程,降低了计算量和对硬件存储器的要求,提高了算法的收敛速度,满足了实时性的需求。仿真结果证实了算法的实效性。     

蒙古文整词计算机生成理论研究

采用面向对象方法,模拟传统蒙古文整词各种形式构成机理,提出了几种蒙古语整词计算机生成数据模型。文章主要依据整词计算机生成三种模型,探讨了传统蒙古文整词计算机最优化生成理论所涉及的精确度、时间复杂度、空间复杂度三项基本要素以及最优化生成必须考虑的整词复杂特征载荷与一体化合一计算知识表示方法和计算结构,证明了“B - J - T= W”数据模型是传统蒙古文整词计算与生成最优化对象模型。     

存储器替换机制及其实现

提出一种解决PACT01一种结合动态可编程逻辑阵列(DPGA)的处理器的新型体系结制中cache的一致性与同步性问题的算法,并且解决多线程支持的快速上下文切换及快速用户级操作问题。存储器替换机制是解决cache的一致性问题及当cache未命中时从局部或远程存储器到cacbe存储器的数据替换问题的一种硬件实现方法,产生冲突的原因是由于多线程并行的写入／读取的位置相同和读或写的位置相同。文中选择的是相联映射策略,同时也选择了最少最近使用LRU算法,即在cache未命中时替换最少最近使用的参考块,为实现LRU算法设置了与每块相对应的计数器。     

Prediction of the binding mode of biarylpropylsulfonamide allosteric AMPA receptor modulators based on docking, GRID molecular interaction fields and 3D-QSAR analysis

A novel approach of combining flexible molecular docking, GRID molecular interaction fields, analysis of ligand-protein hydrogen bond interactions, conformational energy penalties and 3D-QSAR analysis was used to propose a binding mode in the dimer interface of the iGluR2 receptor for the biarylpropylsulfonamide class of positive allosteric AMPA modulators. Possible binding poses were generated by flexible molecular docking. GRID molecular interaction fields of the binding site, ligand-protein hydrogen bonding interactions and conformational energy penalties were used to select the most likely binding mode. The selected binding poses were subjected to a 3D-QSAR analysis using previously published activity data. The resulting model (2 LVs, R2=0.89, q2=0.61) predicted the activities of the compounds in the test set with a standard deviation on error of prediction of 0.17. The proposed binding mode was validated by interpretation of the PLS-coefficient regions from the 3D-QSAR analysis in terms of interactions between the receptor and the modulators.     

Digital microfluidic design and optimization of classic and new fluidic functions for lab on a chip systems

This paper deals with microfluidic studies for lab-on-a-chip development. The first goal was to develop microsystems immediately usable by biologists for complex protocol integrations. All fluid operations are performed on nano-liter droplet independently handled solely by electrowetting on dielectric (EWOD) actuation. A bottom-up architecture was used for chip design due to the development and validation of elementary fluidic designs, which are then assembled. This approach speeds up development and industrialization while minimizing the effort in designing and simplifying chip-fluidic programming. Dispensing reproducibility for 64 nl droplets obtained a CV below 3% and mixing time was only a few seconds. Ease of the integration was demonstrated by performing on chip serial dilutions of 2.8-folds, four times. The second part of this paper concerns the development of new innovative fluidic functions in order to extend EWOD-actuated digital fluidics’ capabilities. Experiments of particle dispensing by EWOD droplet handling are reported. Finally, work is shown concerning the coupling of EWOD actuation and magnetic fields for magnetic bead manipulation.     

蚁群算法参数对物流配送路径问题求解的影响

描述了蚁群系统算法求解物流配送路径问题的过程;通过仿真实验分析了蚁群系统算法中各参数对物流配送路径问题求解的影响.