Generating Questions Based on Semi-Automated and End-to-End Neural Network

With the emergence of large-scale knowledge base, how to use triple information to generate natural questions is a key technology in question answering systems. The traditional way of generating questions require a lot of manual intervention and produce lots of noise. To solve these problems, we propose a joint model based on semi-automated model and End-to-End neural network to automatically generate questions. The semi-automated model can generate question templates and real questions combining the knowledge base and center graph. The End-to-End neural network directly sends the knowledge base and real questions to BiLSTM network. Meanwhile, the attention mechanism is utilized in the decoding layer, which makes the triples and generated questions more relevant. Finally, the experimental results on SimpleQuestions demonstrate the effectiveness of the proposed approach.     

Mathematical modeling of growth of Salmonella spp. and spoilage microorganisms in raw oysters

The main objective of this study was to develop the primary and secondary models to describe the growth kinetics of Salmonella as well as background microorganisms in raw, shucked oysters. Samples, inoculated with a cocktail of two Salmonella serotypes, S. Typhimurium (CICC22956) and S. Enteritidis (CICC21482), were incubated at 4, 8, 12, 16, 20, 25, 30, 33, 37, 40, and 43 °C. Growth of Salmonella was observed at all temperatures, except at 4 °C. The background microorganisms grew at all temperatures. All growth curves clearly exhibited lag, exponential and stationary phases, and were analyzed using the Huang growth model. Three secondary models (Ratkowsky square-root, Huang square-root, and Cardinal parameter models) were compared for evaluating the effect of temperature on bacterial growth rates. Data analysis was performed using IPMP 2013, a free predictive microbiology software tool developed by the USDA ARS.The Cardinal parameters model underestimated the specific rates of the microorganisms at low temperatures. The Huang square-root model was more suitable than the Ratkowsky square-root model for describing the effect of temperature on growth of Salmonella, while the Ratkowsky square-root model, on the other hand, was more suitable for background microorganisms. For both Salmonella and background microorganisms, the logarithms of the lag phase were expressed as linear functions of the logarithms of specific growth rates. The results of this study can be used by the food retailers and regulatory agencies to estimate the microbial shelf-life of raw, shucked oysters.     

REST API技术及在移动通信网络管理中的应用

介绍了目前最炙手可热的REST架构风格，该风格顺应Web2.0的兴起，完美的匹配了云计算时代来临的可扩展要求，在各种应用场景中都得到了充分的表现。根据其技术特点，分析了该风格的API在移动通信网络管理中的应用，从网管系统内部、网管系统之间以及网管系统与上层APP应用之间等多方面对是否适用于REST风格以及如何在合适的位置使用REST API进行了分析。     

云杉八齿小蠹Ips typographus L.虚拟模型构建方法研究

本文以云杉八齿小蠹Ips typographus Linnaeus为例,经标本选取,观察虫体整体形态,绘制整体结构草图,电子显微镜观察局部、得到局部数字图像,分别建立虫体每一部分的高精度细节。把模型每部分拼装在一起组成整体模型,构建出了云杉八齿小蠹三维虚拟昆虫数字化模型。     

Classification of the cardiotocogram data for anticipation of fetal risks using machine learning techniques

The aim of the research is evaluating the classification performances of eight different machine-learning methods on the antepartum cardiotocography (CTG) data. The classification is necessary to predict newborn health, especially for the critical cases. Cardiotocography is used for assisting the obstetricians’ to obtain detailed information during the pregnancy as a technique of measuring fetal well-being, essentially in pregnant women having potential complications. The obstetricians describe CTG shortly as a continuous electronic record of the baby's heart rate took from the mother's abdomen. The acquired information is necessary to visualize unhealthiness of the embryo and gives an opportunity for early intervention prior to happening a permanent impairment to the embryo. The aim of the machine learning methods is by using attributes of data obtained from the uterine contraction (UC) and fetal heart rate (FHR) signals to classify as pathological or normal. The dataset contains 1831 instances with 21 attributes, examined by applying the methods. In the paper, the highest accuracy displayed as 99.2%.     

社交网络中基于网络结构支撑模型的谣言传播研究

结合结构支撑理论，探究节点网络结构支持力的一些性质，提出了社交网络结构中的全网支持力和被支持力的总量一致性，并进一步提出计算节点支持力的方法。谣言作为特殊信息，在支持力不同节点之间的传播特性有所不同，借鉴随机游走模型中的PageRank计算方法，对不同节点支持力的谣言传播以及传播后的辟谣状况进行了仿真模拟，结果表明支持力不同的节点对于谣言传播和辟谣影响明显。     

Reduction of waste biosolids by RAS-ozonation: Model validation and sensitivity analysis for biosolids reduction and nitrification

Biosolids reduction model by return activated sludge ozonation was validated by simulating nitrification data compiled from our pilot-scale and the literature studies. Then, a global sensitivity analysis (GSA) was performed to identify influential and non-influential parameters for biosolids reduction efficiency, change in specific nitrification activity (SNA), and alteration to expected nitrification stability. In general, the model outputs were sensitive to operational and ozone reaction parameters, but not to biochemical parameters. For operational parameters, mainly temperature and initial solids retention time (SRT) influenced all model outputs. For biosolids reduction, increase in the degradability of the influent COD decreased the reduction efficiency. For SNA, the changes were highly dependent on the influent TKN/COD ratio. Our findings also imply that the stability of the nitrification process in ozonated systems should be enhanced at constant MLVSS for warm temperatures, but could be reduced at temperatures below 12 °C and aerated SRTs below 10 days.     

Exactly satisfying initial conditions neural network models for numerical treatment of first Painlevé equation

In this paper, novel computing approach using three different models of feed-forward artificial neural networks (ANNs) are presented for the solution of initial value problem (IVP) based on first Painlevé equation. These mathematical models of ANNs are developed in an unsupervised manner with capability to satisfy the initial conditions exactly using log-sigmoid, radial basis and tan-sigmoid transfer functions in hidden layers to approximate the solution of the problem. The training of design parameters in each model is performed with sequential quadratic programming technique. The accuracy, convergence and effectiveness of the proposed schemes are evaluated on the basis of the results of statistical analyses through sufficient large number of independent runs with different number of neurons in each model as well. The comparisons of these results of proposed schemes with standard numerical and analytical solutions validate the correctness of the design models.     

Event-triggered variable horizon Supervisory Predictive control of hybrid power plants

The supervision of a hybrid power plant, including solar panels, a gas microturbine and a storage unit operating under varying solar power profiles is considered. The Economic Supervisory Predictive controller assigns the power references to the controlled subsystems of the hybrid cell using a financial criterion. A prediction of the renewable sources power is embedded into the supervisor. Results deteriorate when the solar power is unsteady, owing to the inaccuracy of the predictions for a long-range horizon of 10 s. The receding horizon is switched between an upper and a lower value according to the amplitude of the solar power trend. Theoretical results show the relevance of horizon switching, according to a tradeoff between performance and prediction accuracy. Experimental results, obtained in a Hardware In the Loop (HIL) framework, show the relevance of the variable horizon approach. Power amplifiers allow us to simulate virtual components, such as a gas microturbine, and to blend their powers with that of real devices (storage unit, real solar panels). In this case, fuel savings, reaching 15%, obtained under unsteady operating conditions lead to a better overall performance of the hybrid cell. The overall savings obtained in the experiments amount to 12%.     

A new reverse reduce-error ensemble pruning algorithm

Although greedy algorithms possess high efficiency, they often receive suboptimal solutions of the ensemble pruning problem, since their exploration areas are limited in large extent. And another marked defect of almost all the currently existing ensemble pruning algorithms, including greedy ones, consists in: they simply abandon all of the classifiers which fail in the competition of ensemble selection, causing a considerable waste of useful resources and information. Inspired by these observations, an interesting greedy Reverse Reduce-Error (RRE) pruning algorithm incorporated with the operation of subtraction is proposed in this work. The RRE algorithm makes the best of the defeated candidate networks in a way that, the Worst Single Model (WSM) is chosen, and then, its votes are subtracted from the votes made by those selected components within the pruned ensemble. The reason is because, for most cases, the WSM might make mistakes in its estimation for the test samples. And, different from the classical RE, the near-optimal solution is produced based on the pruned error of all the available sequential subensembles. Besides, the backfitting step of RE algorithm is replaced with the selection step of a WSM in RRE. Moreover, the problem of ties might be solved more naturally with RRE. Finally, soft voting approach is employed in the testing to RRE algorithm. The performances of RE and RRE algorithms, and two baseline methods, i.e., the method which selects the Best Single Model (BSM) in the initial ensemble, and the method which retains all member networks of the initial ensemble (ALL), are evaluated on seven benchmark classification tasks under different initial ensemble setups. The results of the empirical investigation show the superiority of RRE over the other three ensemble pruning algorithms.