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.     

Anaerobic co-digestion of food waste and FOG with sewage sludge – realising its potential in Ireland

The severe environmental pollution in many countries is caused by indiscriminate discharge of large quantities of food waste (FW), fat oil and grease (FOG) and sewage sludge (SS) to the environment. There are many possible treatment routes, but anaerobic digestion (AD) is now well accepted for treating several kinds of organic wastes. But AD of FW alone presents some operational challenges because of substrates and variability. Anaerobic co-digestion of two or more substrates is better than single substrate digestion. This can use a plant’s unused capacity, in line with the trend to renewable energy. Co-digestion technology, although well established in many European countries, is still in its infancy in Ireland. There are problems with different regulatory arrangements. They should be resolved. The paper reviews anaerobic co-digestion technology is reviewed, with special focus on possible application in Ireland.     

SDN中基于KMOBPSO的高可靠性控制器部署算法

针对SDN中控制器系统的单节点故障问题,兼顾系统成本和系统时延,应用N+1冗余备份模型来提高SDN控制器部署的可靠性,并将其抽象为多目标优化问题.同时,提出了一种融合K-means聚类算法和遗传算子的多目标二进制粒子群算法——KMOBPSO算法,以求解SDN控制器高可靠性部署问题的解.仿真结果表明,所提算法具有求解精度高、分布均匀、沿Pareto前沿面覆盖广的特点,能够显著提高SDN中控制器部署的可靠性.     

我国钻井装备现状及其相关配套技术

结合目前国外钻机现状以及国际石油钻井市场对钻机的配套要求,分析指出我国目前陆地钻机存在的不足与差距,对我国钻机及其配套技术的发展提出建议。     

Characteristic of polysaccharides from Flammulina velutipes in vitro digestion under salivary,simulated gastric and small intestinal conditions and fermentation by human gut microbiota

In this study, in vitro digestion and fermentation of Flammulina velutipes -derived polysaccharides (FVP) were investigated. It was found that FVP mainly consisted of 48.45% glucose, 15.40% mannose, 14.60% xylose, 11.80% fucose and 9.90% galactose. The -human saliva, simulated gastric and small intestinal juices conditions did not break down the FVP. Based on in vitro fermentation tests, FVP modulated the composition of gut microbiota by elevating the amounts of Bifidobacteriaceae and Bacteroidaceae and reducing the numbers of genera Lachnospiraceae and Enterococcaceae. Meanwhile, FVP affected the synthesis of short-chain fatty acids derived from gut microbiota.     

125I粒子条联合胆道支架植入治疗恶性梗阻性黄疸22例疗效分析

【摘要】　 　目的 研究和探讨125I粒子条联合胆道支架植入治疗恶性梗阻性黄疸的临床疗效。方法 2011年6月—2013年9月收治恶性梗阻性黄疸患者40例。行胆道支架植入术治疗患者18例，为对照组；采用125I粒子条联合胆道支架植入患者22例，为观察组。两组均采用经皮肝穿刺胆道引流（PTCD）并胆道支架植入。结果 观察组与对照组的平均胆道开通时间分别为（8.7 ± 0.7）和（6.2 ± 0.4）个月，两组比较差异有统计学意义（P < 0.05）。观察组与对照组术后平均生存时间分别为（11.4 ± 0.8）和（8.7 ± 0.5）个月，生存时间的差异有统计学意义（P < 0.05）。术后近期疗效及并发症两组之间的比较无明显差异（P > 0.05），但是远期疗效有显著差异（P < 0.05）。结论 与对照组相比，观察组的术后生存时间及胆道开通时间明显延长，对于125I粒子条联合支架治疗恶性梗阻性黄疸值得进一步临床研究。