Learning Query Ambiguity Models by Using Search Logs

dentifying ambiguous queries is crucial to research on personalized Web search and search result diversity. Intuitively, query logs contain valuable information on how many intentions users have when issuing a query. However, previous work showed user clicks alone are misleading in judging a query as being ambiguous or not. In this paper, we address the problem of learning a query ambiguity model by using search logs. First, we propose enriching a query by mining the documents clicked by users and the relevant follow up queries in a session. Second, we use a text classifier to map the documents and the queries into predefined categories. Third, we propose extracting features from the processed data. Finally, we apply a state-of-the-art algorithm, Support Vector Machine (SVM), to learn a query ambiguity classifier. Experimental results verify that the sole use of click based features or session based features perform worse than the previous work based on top retrieved documents. When we combine the two sets of features, our proposed approach achieves the best effectiveness, specifically 86% in terms of accuracy. It significantly improves the click based method by 5.6% and the session based method by 4.6%.     

稳态工作条件下功率晶体管结温的测量与控制

为了在试验周期中了解晶体管的结温,提出一种在功率晶体管稳态工作寿命试验过程中结温的测量与控制方法.着重介绍了基于理想pn结肖克莱方程的结温测量原理,及试验过程中结温测控的技术难点和解决方案,指出了晶体管结温计算中存在的问题和修正办法,实验结果证明了该方法的可行性.     

跟踪式光伏发电系统研究

根据某地实测的太阳辐射数据,仿真比较了配备有单轴跟踪和双轴跟踪等4种跟踪控制的光伏发电系统与固定式光伏发电系统的太阳辐射利用率。并在此基础上对4种跟踪系统的跟踪角控制规律及跟踪控制方式进行了详细的分析,得出倾纬度角单轴跟踪系统控制规律最为简单,算法实施更为实用的结论。同时,也在理论上证明了采用步进式控制方式的跟踪系统能够在保持较高太阳辐射利用率情况下简化控制系统设计,有利于工程设计及应用。     

爆轰法制备与固载钛基氧化物的研究

利用钛酸丁酯与硝酸铵溶液制备的含钛凝胶跟黑索金混合后获得混合炸药, 然后在密闭的爆轰反应釜内引爆柱状混合炸药, 同时将一铝板放置于药柱正下方研究爆轰产物的固载问题。收集到的样品用X射线衍射仪、透射电镜以及扫描电镜进行了表征, 发现样品中含有FeTiO₃, 少量的金红石、锐钛矿及Al₂O₃。钛基氧化物颗粒呈球形, 尺寸大小在微米级别。从扫描电镜分析, 钛基氧化物颗粒被成功地固载于铝板之上, 颗粒约10 µm。爆轰法能够很方便地制备钛基氧化物及制备颗粒-板结构材料。     

条形基础两侧双灰桩加固地基室内模型试验研究

以相似理论为指导,进行了在条形基础两侧设置双灰桩加固地基的室内模型试验,研究了加固后地基的承载力、沉降随荷载变化规律、地基应力的大小及分布规律,分析了该加固技术的承载机理、加固机理和加固效果.试验结果表明,双灰桩可以有效地提高地基承载能力,减小地基沉降量;双灰桩使地基中的应力不易向桩外土扩散,并能使地基应力向更深处扩散...     

Ray tracing on extended hemispherical and elliptical silicon dielectric lenses

In this paper, extended hemispherical and elliptical silicon dielectric lenses are compared by means of ray tracing. It is shown that the optical focuses of two lenses are not at same position though the elliptical lens is synthesized from an extended hemispherical lens by carefully choosing a particular extension length as stated in ref[1]. It may be used to explain why the best compromise between alignment, directivity and Gaussian-coupling effciency will be obtained if the feed antenna is placed at the region between 2.2 mm and 2.4 mm of extension length (for 13.7 mm diameter lens, dielectric contant = 11.7).     

D-S证据理论在时-空信息融合中的应用

多传感器的信息融合技术应用于导航、目标跟踪、目标识别等很多领域。目前用于信息融合的主要方法就是证据理论技术。论文给出了三种基于D-S证据理论的多传感器多个测量周期的信息融合(时-空信息融合)的方法。在仿真实验中对这三种方法进行了比较,分布式有反馈的方法是三种方法中最好的。     

红外双边滤波时域高通非均匀性校正

红外焦平面阵列的非均匀性噪声是制约红外成像质量的主要因素。本文在研究了传统的时域高通滤波法及其两种改进算法的基础上,提出了一种改进的基于双边滤波的非均匀性自适应校正算法,在这种方法中引入了一个由双边滤波系数矩阵推得的二次校正矩阵,该矩阵能够判别原始图像与双边滤波所得图像的差图像中场景的边缘部分,并进行自适应的抑制,使校正参数的计算更加准确。实验部分通过对加模拟噪声图像序列和实际非均匀性图像的校正证明本文的改进算法比其他两种改进算法有更好的校正效果。     

渔药残留检测技术比较分析及其研究进展

我国作为渔药生产使用的大国,存在渔药使用不规范、错用、乱用等现,食品安全检测技术是食品质量保证和食品安全管理的重要基础。相对于其他食品的检测,我国水产品安全检测的技术相对落后,已建立起来的专门针对水产品质量检测的方法相对较少,渔药残留现象日趋严重。目前应用于渔药残留的检测方法主要有色谱质谱技术、免疫分析、电化学分析、毛细管电泳和高分辨率质谱谱库筛查等。本文结合国内外的实际应用情况,对以上检测技术从原理到在渔药残留检测中应用优势、不足及发展方向进行分析比较,使其在渔药检测中有的放矢,发挥更大的能量,以期为养殖企业和相关的研究检测部门,针对不同的检测对象选择最合适的检测技术提供借鉴和帮助。     

Contactless and Rapid Discrimination of Improvised Explosives Realized by Mn2+ Doping Tailored ZnS Nanocrystals

In order to sensitively, selectively, and rapidly detect the constituents relevant to improvised explosive devices (IEDs), the sensing properties of ZnS nanocrystals (NCs) are regulated by tailoring the doping level of Mn²⁺. The responses of the sensors fabricated by ZnS NCs with different Mn‐doping levels (Mn:ZnS) toward the constituents, such as sulphur powder and black powder, generally increases first and then decreases with the increase of the concentration of doped Mn²⁺, and reaches the climate with an atomic ratio of 2.23% at room temperature. The sensory array based on eight sensors of Mn:ZnS NCs can realize the detection of two typical military explosives and six constituents relevant to IEDs within 7 s and can recover in 19 s. Furthermore, the fingerprinting of the constituents is achieved by pattern recognizing the inherent kinetics and thermodynamics of interaction between the sensory array and the constituents. Thus, a simple chemiresistive sensing strategy based on semiconductor NCs which can rapidly, supersensitively, and discriminatively detect the constituents relevant to IEDs is explored for the first time.