Exploring synergies between content-based filtering and Spreading Activation techniques in knowledge-based recommender systems

Recommender systems fight information overload by selecting automatically items that match the personal preferences of each user. The so-called content-based recommenders suggest items similar to those the user liked in the past, using syntactic matching mechanisms. The rigid nature of such mechanisms leads to recommending only items that bear strong resemblance to those the user already knows. Traditional collaborative approaches face up to overspecialization by considering the preferences of other users, which causes other severe limitations. In this paper, we avoid the intrinsic pitfalls of collaborative solutions and diversify the recommendations by reasoning about the semantics of the user’s preferences. Specifically, we present a novel content-based recommendation strategy that resorts to semantic reasoning mechanisms adopted in the Semantic Web, such as Spreading Activation techniques and semantic associations. We have adopted these mechanisms to fulfill the personalization requirements of recommender systems, enabling to discover extra knowledge about the user’s preferences and leading to more accurate and diverse suggestions. Our approach is generic enough to be used in a wide variety of domains and recommender systems. The proposal has been preliminary evaluated by statistics-driven tests involving real users in the recommendation of Digital TV contents. The results reveal the users’ satisfaction regarding the accuracy and diversity of the reasoning-driven content-based recommendations.     

非单调的多层神经网络

介绍了应用非单调函数作为隐层节点的激励函数的前馈多层神经网络的性能。并对以前神经网络中的两个难点：局部极小和收敛速度慢的问题进行分析。表明了采用非单调激励函数的神经网络能够较好地解决局部极小的问题。最后的实验结果也表明了该网络的性能要好于采用Ｓ型函数的神经网络的性能。     

重质碳酸钙活化的研究

采用硬脂酸对1500目重质碳酸钙进行湿法活化,利用正交实验研究了硬脂酸加入量、反应温度和反应时间对活化效果的影响。测定了改性前后重质碳酸钙活化度,并用红外光谱进行表征。     

Activation flotation and mechanism of lime-depressed pyrite with oxalic acid

Flotation tests,contact angle measurements,infrared spectrum analyses,X-ray analyses and computer simulation were carried out in order to study the activation mechanism of lime-depressed pyrite with ox...     

Saccadic velocity and activation: development of a diagnostic tool for assessing energy regulation

The hypothesis was tested that peak velocity of saccadic eye movements in visual motor tasks varies with variables related to energy regulation. The hypothesis is based on the cognitive-energetical performance model of Sanders. An experimental paradigm was developed in which saccadic peak velocity of task-relevant eye movements is measured while a choice reaction task is carried out. Confounding factors of saccadic amplitude and movement direction were controlled. The task was designed in such a way that in each trial subjects performed a target saccade towards an imperative stimulus and a return saccade after the manual response back to the centre of the screen. For both types of saccades the experimental variables were foreperiod duration (short versus long), knowledge of results (with versus without), postsaccadic demand (low versus high) and time on task (five 30-min intervals). In two experiments, there are main and interaction effects of the task variables on peak saccadic velocity. Return saccades are slower than target saccades, but not in the case of high postsaccadic demand. Knowledge of results increases peak saccadic velocity, but more so for return than for target saccades. Time on task leads to a decrease in peak saccadic velocity, which is much stronger for return than for target saccades; furthermore this effect is more pronounced after short than after long foreperiods. Peak saccadic velocity is changed within seconds. The results support the hypothesis. Peak saccadic velocity of task related eye movements reflects energy regulation during task performance. The paradigm will be developed as a diagnostic tool in workload measurement.     

MBE梯度掺杂GaAs光电阴极激活实验研究

本文首次利用分子束外延(MBE)生长了多种由体内到表面掺杂浓度由高到低的梯度掺杂反射式GaAs光电阴极材料,并进行了激活实验,结果表明,表面低掺杂浓度适中,外延层厚度2 μm～3 μm以及衬底为重掺杂p型GaAs的梯度掺杂GaAs光电阴极能够获得较高灵敏度.在优化激活工艺的条件下,梯度掺杂GaAs光电阴极获得了1798μA/lm的最高积分灵敏度,比采用同样方法制备的均匀掺杂GaAs光电阴极高30%以上.梯度掺杂GaAs光电阴极表面掺杂浓度较均匀掺杂的低,第一次给Cs时间较长,第一次Cs、O交替时要调整好Cs/O比,并在整个激活过程中保持不变.一个高量子效率梯度掺杂GaAs光电阴极的获得依赖于梯度掺杂结构和激活工艺两个方面的优化.     

Effect of temperature and α-irradiation on gas permeability for polymeric membrane

In the present study the polyethersulphone (PES) membranes of thickness (35 ±2) μm were prepared by solution cast method. The permeability of these membranes was calculated by varying the temperature and by irradiation of α ions. For the variation of temperature, the gas permeation cell was dipped in a constant temperature water bath in the temperature range from 303–373 K, which is well below the glass transition temperature (498 K). The permeability of H₂ and CO₂ increased with increasing temperature. The PES membrane was exposed by a-source (₉₅Am₂₄₁) of strength (1 μ Ci) in vacuum of the order of 10⁻⁶ torr, with fluence 2.7 × 10⁷ ions/cm². The permeability of H₂ and CO₂ has been observed for irradiated membrane with increasing etching time. The permeability increases with increasing etching time for both gases. There was a sudden change in permeability for both the gases when observed at 18 min etching. At this stage the tracks are visible with optical instrument, which confirms that the pores are generated. Most of pores seen in the micrograph are circular cross-section ones.     

Influences of variable scales and activation functions on the performances of multilayer feedforward neural networks

This paper gives insight into the methods about how to improve the learning capabilities of multilayer feedforward neural networks with linear basis functions in the case of limited number of patterns according to the basic principles of support vector machine (SVM), namely, about how to get the optimal separating hyperplanes. And furthermore, this paper analyses the characteristics of sigmoid-type activation functions, and investigates the influences of absolute sizes of variables on the convergence rate, classification ability and non-linear fitting accuracy of multilayer feedforward networks, and presents the way of how to select suitable activation functions. As a result, this proposed method effectively enhances the learning abilities of multilayer feedforward neural networks by introducing the sum-of-squares weight term into the networks’ error functions and appropriately enlarging the variable components with the help of the SVM theory. Finally, the effectiveness of the proposed method is verified through three classification examples as well as a non-linear mapping one.     

镍氢电池热处理对其内压及活化的影响

采用封口化成工艺的镍氢电池在活化初期及大倍率充电时内压过高,造成电池漏液爬碱,容量下降,寿命变短,安全性差,而且化成时间较长。本文报导了镍氢电池经过热处理以后,可以显著降低大倍率充电时电池内压,缩短化成时间,有利于提高电池的综合性能。     

胶粉的活化改性初步研究

活化剂Ａ和活化剂Ｂ并用、经过热处理可达到活化改性胶粉的目的。掺有活化胶粉的丁苯胶与掺有未活化胶粉的丁苯胶物机性能相比，除扯断伸长率略有下降外，其它静态、动态性能均有所提高。