基于注意力机制与文本信息的用户关系抽取

随着社交媒体的发展,用户之间的关系网络对于社交媒体的分析有很大的帮助。因此,该文主要研究用户好友关系检测。以往的关于用户好友关系抽取的研究主要基于社交媒体上的结构化信息,比如其他好友关系,用户的不同属性等。但是,很多时候用户本身并没有大量的好友信息存在,同时也不一定有很多确定的属性。因此,我们希望基于用户发表的文本信息来对用户关系进行预测。不同于以往的潜在好友推荐算法,该文提出了一种基于注意力机制以及长短时记忆网络(long short-term memory,LSTM)的好友关系预测模型,将好友之间的评论分开处理,通过分析用户之间的评论来判断是否具备一定的好友关系。该模型将好友双方信息拼接后的结果作为输入,并将注意力机制应用于LSTM的输出。实验表明,用户之间的评论对于好友关系预测确实有较大的实际意义,该文提出的模型较之于多个基准系统的效果,取得了明显的提升。在不加入任何其它非文本特征的情况下,实验结果的准确率达到了77%。     

Spare parts inventory management with demand lead times and rationing

We study an inventory system that consists of two demand classes. The orders in the first class need to be satisfied immediately, whereas the orders in the second class are to be filled in a given demand lead time. The two classes are also of different criticality. For this system, we propose a policy that rations the non-critical orders. Under a one-for-one replenishment policy with backordering and for Poisson demand arrivals for both classes, we first derive expressions for the service levels of both classes. The service level for the critical class is an approximation, whereas the service level for the non-critical class is exact. We then conduct a computational study to show that our approximation works reasonably, the benefits of rationing can be substantial, and the incorporation of demand lead time provides more value when the demand class with demand lead time is the critical class. The research is motivated by the spare parts service system of a major capital equipment manufacturer that faces two types of demand. For this company, the critical down orders need to be satisfied immediately, while the less critical maintenance orders can be satisfied after a fixed demand lead time. We conduct a case study with 64 representative parts and show that significant savings (as much as 14% on inventory on hand) are possible through incorporation of demand lead times and rationing.     

市区微蜂窝传播路径损耗模式的比较检验

讨论了几种微蜂窝路径损耗模式,并用我国某城市测量的微蜂窝传播数据对COST231WI,修正COST231WI和Xia模式进行了检验,结果表明修正COST231模式的误差最小,Xia模式次之,COST231WI模式最差。从而得到修正COST231模式比较符合我国实际的微蜂窝传播预测模式。     

Exceptions that prove the rule--Using a theory of motivated social cognition to account for ideological incongruities and political anomalies: Reply to Greenberg and Jonas (2003).

A meta-analysis by J. T. Jost, J. Glaser, A. W. Kruglanski, and F. J. Sulloway (2003) concluded that political conservatism is partially motivated by the management of uncertainty and threat. In this reply to J. Greenberg and E. Jonas (2003), conceptual issues are clarified, numerous political anomalies are explained, and alleged counterexamples are incorporated with a dynamic model that takes into account differences between "young" and "old" movements. Studies directly pitting the rigidity-of-the-right hypothesis against the ideological extremity hypothesis demonstrate strong support for the former. Medium to large effect sizes describe relations between political conservatism and dogmatism and intolerance of ambiguity; lack of openness to experience; uncertainty avoidance; personal needs for order, structure, and closure; fear of death; and system threat. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Effects of situational demand upon social enjoyment and preference in schizotypy.

The socioemotional functioning of schizophrenic and schizotypic individuals is marked by withdrawal, poor organization, and limited emotional displays. Such behavioral tendencies and lack of social enjoyment in schizotypy could be linked to the relative situational demands or role ambiguity inherent in specific social activities. To determine whether high-schizotypy individuals prefer more clearly role-defined social activities (e.g., visiting relatives) to more ambiguous, novel situations (e.g., going alone to a party), the authors gathered reports from 52 high-schizotypy and 60 low-schizotypy individuals on their enjoyment and frequency of engaging in social situations varying in relative situational demand. Parallel reports were obtained from knowledgeable others. Group × Situational Demand interactions revealed the hypothesized pattern of reduced frequency and enjoyment ratings for ambiguous or novel situations by the high-schizotypy participants in both self and others' reports. Groups were more comparable in their reported frequency and enjoyment of less ambiguous situations. Results suggest the importance of situational demands in the socioemotional experience and behavioral withdrawal in schizotypy. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

An approach to efficient business intelligent system for financial prediction

A concept of business intelligent system for financial prediction is considered in this paper. It provides data needed for fast, precise and good business decision support to all levels of management. The aim of the project is the development of a new online analytical processing oriented on case-based reasoning (CBR) where a previous experience for every new problem is taken into account. Methodological aspects have been tested in practice as a part of the management information system development project of “Novi Sad Fair”. A case study of an improved application of CBR in prediction of future payments is discussed in the paper. This paper is originally presented at The International Conference on Hybrid Information Technology 2006, at the special session on “Intelligent Information Systems for Financial Engineering”, November 2006 in Cheju Island, Korea.     

Measurement and prediction of pressure drop in two-phase flow

Experimental data for air–water two-phase co-current flow in two different pipe diameters were used to test the prediction of pressure drop by a number of existing theories and correlations. Several models are shown to be useful for prediction, particularly with the stratified regimes which have proved difficult to handle in the past. The model suggested by Olujic proved to be of particular value.     

On-demand Virtual Neonatal Intensive Care units supporting rural, remote and urban healthcare with Bush Babies Broadband

Premature and ill term babies born in regional Australia must be moved to another hospital with Neonatal Intensive Care Unit (NICU) facilities. Transferred pre-term and critically ill term babies have higher mortality rates and much higher rates of long term disability than similar babies born in hospitals with NICU facilities. This paper details the Bush Babies Broadband project that aims to significantly improve the quality of treatment for babies born in rural, remote as well as urban areas by providing the first on-demand virtual NICU architecture in Australia. Real-time data collected from medical monitors and ventilators attached to the baby, audiovisual streams and static physiological data such as X-ray images are transmitted to the consulting Neonatologist to gain a better picture of the patient's condition than is currently available. The key contribution of this significant research is the infrastructure providing a mechanism for Neonatologists to receive information directly from a regional hospital, thereby preventing, in some cases, the immediate need to move the baby. A key benefit of this framework is that it is available to link regional hospitals with the supporting NICU Neonatologist ‘on demand’ eliminating the need to establish permanent point to point connections. This paper further describes the application of that architecture to a specific pilot connecting the Bathurst regional hospital with Neonatologists within the NICU at Nepean Hospital, Penrith Australia.     

Sensitivity of generation reserve requirements in deregulated power systems

Since demand for power exhibits great variability the amount of firm energy to be purchased to meet real time demand based on forecasts is usually different from the realized demand for that period. The role of generation reserves is to meet the real time fluctuations of power demand. The predictable part of the demand is met through purchases of firm energy. In this paper a model is presented to determine optimal quantities of firm energy and generation reserves to meet random demands. The model is then parameterized introducing a set of factors to perform a sensitivity study. A full factorial experiment is designed to study the impact of five factors on the response variable (i.e., proportion of generation reserves on the total purchased quantity). An example consisting of 640 simulations corresponding to 2⁵ treatment combinations evaluated over 20 randomly generated mean demands is used to identify significant factors on the response variable. Results from the experiment suggest that generation reserve requirements should be adjusted considering changes in significant factors and in the mean demand over the dispatch interval.