Relevance-based language modelling for recommender systems

Relevance-Based Language Models, commonly known as Relevance Models, are successful approaches to explicitly introduce the concept of relevance in the statistical Language Modelling framework of Information Retrieval. These models achieve state-of-the-art retrieval performance in the pseudo relevance feedback task. On the other hand, the field of recommender systems is a fertile research area where users are provided with personalised recommendations in several applications. In this paper, we propose an adaptation of the Relevance Modelling framework to effectively suggest recommendations to a user. We also propose a probabilistic clustering technique to perform the neighbour selection process as a way to achieve a better approximation of the set of relevant items in the pseudo relevance feedback process. These techniques, although well known in the Information Retrieval field, have not been applied yet to recommender systems, and, as the empirical evaluation results show, both proposals outperform individually several baseline methods. Furthermore, by combining both approaches even larger effectiveness improvements are achieved.     

A cross-cultural comparison of expectations in romantic relationships: India and the United States

ABSTRACT

This research investigated romantic expectations in a cross-cultural comparison of India and the United States (US). The study was grounded in expectancy violations theory (EVT), which argues that expectancies are a universal phenomenon whose content is influenced by communicator, relational, and contextual factors. The expectations of romantic partners in the two countries were assessed in conjunction with relational models, sex, and gender roles in a cross-sectional study (N?=?366). Results indicated that romantic partners from the two countries placed different importance on specific expectations. Furthermore, expectations were successfully predicted by one’s culture, relational models, sex, and gender roles.     

The role of business analytics capabilities in bolstering firms’ agility and performance

Many companies invest considerable resources in developing Business Analytics (BA) capabilities to improve their performance. BA can affect performance in many different ways. This paper analyses how BA capabilities affect firms’ agility through information quality and innovative capability. Furthermore, it studies the moderating role of environmental turbulence, both technological and in the market. The proposed model was tested using statistical data from 154 firms with two respondents (CEO and CIO) from each firm. The data were analysed using Partial Least Squares (PLS)/Structured Equation Modelling (SEM). Our results indicate that BA capabilities strongly impact a firm’s agility through an increase in information quality and innovative capability. We also discuss that both market and technological turbulence moderate the influence of firms' agility on firms' performance.     

The development of students' justifications for their positions regarding two theoretical models: Electron cloud or sodium chloride crystal—After engaging in different learning activities

The purpose of this study was to explore high school students' ideas regarding two theoretical scientific models, either electron cloud or sodium chloride crystal, in the context of active learning in small groups. Conversations among peers regarding these models took place during two types of active learning activities: small-group discussion and whole-class debate. The study was conducted in four different high school classes, each of which was in a different school for girls in Oman. The study included 108 grade 10 female students. Two of the classes discussed the electron cloud and the other two classes discussed the sodium chloride model. Qualitative data included students' written responses to prompts, class worksheets, and field notes of student ideas in class debates. In each class, the teacher used a teaching sequence during which the participants expressed their justifications for their positions in writing regarding the particular model on five different occasions, as they progressed through three interactive small group learning activities. The participants' written responses were analyzed using a coding scheme comprising of eight different categories describing the participants' type of justifications regarding the theoretical scientific models: nonsense, approval, mental, experimental, appreciative, external, structural, and modeling. The findings indicated that participants' justifications for their positions regarding theoretical scientific models tended to change over time following each group learning activity. Participants focused their discussion more on external factors, such as the teacher, textbook, religion, and media, after discussions with peers in small groups. In contrast, later their attention focused more on the submicroscopic structural orientations of the model under study during and after engaging in a debating activity. The researchers reasoned that the nature of cognitive demands during each type of active learning activity might play a role in this regard. However, further research to advance the understanding of this phenomenon is needed.     

Where are the numbers? Counting museum visitors in France

ABSTRACT

In the early twenty-first century, the Louvre is the most visited museum in the world. Yet little is known about how visit numbers to French museums developed. Compared to the Anglo-Saxon world, the collection and publication of visit data began late and was initially far from systematic. Some figures were collected in the late nineteenth century, but correspondence from the early twentieth century indicates that not even the Ministry of Fine Arts, overseeing the National Museums, was always aware of them. More complete numbers appeared from 1922 when entrance fees were introduced. However, visitors entering on free days were not yet counted. In the 1930s, data collection for the National Museums was systematized by the Directorate of National Museums, but figures were rarely published. The aim of this article is, therefore, to establish the sources that might be used for a quantitative approach to museum visiting in France and to reflect on the reasons for the initial indifference towards the counting of visitors and the standardisation with international practices over time. The article argues that the triggers for changes in the culture of counting in France were both internal and external. A prolonged debate about the introduction of entrance fees took place from the start of the Third Republic to the interwar years and let to the search for existing numbers in France and abroad. The Fine Arts Administration compiled data about practices in other countries and a number of monographs on the subject were published. The press also frequently referenced foreign examples. These documents provide a fascinating insight into comparisons and emulation of foreign practices at the time, allowing us to rethink the modern obsession with counting as the result of a transnational process.     

Discrete Population Models: Why They Belong in a Differential Equations Course

We make a case for including difference equations, in particular the discrete logistic equation, in basic differential equations courses. Contrasting the behavior of discrete and continuous models enriches students’ understanding of both modeling and differential equations. To facilitate sharing discrete population models with students, some materials including class handouts and an “exploratory” computer assignment are provided.     

What causes prejudice? How may we solve it? Lay beliefs and their relations with classical and modern prejudice and social dominance orientation

This research aimed to examine lay beliefs about the causes of and solutions to ethnic prejudice towards immigrant populations in the Italian context. The research had two main goals: (a) to explore in Italy what non-experts think causes ethnic prejudice against immigrants, and how they suggest we combat it; (b) to assess whether lay perceptions of causes of ethnic prejudice mediate the relationship between SDO and ethnic prejudice, and whether such mediational effects would depend on the form of ethnic prejudice considered (classical vs. modern). In particular, we hypothesized that lay beliefs about the causes of (but not lay beliefs about solutions to) prejudice towards immigrants may mediate the relation between SDO and prejudice towards immigrants. 520 Italian high school students participated in the study and completed a set of questionnaires regarding these issues. Results showed that Italian respondents were especially likely to attribute the causes of ethnic prejudice towards immigrants to ignorance and close-mindedness, and to recommend as main solutions open-mindedness and tolerance of others’ values. Moreover, as predicted, beliefs about causes of ethnic prejudice significantly mediated the relation between SDO and modern (but not classical) prejudice. Implications of our findings for social dominance theory were discussed.     

An integrated model of decision-making in health contexts: the role of science education in health education

Health education is to foster health literacy, informed decision-making and to promote health behaviour. To date, there are several models that seek to explain health behaviour (e.g. the Theory of Planned Behaviour or the Health Belief Model). These models include motivational factors (expectancies and values) that play a role in decision-making in health contexts. In this theoretical paper, it is argued that none of these models makes consequent use of expectancy-value pairs. It is further argued that in order to make these models fruitful for science education and for informed decision-making, models should systematically incorporate knowledge as part of the decision-making process. To fill this gap, this theoretical paper introduces The Integrated Model of Decision-Making in Health Contexts. This model includes three types of knowledge (system health knowledge, action-related health knowledge and effectiveness health knowledge) as influencing factors for motivational factors (perceived health threat, attitude towards health action, attitude towards health outcome and subjective norm) that are formed of expectancy-value pairs and lead to decisions. The model’s potential for health education in science education as well as research implications is discussed.     

日本高校研究生培养方式和管理模式的初步探讨 ——以日本神户大学地球行星科学系为例

本文以日本神户大学地球行星科学系地球动力学实验室为例,对日本高校研究生、留学生的培养方式和学校管理模式进行了初步的探讨,期望能对中国研究生培养及学校的管理提供一些有益的借鉴。     

国外旅游发展社区响应的理论模型述评

中国旅游发展的社区响应领域积累了大量的经验研究,但是这些研究却很少关注旅游发展社区响应的理论分析模型,以至于难以确定已有研究的理论贡献。本文综述了国外文献中有关旅游发展社区响应的9种理论模型,并且比较了这些理论模型的理论内涵、使用情形以及相关的应用研究,对于理解中国国内旅游发展社区响应的经验研究提供了理论分析工具,有助于推进中国旅游学者把他们的研究定位于不同的理论模型脉络下增进该领域的理论贡献。未来研究应关注不同旅游发展阶段、社区利益群体对旅游发展响应的差异及内在原因;关注主-客、客-客、主-主互动对旅游发展响应的差异;吸收情感、权力、政治互信等变量对旅游发展社区响应的影响;关注不同社区群体社会表征对旅游发展的响应机制。