Temporal and Contextual Evaluation of Background Knowledge Discovery for Short Text Classification

Author(s):  
Isak Taksa ◽  
Sarah Zelikovitz ◽  
Amanda Spink

Background Knowledge (BK) plays an essential role in machine learning for short-text and non-topical classification. In this paper the authors present and evaluate two Information Retrieval techniques used to assemble four sets of BK in the past seven years. These sets were applied to classify a commercial corpus of search queries by the apparent age of the user. Temporal and contextual evaluations were used to examine results of various classification scenarios providing insight into choice, significance and range of tuning parameters. The evaluations also demonstrated the impact of the dynamic Web collection on classification results, and the advantages of Automatic Query Expansion (AQE) vs. basic search. The authors discuss other results of this research and its implications on the advancement of short text classification.

2021 ◽  
pp. 63-70
Author(s):  
Inna Shevchenko ◽  
Illia Dmytriiev ◽  
Oksana Dmytriieva

Problem. The global automotive industry has already had an experience of recovery from the global financial crisis of 2008, but the pandemic crisis of 2020 is quite different in nature and pattern of progress: in recent history it has had no analogues and it will be premature to state its completion. Therefore, it is important to determine the impact of the pandemic on the production and sale of cars in order to overcome the negative consequences. To address this issue, the article identifies the sensitivity of this subsector of mechanical engineering to destructive changes in the environment; an analysis of changes in the volume of production and sales of cars by countries of the world over the past period has been made. Goal. The aim of the work is to determine the destructive consequences and trends of the COVID-19 pandemic impact on the global automotive industry, namely the production and sale of cars. Methodology. To determine the impact of the COVID-19 pandemic, a vertical and horizontal analysis of car production and sales in the world has been conducted. Results. The results of the analysis allowed the authors to group the countries of the world by the destructive effects of the pandemic crisis of 2020 for the automotive industry. Originality. The carried out classification of countries by the destructive effects of the COVID-19 pandemic provided an opportunity to gain insight into its impact on the automotive industry, in particular on the production and sale of cars. Practical value. The obtained results can be recommended to identify further ways to overcome the negative effects of the COVID-19 pandemic in the automotive industry.


PEDIATRICS ◽  
1993 ◽  
Vol 92 (4) ◽  
pp. 636-636
Author(s):  
ANDREW F. SHORR

Futterman et al provide interesting insight into the spread of the human immunodeficiency virus (HIV) among adolescents in New York City and into the impact the acquired immunodeficiency syndrome (AIDS) has had on this population. In their conclusion, however, they misrepresent the data regarding HIV and AIDS among youth. More specifically, they write, "Reported AIDS cases among adolescents increased by over 77% over the past two years. . ." By using cumulative percent data for AIDS cases, they distort the true picture. The actual data reveal that the number of AIDS cases in this population has dropped during the past year.


Author(s):  
Nicholas Patterson ◽  
Dhananjay Thiruvady ◽  
Guy Wood-Bradley

This chapter explores the impact that artificial intelligence will make in the education sector and how it will transform the way in which both educators and students interact in the classrooms of the future. The chapter begins with an introduction into the digital education space as well as where artificial intelligence currently sits. When it comes to the transformation of education, the authors explore the educator and student perspectives to ensure both sides requirements are portrayed. Both stakeholders have an equally large learning curve and require more digital literacy than in the past; however, the transformation that artificial intelligence will bring to the table is that educators and students will likely not be trapped with repetitive tasks and can focus on being creative, learning, and teaching. The three elements they explore in this chapter will give insight into work previously completed, research being conducted, and future insights and observations.


Author(s):  
Deboshree Banerji ◽  
Rituparna Das

The economic strength of a country depicts the international standing of a nation and also reflects the significance of the country in moulding the trends of the global economy equally. The Brazilian economy, like many developing economies, has many facets that have developed and matured with time. The Brazilian securities market has undergone much change over the past decade. The reforms that started with the implementation of the “Plano Real” have accelerated the Brazilian market and economy exponentially, thus making the economy one of the major investment destinations, with some calling it the “next superpower.” The fact that the Brazilian economy is a commodities-dominated economy has led the authors to probe into the various nuances related to the securities markets of Brazil, leading to this chapter through which we get a glimpse into the reforms in the securities market and the effect it has on the country as well as the world. The chapter meanders through the development of the Brazilian economy and provides insight into the heart of the Brazilian economy, thereby discussing the effect of the reforms on the economy of the country, how the same strikes the global economy, and the lessons that the country can learn from the other BRICS counterparts, through which it can consolidate its position.


Author(s):  
Brittany Morison

Over the past few decades technology has become ubiquitous, with technology companies gaining increasing insight into the lives of individuals. This paper explores how technology companies use these insights to influence the ability to exercise free and independent decision-making. Through a critical analysis of social nudging, I establish the subtle but significant ways in which individuals can be susceptible to manipulation. Through this lens, I highlight some notable examples of how big tech companies have manipulated individual decision-making and the impact this may have on our democracy. 


2005 ◽  
pp. 185-201
Author(s):  
Jovana Zelic

This paper deals with the impact privatization process has on the performance of Serbian enterprises. Since the most frequently quoted obstacle for good economic performance in the past is the delay in privatization process and enterprise restructuring, the present analysis might help in obtaining a better insight into the problems preventing the acceleration of growth rate in Serbia. Hence, the present work evaluates the relationship between different methods of privatization leading to different ownership structures and the performance of enterprises in Serbia.


Author(s):  
Isak Taksa ◽  
Sarah Zelikovitz ◽  
Amanda Spink

Background knowledge has been actively investigated as a possible means to improve performance of machine learning algorithms. Research has shown that background knowledge plays an especially critical role in three atypical text categorization tasks: short-text classification, limited labeled data, and non-topical classification. This chapter explores the use of machine learning for non-hierarchical classification of search queries, and presents an approach to background knowledge discovery by using information retrieval techniques. Two different sets of background knowledge that were obtained from the World Wide Web, one in 2006 and one in 2009, are used with the proposed approach to classify a commercial corpus of web query data by the age of the user. In the process, various classification scenarios are generated and executed, providing insight into choice, significance and range of tuning parameters, and exploring impact of the dynamic web on classification results.


2011 ◽  
Vol 3 (2) ◽  
pp. 58-73 ◽  
Author(s):  
Ina Markova

This article focuses on the impact of images on reconstructions of the past. In order to analyze the function of images in history textbooks, image-discourse analysis is applied to a case study of Austrian postwar memory. The analysis of recent Austrian history textbooks provides insight into strategies by which notions of Austria as both "victim" and "perpetrator" of the National Socialist regime are held in balance. The article also focuses on the intentional framing of iconic depictions of two central Austrian sites of memory, Heroes' Square (Heldenplatz) and the State Treaty (Staatsvertrag).


1997 ◽  
Vol 161 ◽  
pp. 157-164 ◽  
Author(s):  
Christopher F. Chyba

AbstractOver the past quarter century, our understanding of the impact history of the Solar System has greatly improved. This chapter considers how this increased knowledge affects our evaluation of the chances for other intelligent communicative civilizations in the Galaxy. In addition, the role of impacts is examined for insight into the long-standing debate over whether the evolution of technical intelligence is contingent upon extremely special circumstances, or might instead be a likely outcome of many different but parallel evolutionary pathways.


Sign in / Sign up

Export Citation Format

Share Document