scholarly journals Comparative evaluation of contribution-value plots for machine learning understanding

2021 ◽  
Author(s):  
Dennis Collaris ◽  
Jarke J. van Wijk
Keyword(s):  
Machine Learning ◽  
Visual Analytics ◽  
User Study ◽  
Design Decisions ◽  
Model Interpretation ◽  
Novel Approach ◽  
Definition Of ◽  
The Impact ◽  
The Relationship ◽  
Strict Definition

Abstract The field of explainable artificial intelligence aims to help experts understand complex machine learning models. One key approach is to show the impact of a feature on the model prediction. This helps experts to verify and validate the predictions the model provides. However, many challenges remain open. For example, due to the subjective nature of interpretability, a strict definition of concepts such as the contribution of a feature remains elusive. Different techniques have varying underlying assumptions, which can cause inconsistent and conflicting views. In this work, we introduce local and global contribution-value plots as a novel approach to visualize feature impact on predictions and the relationship with feature value. We discuss design decisions and show an exemplary visual analytics implementation that provides new insights into the model. We conducted a user study and found the visualizations aid model interpretation by increasing correctness and confidence and reducing the time taken to obtain an insight. Graphic Abstract

Sentiment Analysis on UAV-aided Product Comments Based on Machine Learning: From Sentence to Document Level

2020 ◽  
Author(s):  
JINGYANG CAO ◽  
Shirong Yin ◽  
Guoxu Zhang
Keyword(s):  
Machine Learning ◽  
Sentiment Analysis ◽  
Accurate Result ◽  
Supervised Machine Learning ◽  
Hotel Management ◽  
Novel Approach ◽  
Online Comments ◽  
New Perspective ◽  
The Relationship ◽  
Document Level

Abstract This paper presents a novel approach to analyze the sentiment of the product comments from sentence to document level and apply to the customers sentiment analysis on UAV-aided product comments for hotel management. In order to realize the effiffifficient sentiment analysis, a cascaded sentence-to-document sentiment classifification method is investigated. Initially, a supervised machine learning method is applied to explore the sentiment polarity of the sentence (SPS). Afterward, the contribution of the sentence to document (CSD) is calculated by using various statistical algorithms. Lastly, the sentiment polarity of the document (SPD) is determined by the SPS as well as its contribution. Comparative experiments have been established on the basis of hotel online comments, and the outcomes indicate that the proposed method not only raises the effiffifficiency in attaining a more accurate result but also assists immensely in regards to the B5G wireless communication supported by the UAV. The fifindings provide a new perspective that sentence position and its sentiment similarity with document (sentiment condition) dramatically disclose the relationship between sentence and document.

Consumerism

2014 ◽  
pp. 1-17
Author(s):  
Ioanna Papasolomou
Keyword(s):  
Historical Development ◽  
Sustainable Consumption ◽  
Compulsive Buying ◽  
Future Research ◽  
Sustainable Marketing ◽  
Definition Of ◽  
Corporate Social ◽  
The Impact ◽  
The Relationship ◽  
Insight Into

This chapter reveals that the term ‘consumerism' encompasses a number of meanings which create confusion regarding the term. The discussion that follows, attempts to distinguish the different perspectives regarding the term by presenting its historical development and discussing the three definitions that have marked it. It explores the relationship between consumerism, marketing and corporate social responsibility (CSR). The growth of consumerism has led to the over-use of marketing which provided a flourishing ground for compulsive buying and consumption. There is evidence in the literature to suggest that in an era of increasing social problems and environmental challenges, there is a need for CSR and sustainable marketing. In fact, the second definition of consumerism is inextricably linked with CSR and societal marketing. The chapter is conceptual in nature and provides an in-depth review and discussion of some fundamental dimensions associated to consumerism based on the existent literature. The overarching aim is to provide an insight into the evolution and growth of consumerism based on the existent literature related to the topic. The discussion also focuses on exploring the relationship between marketing and consumerism shedding light onto compulsive buying, consumer attitudes and concerns on the micro consumerism issues, sustainable consumption and sustainable marketing. The chapter proceeds to raise some concerns related to the impact of the global economic crisis on consumerism by using as an example Cyprus based on the author's observations and thoughts. The chapter concludes with a list of suggestions to practitioners and directions for future research.

5. Populism and democracy

2017 ◽  
pp. 79-96 ◽  
Author(s):  
Cas Mudde ◽  
Cristóbal Rovira Kaltwasser
Keyword(s):  
Liberal Democracy ◽  
Theoretical Framework ◽  
Complex Relationship ◽  
Clear Definition ◽  
Definition Of ◽  
Electoral Power ◽  
Main Effects ◽  
The Impact ◽  
The Relationship ◽  
Intense Debate

The relationship between populism and democracy has always been a topic of intense debate. Depending on its electoral power and the context in which it arises, populism can work as either a threat to or a corrective for democracy. To better understand this complex relationship, “Populism and democracy” presents a clear definition of (liberal) democracy, which helps to clarify how the latter is positively and negatively affected by populist forces. It then presents an original theoretical framework of the impact of populism on different political regimes, which allows us to distinguish the main effects of populism on the different stages of the process of both democratization and de-democratization.

Microscale Correlations Adoption in Solid Oxide Fuel Cell

2015 ◽  
Vol 12 (4) ◽  
Author(s):  
C. Wang
Keyword(s):  
Particle Size ◽  
Solid Oxide ◽  
Cell Performance ◽  
Oxide Fuel ◽  
Microstructural Parameters ◽  
Novel Approach ◽  
Real Cell ◽  
Microstructure Parameters ◽  
Definition Of ◽  
The Impact

In order to develop a predictive model of real cell performance, firm relationships and assumptions need to be established for the definition of the physical and microstructure parameters for solid oxide fuel cells (SOFCs). This study explores the correlations of microstructure parameters from a microscale level, together with mass transfer and electrochemical reactions inside the electrodes, providing a novel approach to predict SOFC performance numerically. Based on the physical connections and interactions of microstructure parameters, two submodel correlations (i.e., porosity–tortuosity and porosity–particle size ratio) are proposed. Three experiments from literature are selected to facilitate the validation of the numerical results with experimental data. In addition, a sensitivity analysis is performed to investigate the impact of the adopted submodel correlations to the SOFC performance predictions. Normally, the microstructural inputs in the numerical model need to be measured by experiments in order to test the real cell performance. By adopting the two submodel correlations, the simulation can be performed without obtaining relatively hard-to-measure microstructural parameters such as tortuosity and particle size, yet still accurately mimicking a real-world well-structured SOFC operation. By accurately and rationally predicting the microstructural parameters, this study can eventually help to aid the experimental and optimization study of SOFC.

scholarly journals Artificial moral agents: an intercultural perspective

2007 ◽  
Vol 7 ◽  
pp. 129-134
Author(s):  
Michael Nagenborg
Keyword(s):  
Poor Countries ◽  
Moral Agents ◽  
Artificial Moral Agents ◽  
Starting Point ◽  
Different Types ◽  
Good Starting Point ◽  
Cultural Dialogue ◽  
Definition Of ◽  
The Impact ◽  
The Relationship

In this paper I will argue that artificial moral agents (AMAs) are a fitting subject of intercultural information ethics because of the impact they may have on the relationship between information rich and information poor countries. I will give a limiting definition of AMAs first, and discuss two different types of AMAs with different implications from an intercultural perspective. While AMAs following preset rules might raise con-cerns about digital imperialism, AMAs being able to adjust to their user‘s behavior will lead us to the question what makes an AMA ?moral?? I will argue that this question does present a good starting point for an inter-cultural dialogue which might be helpful to overcome the notion of Africa as a mere victim.

1. Fundamental concepts

2013 ◽  
Author(s):  
Kevin Gray ◽  
Susan Francis Gray
Keyword(s):  
Human Rights ◽  
Common Law ◽  
Physical Reality ◽  
England And Wales ◽  
The Core ◽  
The Subject ◽  
Property Concepts ◽  
Definition Of ◽  
The Impact ◽  
The Relationship

Titles in the Core Text series take the reader straight to the heart of the subject, providing focused, concise, and reliable guides for students at all levels. This chapter introduces a number of concepts that are fundamental to an understanding of the contemporary law of land in England and Wales. It discusses: definition of ‘land’ as physical reality; the notion of abstract ‘estates’ in land as the medium of ownership; the relationship between law and equity; the meaning of ‘property’ in land; the impact of human rights on property concepts; the ambivalence of common law perspectives on ‘land’; the statutory organisation of proprietary rights in land; and the underlying policy motivations that drive the contemporary law of land.

Age and gender features of clinical course COVID-19

2021 ◽  
Vol 2021 (2) ◽  
pp. 4-9
Author(s):  
OM Kovalyova ◽  
Keyword(s):  
Clinical Signs ◽  
Scientific Data ◽  
Age And Gender ◽  
Gender And Age ◽  
Different Populations ◽  
And Gender ◽  
Definition Of ◽  
The Impact ◽  
The Relationship ◽  
Gender Features

The article presents the review of modern publications devoted the assessment the impact of age and gender on susceptibility, clinical manifestation and outcome of COVID-19 infection. Statistical data on rate of COVID-19 in relation to age categories and adverse clinical signs of disease in different populations are shown. Old and older ages are the predictors of severe coronavirus course and mortality are emphasized. Gender features of coronavirus infection have been described according to gender cardiology with taken in account the disproportion of hypertension and coronary heart disease in male and female. Due to the scientific research the gender and age peculiarities of immune response to virus infection is considered. The definition of hypothesis “immunosenescence” underlying adverse outcome due to COVID-19 in older patients is taken. Gender peculiarities of COVID-19 are presented by evident scientific data according to the relationship between sex hormone and immune inflammation factors

Complexity and Knowledge in Organisational Context: Concepts, Approaches, Boundaries and Relations

2021 ◽  
Vol 20 (04) ◽  
Author(s):  
Carlos Francisco Bitencourt Jorge ◽  
Marta Lígia Pomim Valentim ◽  
Michael J. D. Sutton ◽  
José Osvaldo de Sordi
Keyword(s):  
Content Analysis ◽  
Web Of Science ◽  
Theoretical Framework ◽  
Organisational Context ◽  
Effective Strategies ◽  
Organisational Knowledge ◽  
Definition Of ◽  
The Impact ◽  
The Relationship

The study sought to understand the relationship among organisations, knowledge and complexity so that managers could develop more effective strategies when working with organisational knowledge and complexity. The theoretical framework of the theme was elaborated from Web of Science and then an analysis of identified approximations, relations and boundaries was carried out. Aiming at greater consistency regarding the approximations and boundaries among the studied themes, we sought complex organisations that contemplated knowledge as a resource. The initial search retrieved 95 articles, and after content analysis was performed, we identified 25 articles considering complex organisations as social organisms and knowledge as a resource. In this sense, difficulties were observed regarding the definition of the concept of complex organisation, as well as regarding the understanding of knowledge as a resource. After the analysis of the 25 articles, eight pointed to some characteristic of complex organisations, and this corpus does now allow to relate and identify the impact of knowledge on complexity, or complexity on knowledge. From these considerations, we discuss ways to manage complexity and knowledge as elements inserted in the organisational context.

scholarly journals Together or Alone: The Price of Privacy in Collaborative Learning

2019 ◽  
Vol 2019 (2) ◽  
pp. 47-65
Author(s):  
Balázs Pejó ◽  
Qiang Tang ◽  
Gergely Biczók
Keyword(s):  
Machine Learning ◽  
Nash Equilibrium ◽  
Privacy Protection ◽  
Machine Learning Algorithms ◽  
Training Dataset ◽  
Privacy Concerns ◽  
Novel Approach ◽  
Player Game ◽  
The Individual ◽  
The Impact

Abstract Machine learning algorithms have reached mainstream status and are widely deployed in many applications. The accuracy of such algorithms depends significantly on the size of the underlying training dataset; in reality a small or medium sized organization often does not have the necessary data to train a reasonably accurate model. For such organizations, a realistic solution is to train their machine learning models based on their joint dataset (which is a union of the individual ones). Unfortunately, privacy concerns prevent them from straightforwardly doing so. While a number of privacy-preserving solutions exist for collaborating organizations to securely aggregate the parameters in the process of training the models, we are not aware of any work that provides a rational framework for the participants to precisely balance the privacy loss and accuracy gain in their collaboration. In this paper, by focusing on a two-player setting, we model the collaborative training process as a two-player game where each player aims to achieve higher accuracy while preserving the privacy of its own dataset. We introduce the notion of Price of Privacy, a novel approach for measuring the impact of privacy protection on the accuracy in the proposed framework. Furthermore, we develop a game-theoretical model for different player types, and then either find or prove the existence of a Nash Equilibrium with regard to the strength of privacy protection for each player. Using recommendation systems as our main use case, we demonstrate how two players can make practical use of the proposed theoretical framework, including setting up the parameters and approximating the non-trivial Nash Equilibrium.

scholarly journals Training confounder-free deep learning models for medical applications

2020 ◽  
Vol 11 (1) ◽  
Author(s):  
Qingyu Zhao ◽  
Ehsan Adeli ◽  
Kilian M. Pohl
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Bone Age ◽  
Magnetic Resonance Images ◽  
Data Sets ◽  
Large Set ◽  
Learning Models ◽  
End To End ◽  
The Impact ◽  
The Relationship

AbstractThe presence of confounding effects (or biases) is one of the most critical challenges in using deep learning to advance discovery in medical imaging studies. Confounders affect the relationship between input data (e.g., brain MRIs) and output variables (e.g., diagnosis). Improper modeling of those relationships often results in spurious and biased associations. Traditional machine learning and statistical models minimize the impact of confounders by, for example, matching data sets, stratifying data, or residualizing imaging measurements. Alternative strategies are needed for state-of-the-art deep learning models that use end-to-end training to automatically extract informative features from large set of images. In this article, we introduce an end-to-end approach for deriving features invariant to confounding factors while accounting for intrinsic correlations between the confounder(s) and prediction outcome. The method does so by exploiting concepts from traditional statistical methods and recent fair machine learning schemes. We evaluate the method on predicting the diagnosis of HIV solely from Magnetic Resonance Images (MRIs), identifying morphological sex differences in adolescence from those of the National Consortium on Alcohol and Neurodevelopment in Adolescence (NCANDA), and determining the bone age from X-ray images of children. The results show that our method can accurately predict while reducing biases associated with confounders. The code is available at https://github.com/qingyuzhao/br-net.

Sign in / Sign up

Export Citation Format

Share Document