scholarly journals An Adaptive Heterogeneous Online Learning Ensemble Classifier for Nonstationary Environments

2021 ◽  
Vol 2021 ◽  
pp. 1-11
Author(s):  
Tinofirei Museba ◽  
Fulufhelo Nelwamondo ◽  
Khmaies Ouahada
Keyword(s):  
Concept Drift ◽  
Predictive Performance ◽  
Dynamic Environments ◽  
Ensemble Classifiers ◽  
Data Generation ◽  
Ensemble Selection ◽  
Different Types ◽  
Real World Datasets ◽  
The Impact ◽  
Dynamic Ensemble Selection

In recent years, the prevalence of technological advances has led to an enormous and ever-increasing amount of data that are now commonly available in a streaming fashion. In such nonstationary environments, the underlying process generating the data stream is characterized by an intrinsic nonstationary or evolving or drifting phenomenon known as concept drift. Given the increasingly common applications whose data generation mechanisms are susceptible to change, the need for effective and efficient algorithms for learning from and adapting to evolving or drifting environments can hardly be overstated. In dynamic environments associated with concept drift, learning models are frequently updated to adapt to changes in the underlying probability distribution of the data. A lot of work in the area of learning in nonstationary environments focuses on updating the learning predictive model to optimize recovery from concept drift and convergence to new concepts by adjusting parameters and discarding poorly performing models while little effort has been dedicated to investigate what type of learning model is suitable at any given time for different types of concept drift. In this paper, we investigate the impact of heterogeneous online ensemble learning based on online model selection for predictive modeling in dynamic environments. We propose a novel heterogeneous ensemble approach based on online dynamic ensemble selection that accurately interchanges between different types of base models in an ensemble to enhance its predictive performance in nonstationary environments. The approach is known as Heterogeneous Dynamic Ensemble Selection based on Accuracy and Diversity (HDES-AD) and makes use of models generated by different base learners to increase diversity to circumvent problems associated with existing dynamic ensemble classifiers that may experience loss of diversity due to the exclusion of base learners generated by different base algorithms. The algorithm is evaluated on artificial and real-world datasets with well-known online homogeneous online ensemble approaches such as DDD, AFWE, and OAUE. The results show that HDES-AD performed significantly better than the other three homogeneous online ensemble approaches in nonstationary environments.

scholarly journals Evolutionary Social Poisson Factorizationfor Temporal Recommendation

2021 ◽  
Vol 14 (1) ◽  
Author(s):  
ChunYan Yin ◽  
YongHeng Chen ◽  
Wanli Zuo
Keyword(s):  
Recommendation Systems ◽  
Predictive Performance ◽  
User Preferences ◽  
Social Model ◽  
User Preference ◽  
Latent Factors ◽  
Factorization Model ◽  
Real World Datasets ◽  
Item Attributes ◽  
The Impact

AbstractPreference-based recommendation systems analyze user-item interactions to reveal latent factors that explain our latent preferences for items and form personalized recommendations based on the behavior of others with similar tastes. Most of the works in the recommendation systems literature have been developed under the assumption that user preference is a static pattern, although user preferences and item attributes may be changed through time. To achieve this goal, we develop an Evolutionary Social Poisson Factorization (EPF$$\_$$ _ Social) model, a new Bayesian factorization model that can effectively model the smoothly drifting latent factors using Conjugate Gamma–Markov chains. Otherwise, EPF$$\_$$ _ Social can obtain the impact of friends on social network for user’ latent preferences. We studied our models with two large real-world datasets, and demonstrated that our model gives better predictive performance than state-of-the-art static factorization models.

scholarly journals A Decision-Based Dynamic Ensemble Selection Method for Concept Drift

2019 ◽  
Author(s):  
Regis Antonio S. Albuquerque ◽  
Albert F. Josua Costa ◽  
Eulanda Miranda dos Santos ◽  
Robert Sabourin ◽  
Rafael Giusti
Keyword(s):  
Concept Drift ◽  
Selection Method ◽  
Ensemble Selection ◽  
Dynamic Ensemble Selection

Investigating the impact of selection criteria in dynamic ensemble selection methods

2018 ◽  
Vol 106 ◽  
pp. 141-153 ◽  
Author(s):  
Jose Augusto S. Lustosa Filho ◽  
Anne M.P. Canuto ◽  
Regivan Hugo Nunes Santiago
Keyword(s):  
Selection Criteria ◽  
Selection Methods ◽  
Ensemble Selection ◽  
The Impact ◽  
Dynamic Ensemble Selection

scholarly journals ADES: A New Ensemble Diversity-Based Approach for Handling Concept Drift

2021 ◽  
Vol 2021 ◽  
pp. 1-17
Author(s):  
Tinofirei Museba ◽  
Fulufhelo Nelwamondo ◽  
Khmaies Ouahada
Keyword(s):  
Machine Learning ◽  
Real World ◽  
Data Streams ◽  
Predictive Models ◽  
Concept Drift ◽  
Dynamic Environments ◽  
Real World Data ◽  
World Data ◽  
Different Types ◽  
Concept Drifts

Beyond applying machine learning predictive models to static tasks, a significant corpus of research exists that applies machine learning predictive models to streaming environments that incur concept drift. With the prevalence of streaming real-world applications that are associated with changes in the underlying data distribution, the need for applications that are capable of adapting to evolving and time-varying dynamic environments can be hardly overstated. Dynamic environments are nonstationary and change with time and the target variables to be predicted by the learning algorithm and often evolve with time, a phenomenon known as concept drift. Most work in handling concept drift focuses on updating the prediction model so that it can recover from concept drift while little effort has been dedicated to the formulation of a learning system that is capable of learning different types of drifting concepts at any time with minimum overheads. This work proposes a novel and evolving data stream classifier called Adaptive Diversified Ensemble Selection Classifier (ADES) that significantly optimizes adaptation to different types of concept drifts at any time and improves convergence to new concepts by exploiting different amounts of ensemble diversity. The ADES algorithm generates diverse base classifiers, thereby optimizing the margin distribution to exploit ensemble diversity to formulate an ensemble classifier that generalizes well to unseen instances and provides fast recovery from different types of concept drift. Empirical experiments conducted on both artificial and real-world data streams demonstrate that ADES can adapt to different types of drifts at any given time. The prediction performance of ADES is compared to three other ensemble classifiers designed to handle concept drift using both artificial and real-world data streams. The comparative evaluation performed demonstrated the ability of ADES to handle different types of concept drifts. The experimental results, including statistical test results, indicate comparable performances with other algorithms designed to handle concept drift and prove their significance and effectiveness.

Quota Women Are Threatening to Men

2017 ◽  
Vol 76 (3) ◽  
pp. 107-116 ◽  
Author(s):  
Klea Faniko ◽  
Till Burckhardt ◽  
Oriane Sarrasin ◽  
Fabio Lorenzi-Cioldi ◽  
Siri Øyslebø Sørensen ◽  
...  
Keyword(s):  
Decision Making ◽  
Organizational Performance ◽  
Gender Stereotypes ◽  
Perceived Threat ◽  
Preferential Treatment ◽  
Public Sector Employees ◽  
Different Types ◽  
Low Levels ◽  
Women In Decision Making ◽  
The Impact

Abstract. Two studies carried out among Albanian public-sector employees examined the impact of different types of affirmative action policies (AAPs) on (counter)stereotypical perceptions of women in decision-making positions. Study 1 (N = 178) revealed that participants – especially women – perceived women in decision-making positions as more masculine (i.e., agentic) than feminine (i.e., communal). Study 2 (N = 239) showed that different types of AA had different effects on the attribution of gender stereotypes to AAP beneficiaries: Women benefiting from a quota policy were perceived as being more communal than agentic, while those benefiting from weak preferential treatment were perceived as being more agentic than communal. Furthermore, we examined how the belief that AAPs threaten men’s access to decision-making positions influenced the attribution of these traits to AAP beneficiaries. The results showed that men who reported high levels of perceived threat, as compared to men who reported low levels of perceived threat, attributed more communal than agentic traits to the beneficiaries of quotas. These findings suggest that AAPs may have created a backlash against its beneficiaries by emphasizing gender-stereotypical or counterstereotypical traits. Thus, the framing of AAPs, for instance, as a matter of enhancing organizational performance, in the process of policy making and implementation, may be a crucial tool to countering potential backlash.

Situational Breakdowns

2019 ◽  
Author(s):  
Anne Nassauer
Keyword(s):  
United States ◽  
Large Scale ◽  
Contextual Factors ◽  
The United States ◽  
Social Outcomes ◽  
Dynamic Processes ◽  
Collective Decisions ◽  
Different Types ◽  
The Impact

This book provides an account of how and why routine interactions break down and how such situational breakdowns lead to protest violence and other types of surprising social outcomes. It takes a close-up look at the dynamic processes of how situations unfold and compares their role to that of motivations, strategies, and other contextual factors. The book discusses factors that can draw us into violent situations and describes how and why we make uncommon individual and collective decisions. Covering different types of surprise outcomes from protest marches and uprisings turning violent to robbers failing to rob a store at gunpoint, it shows how unfolding situations can override our motivations and strategies and how emotions and culture, as well as rational thinking, still play a part in these events. The first chapters study protest violence in Germany and the United States from 1960 until 2010, taking a detailed look at what happens between the start of a protest and the eruption of violence or its peaceful conclusion. They compare the impact of such dynamics to the role of police strategies and culture, protesters’ claims and violent motivations, the black bloc and agents provocateurs. The analysis shows how violence is triggered, what determines its intensity, and which measures can avoid its outbreak. The book explores whether we find similar situational patterns leading to surprising outcomes in other types of small- and large-scale events: uprisings turning violent, such as Ferguson in 2014 and Baltimore in 2015, and failed armed store robberies.

scholarly journals A global analysis of the impact of COVID-19 stay-at-home restrictions on crime

2021 ◽  
Author(s):  
Amy E. Nivette ◽  
Renee Zahnow ◽  
Raul Aguilar ◽  
Andri Ahven ◽  
Shai Amram ◽  
...  
Keyword(s):  
Public Space ◽  
Global Analysis ◽  
Sudden Change ◽  
Interrupted Time Series ◽  
Urban Crime ◽  
Substantial Variation ◽  
Different Types ◽  
Time Series Analyses ◽  
The Impact ◽  
At Home

AbstractThe stay-at-home restrictions to control the spread of COVID-19 led to unparalleled sudden change in daily life, but it is unclear how they affected urban crime globally. We collected data on daily counts of crime in 27 cities across 23 countries in the Americas, Europe, the Middle East and Asia. We conducted interrupted time series analyses to assess the impact of stay-at-home restrictions on different types of crime in each city. Our findings show that the stay-at-home policies were associated with a considerable drop in urban crime, but with substantial variation across cities and types of crime. Meta-regression results showed that more stringent restrictions over movement in public space were predictive of larger declines in crime.

Comprehensive Effects of Short-Term Rental Platforms Across Hotel Types in U.S. and International Destinations

2021 ◽  
pp. 193896552110335
Author(s):  
John W. O’Neill ◽  
Jihwan Yeon
Keyword(s):  
Service Providers ◽  
Comprehensive Analysis ◽  
Competitive Market ◽  
Substitution Effects ◽  
Short Term ◽  
Hotel Performance ◽  
The World ◽  
Different Types ◽  
Location Type ◽  
The Impact

In recent years, short-term rental platforms in the lodging sector, including Airbnb, VRBO, and HomeAway, have received extensive attention and emerged as potentially alternative suppliers of services traditionally provided by established commercial accommodation providers, that is, hotels. Short-term rentals have dramatically increased the available supply of rooms for visitors to multiple international destinations, potentially siphoning demand away from hotels to short-term rental businesses. In a competitive market, an increase in supply with constant demand would negatively influence incumbent service providers. In this article, we examine the substitution effects of short-term rental supply on hotel performance in different cities around the world. Specifically, we comprehensively investigate the substitution effects of short-term rental supply on hotel performance based on hotel class, location type, and region. Furthermore, we segment the short-term rental supply based on its types of accommodations, that is, shared rooms, private rooms, and entire homes, and both examine and quantify the differential effects of these types of short-term rentals on different types of hotels. This study offers a comprehensive analysis regarding the impact of multiple short-term rental platforms on hotel performance and offers both conceptual and practical insights regarding the nature and extent of the effects that were identified.

Double, triple or quadruple hits? Exploring the impact of cybercrime on victims in the Netherlands

2021 ◽  
pp. 026975802110106
Author(s):  
Raoul Notté ◽  
E.R. Leukfeldt ◽  
Marijke Malsch
Keyword(s):  
Public Opinion ◽  
Literature Review ◽  
The Netherlands ◽  
Public Discourse ◽  
Moral Judgments ◽  
Victim Blaming ◽  
High Prevalence ◽  
Different Types ◽  
The Impact ◽  
Insight Into

This article explores the impact of online crime victimisation. A literature review and 41 interviews – 19 with victims and 22 with experts – were carried out to gain insight into this. The interviews show that most impacts of online offences correspond to the impacts of traditional offline offences. There are also differences with offline crime victimisation. Several forms of impact seem to be specific to victims of online crime: the substantial scale and visibility of victimhood, victimisation that does not stop in time, the interwovenness of online and offline, and victim blaming. Victims suffer from double, triple or even quadruple hits; it is the accumulation of different types of impact, enforced by the limitlessness in time and space, which makes online crime victimisation so extremely invasive. Furthermore, the characteristics of online crime victimisation greatly complicate the fight against and prevention of online crime. Finally, the high prevalence of cybercrime victimisation combined with the severe impact of these crimes seems contradictory with public opinion – and associated moral judgments – on victims. Further research into the dominant public discourse on victimisation and how this affects the functioning of the police and victim support would be valuable.

scholarly journals Can Architectural Surfaces Capture Atmospheric Particulate Matter Like Trees? A Design Strategy to Mimic Leaf Traits

2021 ◽  
Vol 13 (14) ◽  
pp. 7637
Author(s):  
Taekyoung Lee ◽  
Jieun Cha ◽  
Sohyun Sung
Keyword(s):  
Leaf Traits ◽  
Design Strategy ◽  
Atmospheric Particulate Matter ◽  
Porous Surface ◽  
Base Surface ◽  
Different Types ◽  
Pm2.5 And Pm10 ◽  
Micro Morphology ◽  
The Impact ◽  
Cluster Pattern

Trees’ ability to capture atmospheric Particular Matter (PM) is related to morphological traits (shape, size, and micro-morphology) of the leaves. The objectives of this study were (1) to find out whether cluster pattern of the leaves is also a parameter that affects trees’ PM capturing performance and (2) to apply the cluster patterns of the leaves on architectural surfaces to confirm its impact on PM capturing performance. Two series of chamber experiments were designed to observe the impact of cluster patterns on PM capturing performance whilst other influential variables were controlled. First, we exposed synthetic leaf structures of different cluster patterns (a large and sparsely arranged cluster pattern and a small and densely arranged cluster pattern) to artificially generated PM in a chamber for 60 min and recorded the changing levels of PM2.5 and PM10 every minute. The results confirmed that the small and densely arranged cluster pattern has more significant effect on reducing PM2.5 and PM10 than the large and sparsely arranged cluster pattern. Secondly, we created three different types of architectural surfaces mimicking the cluster patterns of the leaves: a base surface, a folded surface, and a folded and porous surface. The surfaces were also exposed to artificially generated PM in the chamber and the levels of PM2.5 and PM10 were recorded. The results confirmed that the folded and porous surface has a more significant effect on reducing PM2.5 and PM10 than other surfaces. The study has confirmed that the PM capturing performance of architectural surfaces can be improved by mimicking cluster pattern of the leaves.

Sign in / Sign up

Export Citation Format

Share Document