相关论文
2021 - HCINLP
Challenges in Designing Games with a Purpose for Abusive Language Annotation

In this paper we discuss several challenges related to the development of a 3D game, whose goal is to raise awareness on cyberbullying while collecting linguistic annotation on offensive language. The game is meant to be used by teenagers, thus raising a number of issues that need to be tackled during development. For example, the game aesthetics should be appealing for players belonging to this age group, but at the same time all possible solutions should be implemented to meet privacy requirements. Also, the task of linguistic annotation should be possibly hidden, adopting so-called orthogonal game mechanics, without affecting the quality of collected data. While some of these challenges are being tackled in the game development, some others are discussed in this paper but still lack an ultimate solution.

2013 - The Journal of applied psychology
Blaming the organization for abusive supervision: the roles of perceived organizational support and supervisor's organizational embodiment.

Why do employees who experience abusive supervision retaliate against the organization? We apply organizational support theory to propose that employees hold the organization partly responsible for abusive supervision. Depending on the extent to which employees identify the supervisor with the organization (i.e., supervisor's organizational embodiment), we expected abusive supervision to be associated with low perceived organizational support (POS) and consequently with retribution against the organization. Across 3 samples, we found that abusive supervision was associated with decreased POS as moderated by supervisor's organizational embodiment. In turn, reduced POS was related to heightened counterproductive work behavior directed against the organization and lowered in-role and extra-role performance. These findings suggest that employees partly attribute abusive supervision to negative valuation by the organization and, consequently, behave negatively toward and withhold positive contributions to it.

2020 - ArXiv
Towards Ethics by Design in Online Abusive Content Detection

To support safety and inclusion in online communications, significant efforts in NLP research have been put towards addressing the problem of abusive content detection, commonly defined as a supervised classification task. The research effort has spread out across several closely related sub-areas, such as detection of hate speech, toxicity, cyberbullying, etc. There is a pressing need to consolidate the field under a common framework for task formulation, dataset design and performance evaluation. Further, despite current technologies achieving high classification accuracies, several ethical issues have been revealed. We bring ethical issues to forefront and propose a unified framework as a two-step process. First, online content is categorized around personal and identity-related subject matters. Second, severity of abuse is identified through comparative annotation within each category. The novel framework is guided by the Ethics by Design principle and is a step towards building more accurate and trusted models.

2019 - 2019 International Seminar on Research of Information Technology and Intelligent Systems (ISRITI)
Abusive Language Detection on Indonesian Online News Comments

Abusive language is an expression used by a person with insulting delivery of any person's aspect. In the modern era, the use of harsh words is often found on the internet, one of them is in the comment section of online news articles which contains harassment, insult, or a curse. An abusive language detection system is important to prevent the negative effect of such comments. Detecting abusive language in the online comment section is a challenge since abusive languages can be expressed in various words. Moreover, only a few studies have been conducted in Indonesian language. In this paper, we present an Indonesian abusive language detection system by tackling this problem as a classification task and solving it using the following classifiers: Naive Bayes, SVM, and KNN. We also performed feature selection procedure based on Mutual Information value between words. The experimental results show that SVM is the best classifier for detecting the abusive language in news comment with an accuracy score of 90,19% and the use of Mutual Information able to improve the classification accuracy by 1.63%. Mutual Information can increase the accuracy performance of the classifier.

2019 - 2019 International Seminar on Research of Information Technology and Intelligent Systems (ISRITI)
Hate Speech and Abusive Language Classification using fastText

Hate speeches are defined as utterances, writings, actions, or performances that are intended to incite violence or prejudice against a person on the basis of the characteristics of a particular group that he or she is representing, such as race, ethnicity. In this study, we built a hate speech classification model using word representation with continous bag of words (CBOW) and fastText algorithm. This algorithms was chosen, because it is able to achieve a good performance, specially in the case of rare words by making use of character level information. Based on this result, we can see that there is no single, universal variations that outperform other. But in general, models that use pre-trained vectors from Wiki outperform models that do not use pre-trained vectors.

2019 - RANLP
Investigating Multilingual Abusive Language Detection: A Cautionary Tale

Abusive language detection has received much attention in the last years, and recent approaches perform the task in a number of different languages. We investigate which factors have an effect on multilingual settings, focusing on the compatibility of data and annotations. In the current paper, we focus on English and German. Our findings show large differences in performance between the two languages. We find that the best performance is achieved by different classification algorithms. Sampling to address class imbalance issues is detrimental for German and beneficial for English. The only similarity that we find is that neither data set shows clear topics when we compare the results of topic modeling to the gold standard. Based on our findings, we can conclude that a multilingual optimization of classifiers is not possible even in settings where comparable data sets are used.

2010
Developing child protection strategies:A Critical Study of Offenders’ Use of Information Technology for the Sexual Exploitation of Children

The overall aim of this thesis is to critically explore offenders’ use of information technology for the sexual exploitation of children, focusing upon child abusive material and grooming, in relation to the societal response, i.e. legal and technological regulation models. The following aspects are highlighted (i) How can offenders’ use of information technology for child abusive material and grooming be understood in relation to current regulation models? (i) What alternative models for regulation of child abusive material and grooming could be proposed? (iii) What are the implications of applying a critical approach? The motivation for conducting this research has been to contribute, with empirically based research, to the development of effective child protection strategies in relation to child abusive material and grooming. The empirical material used in this thesis consists of court and police records and interviews with offenders. The result shows that offenders’ use of information technology for child abusive material and grooming is more complex and multifaceted than current regulation models have managed to envisage. It has been recognized that the offenders are aware of the illegality in their activities and thus the risk of being observed by law enforcement and have therefore developed different technological and social strategies to be able to continue with their criminal activities. Therefore, this thesis suggests that existing regulation models such as law and the use of technology for filtering should be re-evaluated and that further dimensions such as norms and markets should be considered.

2017 - Proceedings of the International Conference on Web Intelligence
Presenting a labelled dataset for real-time detection of abusive user posts

Social media sites facilitate users in posting their own personal comments online. Most support free format user posting, with close to real-time publishing speeds. However, online posts generated by a public user audience carry the risk of containing inappropriate, potentially abusive content. To detect such content, the straightforward approach is to filter against blacklists of profane terms. However, this lexicon filtering approach is prone to problems around word variations and lack of context. Although recent methods inspired by machine learning have boosted detection accuracies, the lack of gold standard labelled datasets limits the development of this approach. In this work, we present a dataset of user comments, using crowdsourcing for labelling. Since abusive content can be ambiguous and subjective to the individual reader, we propose an aggregated mechanism for assessing different opinions from different labellers. In addition, instead of the typical binary categories of abusive or not, we introduce a third class of 'undecided' to capture the real life scenario of instances that are neither blatantly abusive nor clearly harmless. We have performed preliminary experiments on this dataset using best practice techniques in text classification. Finally, we have evaluated the detection performance of various feature groups, namely syntactic, semantic and context-based features. Results show these features can increase our classifier performance by 18% in detection of abusive content.

2020 - ArXiv
Directions in Abusive Language Training Data: Garbage In, Garbage Out

Data-driven analysis and detection of abusive online content covers many different tasks, phenomena, contexts, and methodologies. This paper systematically reviews abusive language dataset creation and content in conjunction with an open website for cataloguing abusive language data. This collection of knowledge leads to a synthesis providing evidence-based recommendations for practitioners working with this complex and highly diverse data.

2019 - Proceedings of the Third Workshop on Abusive Language Online
Multi-label Hate Speech and Abusive Language Detection in Indonesian Twitter

Hate speech and abusive language spreading on social media need to be detected automatically to avoid conflicts between citizens. Moreover, hate speech has a target, category, and level that also need to be detected to help the authority in prioritizing which hate speech must be addressed immediately. This research discusses multi-label text classification for abusive language and hate speech detection including detecting the target, category, and level of hate speech in Indonesian Twitter using machine learning approaches with Support Vector Machine (SVM), Naive Bayes (NB), and Random Forest Decision Tree (RFDT) classifier and Binary Relevance (BR), Label Power-set (LP), and Classifier Chains (CC) as the data transformation method. We used several kinds of feature extractions which are term frequency, orthography, and lexicon features. Our experiment results show that in general the RFDT classifier using LP as the transformation method gives the best accuracy with fast computational time.

2017 - Proceedings of the 2017 CHI Conference on Human Factors in Computing Systems
The Bag of Communities: Identifying Abusive Behavior Online with Preexisting Internet Data

Since its earliest days, harassment and abuse have plagued the Internet. Recent research has focused on in-domain methods to detect abusive content and faces several challenges, most notably the need to obtain large training corpora. In this paper, we introduce a novel computational approach to address this problem called Bag of Communities (BoC)---a technique that leverages large-scale, preexisting data from other Internet communities. We then apply BoC toward identifying abusive behavior within a major Internet community. Specifically, we compute a post's similarity to 9 other communities from 4chan, Reddit, Voat and MetaFilter. We show that a BoC model can be used on communities "off the shelf" with roughly 75% accuracy---no training examples are needed from the target community. A dynamic BoC model achieves 91.18% accuracy after seeing 100,000 human-moderated posts, and uniformly outperforms in-domain methods. Using this conceptual and empirical work, we argue that the BoC approach may allow communities to deal with a range of common problems, like abusive behavior, faster and with fewer engineering resources.

2010 - Journal of affective disorders
The determinants of family carers' abusive behaviour to people with dementia: results of the CARD study.

BACKGROUND Although dementia and elder abuse prevention are political priorities, there are no evidence-based interventions to reduce abuse by family carers. We have limited understanding of why some family carers, but not others in similar circumstances, behave abusively. We aimed to test our hypothesis, that more anxious dementia carers report more abusive behaviours, and dysfunctional coping strategies and carer burden mediate this relationship. METHOD We interviewed 220 family/friend dementia carers from Essex and London Community Mental Health Teams. We used the revised Modified Conflict Tactics Scale to measure abuse. RESULTS More anxious and depressed carers reported more abuse; this relationship was mediated by using dysfunctional coping strategies and higher burden. Abuse was predicted by: spending more hours caring, experiencing more abusive behaviour from care recipients and higher burden. LIMITATIONS This was a cross-sectional study so we cannot confirm directions of causality. While many carers were willing to report abusive actions, some may not have been and our numbers may be an underestimate. CONCLUSION Anxious and depressed carers are particularly likely to report abusive behaviour when asked. Testing interventions directed at reducing carer anxiety, depression or changing unhelpful coping strategies, and/or reducing care recipient aggression where possible, is a logical and urgent next step.

2020 - Internet Res.
Abusive supervisors and employees who cyberloaf

The purpose of this study is to examine the effects of abusive supervision on cyberloafing behavior, to test the mediating role of PsyCap in this relationship and the extent to which these relationships are moderated by psychological contract breach.,A total of 394 full-time managers across different Indian organizations served as the sample for this study.,The results revealed that the abusive supervision and PsyCap are significantly correlated with cyberloafing, the relationship between abusive supervision and cyberloafing is partially mediated by PsyCap and the impact of abusive supervision and PsyCap on cyberloafing is moderated by psychological contract breach such that the effects of abusive supervision and PsyCap on cyberloafing are stronger when employees perceive high psychological contract breach.,A cross-sectional design and use of self-reported questionnaires are a few limitations of this study.,This is one of the first studies examining cyberloafing in response to abusive supervision and one of the few attempts to examine the effects of abusive supervision on individual resources (PsyCap) in response to workplace mistreatment. This study is also the first to examine these phenomena in the Indian context.

1989 - Issues in mental health nursing
A process of entrapment in and recovery from an abusive relationship.

The purpose of the study was to describe the experience of being abused within the context of a significant relationship and to explain how the nature of the relationship influences the choices a woman makes over time. The sample consisted of 30 women who were currently in or who had already left an abusive relationship. A semistructured, open-ended interview was used to obtain data on the duration, frequency, and severity of abuse sustained by women while in abusive relationships. Data were analyzed using constant comparative and domain analyses. An independent analyst determined that codes developed were reliable. A four-phase process describing entrapment in and recovery from an abusive relationship was generated: binding, enduring, disengaging, and recovering are phases through which a woman passes progressively as the meaning she ascribes to her abusive experience, her interactions with her partner, and her self-change. The process is cumulative and multidimensional.

2003 - Trauma, Violence, & Abuse
Leaving An Abusive Partner

Four facets of leaving an abusive relationship are reviewed: (a) factors related to initially leaving an abusive partner; (b) the process of leaving an abusive relationship; (c) the psychological well-being of survivors after leaving; and (d) the predictors of this well-being. The conceptual and methodological limitations of studies in each of these areas are presented. Consistently found predictors of leaving include both material and psychological factors. Because battered women typically undergo several shifts in their thinking about the abuse before leaving permanently, research on leaving as a process is highlighted. A stress-process framework is used to explain the seemingly paradoxical finding that some women just out of the abusive relationship may have greater psychological difficulties than those who are still in it. For those experiencing the most stress, psychological health can worsen over time. Researchers and practitioners need to pay more attention to the plight of women who have left abusive partners.

2016 - Proceedings of the 25th International Conference on World Wide Web
Abusive Language Detection in Online User Content

Detection of abusive language in user generated online content has become an issue of increasing importance in recent years. Most current commercial methods make use of blacklists and regular expressions, however these measures fall short when contending with more subtle, less ham-fisted examples of hate speech. In this work, we develop a machine learning based method to detect hate speech on online user comments from two domains which outperforms a state-of-the-art deep learning approach. We also develop a corpus of user comments annotated for abusive language, the first of its kind. Finally, we use our detection tool to analyze abusive language over time and in different settings to further enhance our knowledge of this behavior.

2019 - Journal of Business and Psychology
A Social Exchange Perspective of Abusive Supervision and Knowledge Sharing: Investigating the Moderating Effects of Psychological Contract Fulfillment and Self-Enhancement Motive

Given the importance of knowledge management in this competitive environment, the purpose of the present study is to fill the gap in contemporary literatures of knowledge sharing behavior and abusive supervision by observing the main effect, mechanism, and moderators. Based on social exchange perspective, we propose a theoretical model that links abusive supervision to employee knowledge sharing as mediated by leader-member exchange (LMX) with conditional processes. Employing a sample of 184 supervisor-subordinate dyads, we carried out a survey in large listed companies in South Korea. To test our hypotheses, we conducted multiple regression analyses and used bootstrapping procedures. Our results suggest that LMX mediates the abusive supervision and knowledge sharing relationship. Most significantly, findings show that this mediated relationship is contingent on the level of psychological contract fulfillment and self-enhancement motive. One of the most critical implications of our work is that negative influence of hostile behaviors of supervisors on knowledge sharing via LMX may actually be attenuated by perceptions of employees formed both from the organization (i.e., psychological contract fulfillment) and from oneself (i.e., self-enhancement motive). Moreover, it also provides practical insights for both the management practitioner and the organization. Extending from earlier studies, this research enriches our understanding of organizational behavior research by demonstrating an overall complete picture of a moderated-mediation model between abusive supervision and knowledge sharing by uncovering a mediator explaining the mechanism and moderators buffering the negative effect of abusive supervision.

2008 - The Journal of applied psychology
Abusive supervision and subordinates' organization deviance.

The authors developed an integrated model of the relationships among abusive supervision, affective organizational commitment, norms toward organization deviance, and organization deviance and tested the framework in 2 studies: a 2-wave investigation of 243 supervised employees and a cross-sectional study of 247 employees organized into 68 work groups. Path analytic tests of mediated moderation provide support for the prediction that the mediated effect of abusive supervision on organization deviance (through affective commitment) is stronger when employees perceive that their coworkers are more approving of organization deviance (Study 1) and when coworkers perform more acts of organization deviance (Study 2).

2009 - Organizational Behavior and Human Decision Processes
Abusive supervision, intentions to quit, and employees' workplace deviance: A power/dependence analysis

We conducted a two-study examination of relationships between abusive supervision and subordinates' workplace deviance. Consistent with predictions derived from power/dependence theory, the results of a cross-sectional study with employees from three organizations suggest that abusive supervision is more strongly associated with subordinates' organization deviance and supervisor-directed deviance when subordinates' intention to quit is higher. The results also support the prediction that when intention to quit is higher, abusive supervision is more strongly associated with supervisor-directed deviance than with organization-directed deviance. These results were replicated in a second study, a two-wave investigation of people employed in a variety of industries and occupations.

2006 - Personnel Psychology
PROCEDURAL INJUSTICE, VICTIM PRECIPITATION, AND ABUSIVE SUPERVISION

We used data collected from a field survey of 334 supervisor‐subordinate dyads to test a model of the antecedents of abusive supervision. Path analytic tests of moderated mediation provided support for our prediction that supervisors’ depression mediates the relationship between supervisors’ procedural justice and subordinates’ perceptions of their supervisors’ abusiveness and that the mediation framework is stronger when subordinates are higher in negative affectivity. We discuss the study’s implications for theory, research, and practice.

Consequences of Abusive Supervision

B. Tepper
|
Bennett J. Tepper

Abstract:Drawing on justice theory, the author examined the consequences of abusive supervisor behavior. As expected, subordinates who perceived their supervisors were more abusive were more likely to quit their jobs. For subordinates who remained with their jobs, abusive supervision was associated with lower job and life satisfaction, lower normative and affective commitment, and higher continuance commitment, conflict between work and family, and psychological distress. Organizational justice mediated most of these effects, and job mobility moderated some of the deleterious effects of abusive supervision.

参考文献

[1]  L. Walker.  The Battered Woman , 1979 .

[2]  Christina Maslach,et al.  THE TRUTH ABOUT BURNOUT , 1997 .

[3]  R. Folger,et al.  RETALIATION IN THE WORKPLACE: THE ROLES OF DISTRIBUTIVE, PROCEDURAL, AND INTERACTIONAL JUSTICE , 1997 .

[4]  John P. Meyer,et al.  Commitment in the Workplace: Theory, Research, and Application , 1997 .

[5]  T. Judge,et al.  Job and life attitudes of male executives. , 1994, The Journal of applied psychology.

[6]  M. Shepard,et al.  The Abusive Behavior Inventory , 1992 .

[7]  B. Tepper,et al.  Upward Maintenance Tactics in Supervisory Mentoring and Nonmentoring Relationships , 1995 .

[8]  J. V. Maanen,et al.  Toward a theory of organizational socialization , 1977 .

[9]  Robert McMurrian,et al.  Development and validation of work–family conflict and family–work conflict scales. , 1996 .

[10]  J. S. Adams,et al.  Equity Theory Revisited: Comments and Annotated Bibliography , 1976 .

[11]  P. M. Podsakoff,et al.  Self-Reports in Organizational Research: Problems and Prospects , 1986 .

[12]  Michael A. Campion,et al.  Meaning and measurement of turnover: Comparison of alternative measures and recommendations for research. , 1991 .

[13]  Robert A. Baron,et al.  Aggression in the workplace. , 1997 .

[14]  Roderick M. Kramer,et al.  Power and influence in organizations , 1998 .

[15]  R. Bies,et al.  Two Faces of the Powerless: Coping with Tyranny in Organizations , 1998 .

[16]  T. Tyler,et al.  The Social Psychology of Procedural Justice , 1988 .

[17]  A. Bandura.  Social Foundations of Thought and Action: A Social Cognitive Theory , 1985 .

[18]  P. Sweeney,et al.  Distributive and procedural justice as predictors of satisfaction with personal and organizational outcomes. , 1992 .

[19]  L. Radloff.  The CES-D Scale , 1977 .

[20]  R. Larsen,et al.  The Satisfaction with Life Scale , 1985, Journal of personality assessment.

[21]  R. Folger,et al.  Effects of Procedural and Distributive Justice on Reactions to Pay Raise Decisions , 1989 .

[22]  R. Bennett,et al.  A TYPOLOGY OF DEVIANT WORKPLACE BEHAVIORS: A MULTIDIMENSIONAL SCALING STUDY , 1995 .

[23]  R. Bies,et al.  Beyond distrust: "Getting even" and the need for revenge. , 1996 .

[24]  R. H. Moorman,et al.  JUSTICE AS A MEDIATOR OF THE RELATIONSHIP BETWEEN METHODS OF MONITORING AND ORGANIZATIONAL CITIZENSHIP BEHAVIOR , 1993 .

[25]  Russell Cropanzano,et al.  Perceived fairness of employee drug testing as a predictor of employee attitudes and job performance. , 1991, Journal of Applied Psychology.

[26]  K. Rospenda,et al.  Mental health consequences and correlates of reported medical student abuse. , 1992, JAMA.

[27]  D. Sheehan,et al.  A pilot study of medical student 'abuse'. Student perceptions of mistreatment and misconduct in medical school. , 1990, JAMA.

[28]  B. Raymond,et al.  Psychological Abuse among College Women in Dating Relationships , 1989, Perceptual and motor skills.

[29]  D. A. Kenny,et al.  The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. , 1986, Journal of personality and social psychology.

[30]  J. Greenberg,et al.  Organizational Justice: Yesterday, Today, and Tomorrow , 1990 .

[31]  John P. Meyer,et al.  Commitment to organizations and occupations: Extension and test of a three-component conceptualization. , 1993 .

[32]  C. Ross,et al.  Social Causes of Psychological Distress , 1990 .

[33]  S. Gilliland,et al.  Effects of procedural and distributive justice on reactions to a selection system. , 1994 .

[34]  R. Bies.  Interactional justice : communication criteria of fairness , 1986 .

[35]  Chester A. Schriesheim,et al.  Influence tactics used by subordinates: A theoretical and empirical analysis and refinement of the Kipnis, Schmidt, and Wilkinson subscales. , 1990 .

[36]  Norbert K Tanzer,et al.  What people regard as unjust: Types and structures of everyday experiences of injustice. , 1990 .

[37]  Jeffrey H. Greenhaus,et al.  Sources of Conflict Between Work and Family Roles , 1985 .

[38]  R. Tolman.  The Development of a Measure of Psychological Maltreatment of Women by Their Male Partners , 1989, Violence and Victims.

[39]  L. Keashly,et al.  Abusive Behavior in the Workplace: A Preliminary Investigation , 1994, Violence and Victims.

[40]  Robert A. Baron,et al.  Criticism (informal negative feedback) as a source of perceived unfairness in organizations: Effects, mechanisms, and countermeasures. , 1993 .

[41]  Blake E. Ashforth,et al.  Petty Tyranny in Organizations , 1994 .

[42]  G. Leventhal.  What Should Be Done with Equity Theory? New Approaches to the Study of Fairness in Social Relationships. , 1976 .

[43]  C. Molidor,et al.  Physical, sexual and psychological abuse in high school dating relationships: Prevalence rates and self-esteem issues , 1996 .

[44]  David G. Allen,et al.  Integrating justice constructs into the turnover process : a test of a referent cognitions model , 1997 .

[45]  K. Bowers.  Pain, anxiety, and perceived control. , 1968, Journal of consulting and clinical psychology.

引用
Variable Mediated Model
2016
The Relationship between Islamic Piety (Taqwa) and Workplace Deviance with Organizational Justice as a Moderator
2015
Building A Framework For Two Internal Organizational Approaches To Resolving And Preventing Workplace Bullying: Alternative Dispute Resolution And Training
2009
Exploring the impact of workplace cyberbullying on trainee doctors
Medical education
2015
Workplace bullying and intention to leave among schoolteachers in China: The mediating effect of affective commitment
2009
Workplace Abuse: Problems And Consequences
2011
Passive leadership and online interaction: The mediating effects of job autonomy and employee resilience
PACIS
2019
What happens at home does not stay at home: The role of family and romantic partner conflict in destructive leadership.
Stress and Health
2019
Blaming the organization for abusive supervision: the roles of perceived organizational support and supervisor's organizational embodiment.
The Journal of applied psychology
2013
When and How Subordinate Performance Leads to Abusive Supervision: A Social Dominance Perspective
2018
Sales force social exchange in problem resolution situations
2009
Tyrant Leaders as e-Government Service Promoters: The Role of Transparency and Tyranny in the Implementation of e-Government Service
EGOVIS
2017
Support, undermining, and newcomer socialization: : fitting in during the first 90 days.
2013
PROCEDURAL INJUSTICE, VICTIM PRECIPITATION, AND ABUSIVE SUPERVISION
2006
Reviewing employee turnover: focusing on proximal withdrawal states and an expanded criterion.
Psychological bulletin
2012
Understanding Social Loafing: The Role of Justice Perceptions and Exchange Relationships
2003
Does power distance exacerbate or mitigate the effects of abusive supervision? It depends on the outcome.
The Journal of applied psychology
2012
Personality and counterproductive work behaviour: Using conservation of resources theory to narrow the profile of deviant employees
2011
A Social Context Model of Envy and Social Undermining
2012
Does taking the good with the bad make things worse? How abusive supervision and leader–member exchange interact to impact need satisfaction and organizational deviance
2012