Text style can reveal sensitive attributes of the author (e.g. age and race) to the reader, which can, in turn, lead to privacy violations and bias in both human and algorithmic decisions based on text. For example, the style of writing in job applic
ations might reveal protected attributes of the candidate which could lead to bias in hiring decisions, regardless of whether hiring decisions are made algorithmically or by humans. We propose a VAE-based framework that obfuscates stylistic features of human-generated text through style transfer, by automatically re-writing the text itself. Critically, our framework operationalizes the notion of obfuscated style in a flexible way that enables two distinct notions of obfuscated style: (1) a minimal notion that effectively intersects the various styles seen in training, and (2) a maximal notion that seeks to obfuscate by adding stylistic features of all sensitive attributes to text, in effect, computing a union of styles. Our style-obfuscation framework can be used for multiple purposes, however, we demonstrate its effectiveness in improving the fairness of downstream classifiers. We also conduct a comprehensive study on style-pooling's effect on fluency, semantic consistency, and attribute removal from text, in two and three domain style transfer.
Class imbalance is a common challenge in many NLP tasks, and has clear connections to bias, in that bias in training data often leads to higher accuracy for majority groups at the expense of minority groups. However there has traditionally been a dis
connect between research on class-imbalanced learning and mitigating bias, and only recently have the two been looked at through a common lens. In this work we evaluate long-tail learning methods for tweet sentiment and occupation classification, and extend a margin-loss based approach with methods to enforce fairness. We empirically show through controlled experiments that the proposed approaches help mitigate both class imbalance and demographic biases.
Current abusive language detection systems have demonstrated unintended bias towards sensitive features such as nationality or gender. This is a crucial issue, which may harm minorities and underrepresented groups if such systems were integrated in r
eal-world applications. In this paper, we create ad hoc tests through the CheckList tool (Ribeiro et al., 2020) to detect biases within abusive language classifiers for English. We compare the behaviour of two BERT-based models, one trained on a generic hate speech dataset and the other on a dataset for misogyny detection. Our evaluation shows that, although BERT-based classifiers achieve high accuracy levels on a variety of natural language processing tasks, they perform very poorly as regards fairness and bias, in particular on samples involving implicit stereotypes, expressions of hate towards minorities and protected attributes such as race or sexual orientation. We release both the notebooks implemented to extend the Fairness tests and the synthetic datasets usable to evaluate systems bias independently of CheckList.
Data in general encodes human biases by default; being aware of this is a good start, and the research around how to handle it is ongoing. The term bias' is extensively used in various contexts in NLP systems. In our research the focus is specific to
biases such as gender, racism, religion, demographic and other intersectional views on biases that prevail in text processing systems responsible for systematically discriminating specific population, which is not ethical in NLP. These biases exacerbate the lack of equality, diversity and inclusion of specific population while utilizing the NLP applications. The tools and technology at the intermediate level utilize biased data, and transfer or amplify this bias to the downstream applications. However, it is not enough to be colourblind, gender-neutral alone when designing a unbiased technology -- instead, we should take a conscious effort by designing a unified framework to measure and benchmark the bias. In this paper, we recommend six measures and one augment measure based on the observations of the bias in data, annotations, text representations and debiasing techniques.
Administrative responsibility is a real
guarantee of respecting environmental laws. If the
administration aims to protect the environment and
create sustainable development, we should help it and
expand its responsibility. Development should not
take
priority over the environment and resources’ health.
This study aimed to identify the impact of budget participation on each of managerial
performance, organizational commitment and distributive fairness, to investigate the
impact of organizational commitment and distributive fairness on managerial p
erformance,
and to study the impact of distributive fairness on organizational commitment.
To achieve the study objectives, a questionnaire was used and distributed on the
Agricultural Firms in the Syrian Coast. This questionnaire was distributed to72
respondents, but only 50 questionnaires were completed and received.
The results showed that budgetary participation has a positive impact on each of the
manage rial performance, organizational commitment and distributive fairness. And a
positive impact each of the Organizational Commitment and the Distributive Fairnesson
the managerial performance. At last the results appeared that there is no significant impact
for Distributive Fairnesson budget goal commitment.
The policy of punishment in the community is no longer sufficient to achieve the
interest of offenders' deterring and warding off the corruption. Nowadays,
legislation takes care of the offender's situation after the punishment, the method of
trea
ting the effects of the previous sanction and the implemented sentence in his
eligibility to return as a useful member in his society.
The Islamic Law is the law of Justice and Charity so the eternal wound of eligibility
and the permanent derogation of rights are not compatible with its rules aimed at
the conservation of human's interests and the maintenance of his dignity.
We find, in its rules and verdicts, the ways that are capable of achieving the
subsequent care for convicts, their rehabilitation and erasing the effects of the
sanctions on their eligibility.
This research aims at the study of the impact of repentance, applied its pillars and
terms, in the treatment of the effects of offence and punishment on the offender's
fairness and his eligibility to the general mandates as the eligibility of testimony and
judicature.
The scholars of jurisprudence (jurists) agree to accept the repentance of who is
sentenced and the return of his fairness in all sanctions of bounds, retribution and
punitive crimes.
The controversy is in the cases of the bounds of defamation and the perjurer and the
return of his eligibility and mandate if he repents and the eligibility of the bounded
in his bound.
These issues have been addressed in this research by this comparative juridical
study.
