1,721,051 research outputs found
A Framework for Biometric Recognition in Online Content Delivery Platforms
No full textIn this paper, we introduce a modular framework that aims to empower online platforms with biometric-related capabilities, minimizing the user's interaction cost. First, we describe core concepts and architectural aspects characterizing the proposed framework. Then, as a use case, we integrate it in an e-learning platform to provide biometric recognition at the time of login and continuous identity verification in well-rounded areas of the platform
Controlling user access to cloud-connected mobile applications by means of biometrics
No full textCloud-connected mobile applications are becoming a popular solution for ubiquitous access to online services, such as cloud data storage platforms. The adoption of such applications has security and privacy implications that are making individuals hesitant to migrate sensitive data to the cloud; thus, new secure authentication protocols are needed. In this article, we propose a continuous-authentication approach integrating physical (face) and behavioral (touch and hand movements) biometrics to control user access to cloud-based mobile services, going beyond one-time login. Experimental results show the security–usability tradeoff achieved by our approach
Countering Bias in Personalized Rankings From Data Engineering to Algorithm Development
No full textThis tutorial presents recent advances on the assessment and mitigation of data and algorithmic bias in personalized rankings. We first introduce fundamental concepts and definitions associated with bias issues, covering the state of the art and describing real-world examples of how bias can impact ranking algorithms from several perspectives (e.g., ethics and system's objectives). Then, we continue with a systematic presentation of techniques to uncover, assess, and mitigate biases along the personalized ranking design process, with a focus on the role of data engineering in each step of the pipeline. Handson parts provide attendees with concrete implementations of bias mitigation algorithms, in addition to processes and guidelines on how data is organized and manipulated by these algorithms. The tutorial leverages open-source tools and public datasets, engaging attendees in designing bias countermeasures and in articulating impacts on stakeholders. We finally showcase open issues and future directions in this vibrant and rapidly evolving research area (Website: https://biasinrecsys.github.ioticde2021/).ML4E
Evaluation framework for context-aware speaker recognition in noisy smart living environments
Get PDFThe integration of voice control into connected devices is expected to improve the efficiency and comfort of our daily lives. However, the underlying biometric systems often impose constraints on the individual or the environment during interaction (e.g., quiet surroundings). Such constraints have to be surmounted in order to seamlessly recognize individuals. In this paper, we propose an evaluation framework for speaker recognition in noisy smart living environments. To this end, we designed a taxonomy of sounds (e.g., home-related, mechanical) that characterize representative indoor and outdoor environments where speaker recognition is adopted. Then, we devised an approach for off-line simulation of challenging noisy conditions in vocal audios originally collected under controlled environments, by leveraging our taxonomy. Our approach adds a (combination of) sound(s) belonging to the target environment into the current vocal example. Experiments on a large-scale public dataset and two state-of-the-art speaker recognition models show that adding certain background sounds to clean vocal audio leads to a substantial deterioration of recognition performance. In several noisy settings, our findings reveal that a speaker recognition model might end up to make unreliable decisions. Our framework is intended to help system designers evaluate performance deterioration and develop speaker recognition models more robust to smart living environments
Interplay between upsampling and regularization for provider fairness in recommender systems
Get PDFConsidering the impact of recommendations on item providers is one of the duties of multi-sided recommender systems. Item providers are key stakeholders in online platforms, and their earnings and plans are influenced by the exposure their items receive in recommended lists. Prior work showed that certain minority groups of providers, characterized by a common sensitive attribute (e.g., gender or race), are being disproportionately affected by indirect and unintentional discrimination. Our study in this paper handles a situation where (i) the same provider is associated with multiple items of a list suggested to a user, (ii) an item is created by more than one provider jointly, and (iii) predicted user-item relevance scores are biasedly estimated for items of provider groups. Under this scenario, we assess disparities in relevance, visibility, and exposure, by simulating diverse representations of the minority group in the catalog and the interactions. Based on emerged unfair outcomes, we devise a treatment that combines observation upsampling and loss regularization, while learning user-item relevance scores. Experiments on real-world data demonstrate that our treatment leads to lower disparate relevance. The resulting recommended lists show fairer visibility and exposure, higher minority item coverage, and negligible loss in recommendation utility.ML4E
Learning analytics tool for usability assessment in Moodle environments
Get PDFEvaluating user experience is a challenging task, particularly in e-learning. Existing e-learning systems are limited in their ability of being evaluated based on the user interfaces because current evaluation approaches are usually expensive in time and organization and require active users’ participation. Moreover, a usability assessment is needed whenever a new version of the “user interface tailored for a given type of device (e.g. laptop, tablet, smartphone) is developed”. In this paper, we get around the problem leveraging on the increasing adoption of analytics tools in e-learning and on logs transparently tracked by e-learning platforms. We introduce an automated analytics approach aiming at assessing the usability of both desktop and mobile user interfaces of a Learning Management System through specific native indicators (e.g. efficiency and satisfaction). They are defined as comparable scores and calculated automatically based on the tracked log files. In order to put the proposed approach into practice, we implemented it on the Moodle LMS. Our contribution promises to reduce both time and cost for usability assessment of user interfaces in e-learning, while ensuring adaptability to different devices and systems
Leveraging continuous multi-modal authentication for access control in mobile cloud environments
No full textMobile cloud computing integrates cloud computing into mobile environments, allowing users to use data, infrastructure, platforms, and applications on the cloud from their mobile devices. However, accessing and exploiting cloud-based resources and services is associated with numerous security implications (e.g. authentication and authorization) which represent the major barriers making individuals and organizations hesitant to migrate data or processing to the cloud. Using biometric techniques is increasingly emerging to secure such users' assets. In this paper, we propose a bi-modal continuous authentication approach integrating face and touch biometrics into mobile cloud environments, going beyond traditional one-off authentication. The system reacts to sliding windows of recent users actions to dynamically update the trust in genuineness for the current user. For each biometric trait, it calculates the similarity scores resulting from the comparison between probes and templates, then they are fused together. If the fusion score is above a given threshold, the system rewards the current user; otherwise, they are penalized. In case the trust in the current user drops below a predefined threshold, the system raises an alarm. Experimental results indicate the advantage offered by our approach on performance of continuous authentication systems, providing a good security-usability tradeoff to mobile cloud environments
Improving Fairness in Speaker Recognition
No full textThe human voice conveys unique characteristics of an individual, making voice biometrics a key technology for verifying identities in various industries. Despite the impressive progress of speaker recognition systems in terms of accuracy, a number of ethical and legal concerns has been raised, specifically relating to the fairness of such systems. In this paper, we aim to explore the disparity in performance achieved by state-of-the-art deep speaker recognition systems, when different groups of individuals characterized by a common sensitive attribute (e.g., gender) are considered. In order to mitigate the unfairness we uncovered by means of an exploratory study, we investigate whether balancing the representation of the different groups of individuals in the training set can lead to a more equal treatment of these demographic groups. Experiments on two state-of-the-art neural architectures and a large-scale public dataset show that models trained with demographically-balanced training sets exhibit a fairer behavior on different groups, while still being accurate. Our study is expected to provide a solid basis for instilling beyond-accuracy objectives (e.g., fairness) in speaker recognition
Equality of Learning Opportunity via Individual Fairness in Personalized Recommendations
Get PDFOnline education platforms play an increasingly important role in mediating the success of individuals' careers. Therefore, while building overlying content recommendation services, it becomes essential to guarantee that learners are provided with equal recommended learning opportunities, according to the platform principles, context, and pedagogy. Though the importance of ensuring equality of learning opportunities has been well investigated in traditional institutions, how this equality can be operationalized in online learning ecosystems through recommender systems is still under-explored. In this paper, we shape a blueprint of the decisions and processes to be considered in the context of equality of recommended learning opportunities, based on principles that need to be empirically-validated (no evaluation with live learners has been performed). To this end, we first provide a formalization of educational principles that model recommendations' learning properties, and a novel fairness metric that combines them to monitor the equality of recommended learning opportunities among learners. Then, we envision a scenario wherein an educational platform should be arranged in such a way that the generated recommendations meet each principle to a certain degree for all learners, constrained to their individual preferences. Under this view, we explore the learning opportunities provided by recommender systems in a course platform, uncovering systematic inequalities. To reduce this effect, we propose a novel post-processing approach that balances personalization and equality of recommended opportunities. Experiments show that our approach leads to higher equality, with a negligible loss in personalization. This paper provides a theoretical foundation for future studies of learners' preferences and limits concerning the equality of recommended learning opportunities.ML4E
Adversarial optimization for dictionary attacks on speaker verification
No full textIn this paper, we assess vulnerability of speaker verification systems to dictionary attacks. We seek master voices, i.e., adversarial utterances optimized to match against a large number of users by pure chance. First, we perform menagerie analysis to identify utterances which intrinsically hold this property. Then, we propose an adversarial optimization approach for generating master voices synthetically. Our experiments show that, even in the most secure configuration, on average, a master voice can match approx. 20% of females and 10% of males without any knowledge about the population. We demonstrate that dictionary attacks should be considered as a feasible threat model for sensitive and high-stakes deployments of speaker verification
- …
