With the growth of Internet and technology in the past decade, online learning has become increasingly popular and evolved. Online examination is an integral and vital component of online learning. Student assessment in online learning is largely submitted remotely without any face-to-face interaction and therefore, student authentication is widely seen as one of the major challenges. This study aims to investigate potential threats to student authentication in the online examinations and analyzing the benefits and limitations of the existing authentication approaches. We propose the use of challenge questions for student authentication in the online examinations. For this purpose, we designed a profile based authentication framework (PBAF) together with a user-id and password for the authentication of students during online examinations, utilizing a cohort of personal and academic questions as challenge questions. We conducted an empirical study on a group of online students from local and overseas Universities. The result shows the impact of questions type on the usability, in particular the amount of time taken by the introduction of the proposed approach. We also conducted a post experiment survey to collect student feedback on the proposed technique.

VRAM reconstruction of the perineum can be safely performed following APE with results that compare favourably with other techniques. Most flap complications are minor, although these are still associated with an increase in the length of hospital stay.

Online examinations are an integral component of online learning environments and research studies have identified academic dishonesty as a critical threat to the credibility of such examinations. Academic dishonesty exists in many forms. Collusion is seen as a major security threat, wherein a student invites a third party for help or to impersonate him or her in an online examination. This work aims to investigate the authentication of students using text-based and image-based challenge questions. The study reported in this paper involved 70 online participants from nine countries completing a five week online course and simulating an abuse case scenario. The results of a usability analysis suggested that i) image-based questions are more usable than text-based questions (p < 0.01) and ii) using a more flexible data entry method increased the usability of text-based questions (p < 0.01). An impersonation abuse scenario was simulated to test the influence of sharing with different database sizes. The findings revealed that iii) an increase in the number of questions shared for impersonation increased the success of an impersonation attack and the results showed a significant linear trend (p < 0.01). However, the number of correct answers decreased when the attacker had to memorize and answer the questions in an invigilated online examination or their response to questions was timed. The study also revealed that iv) Educ Inf Technol (2019) an increase in the size of challenge question database decreased the success of an impersonation attack (p < 0.01).

In exemplar-based speech enhancement systems, lower dimensional features are preferred over the full-scale DFT features for their reduced computational complexity and the ability to better generalize for the unseen cases. But in order to obtain the Wiener-like filter for noisy DFT enhancement, the speech and noise estimates obtained in the feature space need to be mapped to the DFT space, which yield a low-rank approximation of the estimates resulting in a sub-optimal filter. This paper proposes a novel method using coupled dictionaries where the exemplars for the required feature space and the DFT space are jointly extracted and the estimates are directly obtained in the DFT space following the decomposition in the chosen feature space. Simulation experiments revealed that the proposed approach, where the activations of exemplars calculated using the Mel resolution are directly used to obtain the Wiener filter in the DFT space, results in improved signal-to-distortion ratio (SDR) when compared to the system without coupled dictionaries. To further motivate the use of coupled dictionaries, the paper also investigates the use of modulation envelope features for the exemplar-based speech enhancement.

Hearing aid users are challenged in listening situations with noise and especially speech-on-speech situations with two or more competing voices. Specifically, the task of attending to and segregating two competing voices is particularly hard, unlike for normal-hearing listeners, as shown in a small sub-experiment. In the main experiment, the competing voices benefit of a deep neural network (DNN) based stream segregation enhancement algorithm was tested on hearing-impaired listeners. A mixture of two voices was separated using a DNN and presented to the two ears as individual streams and tested for word score. Compared to the unseparated mixture, there was a 13%-point benefit from the separation, while attending to both voices. If only one output was selected as in a traditional target-masker scenario, a larger benefit of 37%-points was found. The results agreed well with objective metrics and show that for hearing-impaired listeners, DNNs have a large potential for improving stream segregation and speech intelligibility in difficult scenarios with two equally important targets without any prior selection of a primary target stream. An even higher benefit can be obtained if the user can select the preferred target via remote control.