Challenges and opportunities of multimodal data in human learning: The computer science students' perspective

Multimodal data have the potential to explore emerging learning practices that extend human cognitive capacities. A critical issue stretching in many multimodal learning analytics (MLA) systems and studies is the current focus aimed at supporting researchers to model learner behaviours, rather than directly supporting learners. Moreover, many MLA systems are designed and deployed without learners' involvement. We argue that in order to create MLA interfaces that directly support learning, we need to gain an expanded understanding of how multimodal data can support learners' authentic needs. We present a qualitative study in which 40 computer science students were tracked in an authentic learning activity using wearable and static sensors. Our findings outline learners' curated representations about multimodal data and the non-technical challenges in using these data in their learning practice. The paper discusses 10 dimensions that can serve as guidelines for researchers and designers to create effective and ethically aware student-facing MLA innovations.     

Optimization of the grapes reception process

Neural Computing and Applications - Grapes reception is a key process in wine production. The harvest days are extremely challenging days in managing the reception of the grapes, as the winery...     

How far can one compress digital fault records? Analysis of a matching pursuit-based algorithm

In this paper, we verify how far electric disturbance signals can be compressed without compromising the analysis of encoded fault records. A recently proposed compression algorithm, referred to as Damped Sinusoidal Matching Pursuit (DSMP) has the remarkable feature of obtaining both compact and physically interpretable representations. However, for fault analysis applications, one is primarily interested in how accurate can be the analysis performed on compressed signals, instead of evaluating mean-squared error figures. Unlike previous works in digital fault records compression, the performance of the DSMP compression method is evaluated using a protocol based on fault analysis procedures commonly performed by expert engineers. This protocol is applied for comparing the results obtained in the analysis of both uncompressed records and their compressed versions at different compression ratios. The results show that the DSMP is a reliable compression system since it achieves high compression ratios (6.4:1) without causing fault analysis misinterpretation.     

Flow-based local optimization for image-to-geometry projection

The projection of a photographic data set on a 3D model is a robust and widely applicable way to acquire appearance information of an object. The first step of this procedure is the alignment of the images on the 3D model. While any reconstruction pipeline aims at avoiding misregistration by improving camera calibrations and geometry, in practice a perfect alignment cannot always be reached. Depending on the way multiple camera images are fused on the object surface, remaining misregistrations show up either as ghosting or as discontinuities at transitions from one camera view to another. In this paper we propose a method, based on the computation of Optical Flow between overlapping images, to correct the local misalignment by determining the necessary displacement. The goal is to correct the symptoms of misregistration, instead of searching for a globally consistent mapping, which might not exist. The method scales up well with the size of the data set (both photographic and geometric) and is quite independent of the characteristics of the 3D model (topology cleanliness, parametrization, density). The method is robust and can handle real world cases that have different characteristics: low level geometric details and images that lack enough features for global optimization or manual methods. It can be applied to different mapping strategies, such as texture or per-vertex attribute encoding.     

Interactive quadrangulation with Reeb atlases and connectivity textures

Creating high-quality quad meshes from triangulated surfaces is a highly nontrivial task that necessitates consideration of various application specific metrics of quality. In our work, we follow the premise that automatic reconstruction techniques may not generate outputs meeting all the subjective quality expectations of the user. Instead, we put the user at the center of the process by providing a flexible, interactive approach to quadrangulation design. By combining scalar field topology and combinatorial connectivity techniques, we present a new framework, following a coarse to fine design philosophy, which allows for explicit control of the subjective quality criteria on the output quad mesh, at interactive rates. Our quadrangulation framework uses the new notion of Reeb atlas editing, to define with a small amount of interactions a coarse quadrangulation of the model, capturing the main features of the shape, with user prescribed extraordinary vertices and alignment. Fine grain tuning is easily achieved with the notion of connectivity texturing, which allows for additional extraordinary vertices specification and explicit feature alignment, to capture the high-frequency geometries. Experiments demonstrate the interactivity and flexibility of our approach, as well as its ability to generate quad meshes of arbitrary resolution with high-quality statistics, while meeting the user’s own subjective requirements.     

Modeling of the condyle elements within a biomechanical knee model

The development of a computational multibody knee model able to capture some of the fundamental properties of the human knee articulation is presented. This desideratum is reached by including the kinetics of the real knee articulation. The research question is whether an accurate modeling of the condyle contact in the knee will lead to reproduction of the complex combination of flexion/extension, abduction/adduction, and tibial rotation observed in the real knee. The model is composed by two anatomic segments, the tibia and the femur, whose characteristics are functions of the geometric and anatomic properties of the real bones. The biomechanical model characterization is developed under the framework of multibody systems methodologies using Cartesian coordinates. The type of approach used in the proposed knee model is the joint surface contact conditions between ellipsoids, representing the two femoral condyles, and points, representing the tibial plateau and the menisci. These elements are closely fitted to the actual knee geometry. This task is undertaken by considering a parameter optimization process to replicate experimental data published in the literature, namely that by Lafortune and his coworkers in 1992. Then kinematic data in the form of flexion/extension patterns are imposed on the model corresponding to the stance phase of the human gait. From the results obtained, by performing several computational simulations, it can be observed that the knee model approximates the average secondary motion patterns observed in the literature. Because the literature reports considerable inter-individual differences in the secondary motion patterns, the knee model presented here is also used to check whether it is possible to reproduce the observed differences with reasonable variations of bone shape parameters. This task is accomplished by a parameter study, in which the main variables that define the geometry of condyles are taken into account. It was observed that the data reveal a difference in secondary kinematics of the knee in flexion versus extension. The likely explanation for this fact is the elastic component of the secondary motions created by the combination of joint forces and soft tissue deformations. The proposed knee model is, therefore, used to investigate whether this observed behavior can be explained by reasonable elastic deformations of the points representing the menisci in the model.     

Automatic stress exaggeration by prosody modification to assist language learners perceive sentence stress

In this paper, we propose a set of automatic stress exaggeration methods that can enlarge the differences between stressed and unstressed syllables. Our stress exaggeration methods can be used in computer-aided language learning systems to assist second language learners perceive stress patterns. The intention of our automatic stress exaggeration methods is to support hyper-pronunciation training which is commonly used in classrooms by teachers. In hyper-pronunciation training, exaggeration is used to help learners increase their awareness of acoustic features and effectively apply these features into their pronunciation. Duration, pitch and intensity have been claimed to be the main acoustic features that are closely related to stress in English language. Thus, four stress exaggeration methods are proposed in this paper: (i) duration-based stress exaggeration, (ii) pitch-based stress exaggeration, (iii) intensity-based stress exaggeration, and (iv) a combined stress exaggeration method that integrates the duration-based, pitch-based and intensity-based exaggeration methods. Our perceptual experimental results show that resynthesised stimuli by our proposed stress exaggerated methods can help learners of English as a Second Language (ESL) better perceive English stress patterns significantly.     

Evolutionary Generalized Radial Basis Function neural networks for improving prediction accuracy in gene classification using feature selection

Radial Basis Function Neural Networks (RBFNNs) have been successfully employed in several function approximation and pattern recognition problems. The use of different RBFs in RBFNN has been reported in the literature and here the study centres on the use of the Generalized Radial Basis Function Neural Networks (GRBFNNs). An interesting property of the GRBF is that it can continuously and smoothly reproduce different RBFs by changing a real parameter τ. In addition, the mixed use of different RBF shapes in only one RBFNN is allowed. Generalized Radial Basis Function (GRBF) is based on Generalized Gaussian Distribution (GGD), which adds a shape parameter, τ, to standard Gaussian Distribution. Moreover, this paper describes a hybrid approach, Hybrid Algorithm (HA), which combines evolutionary and gradient-based learning methods to estimate the architecture, weights and node topology of GRBFNN classifiers. The feasibility and benefits of the approach are demonstrated by means of six gene microarray classification problems taken from bioinformatic and biomedical domains. Three filters were applied: Fast Correlation-Based Filter (FCBF), Best Incremental Ranked Subset (BIRS), and Best Agglomerative Ranked Subset (BARS); this was done in order to identify salient expression genes from among the thousands of genes in microarray data that can directly contribute to determining the class membership of each pattern. After different gene subsets were obtained, the proposed methodology was performed using the selected gene subsets as new input variables. The results confirm that the GRBFNN classifier leads to a promising improvement in accuracy.     

Strong accumulators from collision-resistant hashing

Accumulator schemes were introduced in order to represent a large set of values as one short value called the accumulator. These schemes allow one to generate membership proofs, that is, short witnesses that a certain value belongs to the set. In universal accumulator schemes, efficient proofs of non-membership can also be created. Li et?al. (Proceedings of applied cryptography and network security—ACNS ’07, LNCS, vol 4521, 2007), building on the work of Camenisch and Lysyanskaya (Advances in cryptology—proceedings of Crypto ’02, LNCS, vol 2442. Springer, Berlin, pp 61–76, 2002), proposed an efficient accumulator scheme, which relies on a trusted accumulator manager. Specifically, a manager that correctly performs accumulator updates. In this work, we introduce the notion of strong universal accumulator schemes, which are similar in functionality to universal accumulator schemes, but do not assume the accumulator manager is trusted. We also formalize the security requirements for such schemes. We then give a simple construction of a strong universal accumulator scheme, which is provably secure under the assumption that collision-resistant hash functions exist. The weaker requirement on the accumulator manager comes at a price; our scheme is less efficient than known universal accumulator schemes—the size of (non)membership witnesses is logarithmic in the size of the accumulated set in contrast to constant in the scheme of Camenisch and Lysyanskaya. Finally, we show how to use strong universal accumulators to solve a problem of practical relevance, the so-called e-Invoice Factoring Problem.     

Propagation of transverse bulk and surface acoustic waves in LiNbO (3) variable time-delay devices

Experimental measurements are reported on voltage-controlled acoustic time-delay lines operating at 1 GHz in the nearly pure shear-horizontal (S-H) mode in 38 degrees rotated Y-cut LiNbO(3). The high-acoustic velocity (4800 m/s) in conjunction with the large electroacoustic effect exhibited by this orientation allows high-frequency operation and optimum time-delay tuning sensitivity with a planar, single surface, device geometry. The authors demonstrate fractional time delay of 0.3x10(-6) V(-1 ) for surface electrodes that produce an in-plane E-field. However, the simultaneous excitation and propagation of both a leaky surface-acoustic wave (LSAW) and surface skimming bulk wave (SSBW), both as (nearly pure) S-H waves in these devices, seriously restricts the extent to which it is possible to maximize the time delay modulation sensitivity by reducing electrode gap spacing as done in similar SAW devices. The LSAW and surface-skimming body wave (SSBW) propagate at nearly the same velocity on a free surface, and perturbation of their velocity and relative attenuation rates by surface electrodes causes pronounced interference effects between the two modes for some device geometries.