Examining the Assessment Opportunities for Cultural Connectedness for Student Learning : A Sociocultural Analysis

The disparity that exists between the highest and lowest achievers together with deficit approaches to teaching, learning and assessment raise serious equity issues related to fairness, validity, culture and access, which were analysed in a recent Australian Research Council funded project. This chapter explores the potential that exists for teachers to work with Indigenous Teacher Assistants (ITAs) to secure cultural connectedness in teaching, learning and assessment of Indigenous students. The study was a design experiment, which was conducted in seven Catholic and Independent primary schools in northern Queensland and involved semi-structured focus group interviews with Year 4 and 6 Indigenous students, principals, teachers and Indigenous Teacher Assistants. Classroom observations and document analyses were also conducted. This corpus of data was analysed using a sociocultural theoretical lens. The use of a sociocultural analysis helped to identify cultural influences, Indigenous students’ funds of knowledge and values. The information from this analysis was made explicit to teachers to demonstrate how they could enhance their pedagogic and assessment practices by embracing and extending the cultural spaces for learning and teaching of Indigenous students. The way in which teachers construct their interactions for greater cultural connectedness and enhanced learning would appear to rely on relationship building with Indigenous staff, Indigenous students’ cultural knowledge, and improved understanding of assessment and related equity issues.

Transitioning from dependent to more independent learners : Using video tutorials as a means of helping students help themselves

This paper examines the use of short video tutorials in a post-graduate accounting subject, as a means of helping students transition from dependent to more independent learners. Five short (three to five minute) video tutorials were introduced in an effort to shift the reliance for learning from the lecturer to the student. Students’ usage of video tutorials, comments by students, and reliance on teaching staff for individual assistance were monitored over three semesters from 2008 to 2009. Interviews with students were then conducted in late 2009 to more comprehensively evaluate the use and benefits of video tutorials. Findings reveal preliminary but positive outcomes in terms of both more efficient teaching and more effective learning.

Real-time video event detection in crowded scenes using MPEG derived features : a multiple instance learning approach

This paper presents an investigation into event detection in crowded scenes, where the event of interest co-occurs with other activities and only binary labels at the clip level are available. The proposed approach incorporates a fast feature descriptor from the MPEG domain, and a novel multiple instance learning (MIL) algorithm using sparse approximation and random sensing. MPEG motion vectors are used to build particle trajectories that represent the motion of objects in uniform video clips, and the MPEG DCT coefficients are used to compute a foreground map to remove background particles. Trajectories are transformed into the Fourier domain, and the Fourier representations are quantized into visual words using the K-Means algorithm. The proposed MIL algorithm models the scene as a linear combination of independent events, where each event is a distribution of visual words. Experimental results show that the proposed approaches achieve promising results for event detection compared to the state-of-the-art.

The why dimension - opening frontiers for digital learning

Digital learning has come a long way from the days of simple 'if-then' queries. It is now enabled by countless innovations that support knowledge sharing, openness, flexibility, and independent inquiry. Set against an evolutionary context this study investigated innovations that directly support human inquiry. Specifically, it identified five activities that together are defined as the 'why dimension' – asking, learning, understanding, knowing, and explaining why. Findings highlight deficiencies in mainstream search-based approaches to inquiry, which tend to privilege the retrieval of information as distinct from explanation. Instrumental to sense-making, the 'why dimension' provides a conceptual framework for development of 'sense-making technologies'.

Translations of industry-based student learning and academic performances

We hypothesized that Industry based learning and teaching, especially through industry assigned student projects or training programs, is an integral part of science, technology, engineering and mathematics (STEM) education. In this paper we show that industry-based student training and experience increases students’ academic performances independent to the organizational parameters and contexts. The literature on industry-based student training focuses on employability and the industry dimension, and neglects in many ways the academic dimension. We observed that the association factors between academic attributes and contributions of industry-based student training are central and vital to the technological learning experiences. We explore international initiatives and statistics collected of student projects in two categories: Industry based learning performances and on campus performances. The data collected were correlated to five (5) universities in different industrialized countries, e.g., Australia N=545, Norway N=279, Germany N=74, France N=107 and Spain N=802 respectively. We analyzed industry-based student training along with company assigned student projects compared with in comparisons to campus performance. The data that suggests a strong correlation between industry-based student training per se and improved performance profiles or increasing motivation shows that industry-based student training increases student academic performance independent of organizational parameters and contexts. The programs we augmented were orthogonal to each other however, the trend of the students’ academic performances are identical. An isolated cohort for the reported countries that opposed our hypothesis warrants further investigation.

Contribution of industry-based student learning performance for the STEM education

We hypothesized that Industry based learning and teaching, especially through company assigned student projects or training programs, is an integral part of science, technology, engineering and mathematics (STEM) education. In this paper we show that industry-based student training and experience increases students’ academic performances independent to the organizational parameters and contexts. The literature on industry-based student training focuses on employability and the industry dimension, and neglects in many ways the academic dimension. We observed that the association factors between academic attributes and contributions of industry-based student training are central and vital to the technological learning experiences. We explore international initiatives and statistics collected of student projects in two categories: Industry based learning performances and on campus performances. The data collected were correlated to five (5) universities in different industrialized countries, e.g., Australia N=545 projects, Norway N=279, Germany N=74, France N=107 and Spain N=802. We analyzed industry-based student training along with company assigned student projects compared with in comparisons to campus performance. The data that suggests a strong correlation between industry-based student training per se and improved performance profiles or increasing motivation shows that industry-based student training increases student academic performance independent of organizational parameters and contexts. The programs we augmented were orthogonal to each other however, the trend of the students’ academic performances are identical. An isolated cohort for the reported countries that opposed our hypothesis warrants further investigation.

The Learning Component of Dynamic Allocation Indices

For a multiarmed bandit problem with exponential discounting the optimal allocation rule is defined by a dynamic allocation index defined for each arm on its space. The index for an arm is equal to the expected immediate reward from the arm, with an upward adjustment reflecting any uncertainty about the prospects of obtaining rewards from the arm, and the possibilities of resolving those uncertainties by selecting that arm. Thus the learning component of the index is defined to be the difference between the index and the expected immediate reward. For two arms with the same expected immediate reward the learning component should be larger for the arm for which the reward rate is more uncertain. This is shown to be true for arms based on independent samples from a fixed distribution with an unknown parameter in the cases of Bernoulli and normal distributions, and similar results are obtained in other cases.

Learning Nonlinear Visual Processing from Natural Images

The paradigm of computational vision hypothesizes that any visual function -- such as the recognition of your grandparent -- can be replicated by computational processing of the visual input. What are these computations that the brain performs? What should or could they be? Working on the latter question, this dissertation takes the statistical approach, where the suitable computations are attempted to be learned from the natural visual data itself. In particular, we empirically study the computational processing that emerges from the statistical properties of the visual world and the constraints and objectives specified for the learning process. This thesis consists of an introduction and 7 peer-reviewed publications, where the purpose of the introduction is to illustrate the area of study to a reader who is not familiar with computational vision research. In the scope of the introduction, we will briefly overview the primary challenges to visual processing, as well as recall some of the current opinions on visual processing in the early visual systems of animals. Next, we describe the methodology we have used in our research, and discuss the presented results. We have included some additional remarks, speculations and conclusions to this discussion that were not featured in the original publications. We present the following results in the publications of this thesis. First, we empirically demonstrate that luminance and contrast are strongly dependent in natural images, contradicting previous theories suggesting that luminance and contrast were processed separately in natural systems due to their independence in the visual data. Second, we show that simple cell -like receptive fields of the primary visual cortex can be learned in the nonlinear contrast domain by maximization of independence. Further, we provide first-time reports of the emergence of conjunctive (corner-detecting) and subtractive (opponent orientation) processing due to nonlinear projection pursuit with simple objective functions related to sparseness and response energy optimization. Then, we show that attempting to extract independent components of nonlinear histogram statistics of a biologically plausible representation leads to projection directions that appear to differentiate between visual contexts. Such processing might be applicable for priming, \ie the selection and tuning of later visual processing. We continue by showing that a different kind of thresholded low-frequency priming can be learned and used to make object detection faster with little loss in accuracy. Finally, we show that in a computational object detection setting, nonlinearly gain-controlled visual features of medium complexity can be acquired sequentially as images are encountered and discarded. We present two online algorithms to perform this feature selection, and propose the idea that for artificial systems, some processing mechanisms could be selectable from the environment without optimizing the mechanisms themselves. In summary, this thesis explores learning visual processing on several levels. The learning can be understood as interplay of input data, model structures, learning objectives, and estimation algorithms. The presented work adds to the growing body of evidence showing that statistical methods can be used to acquire intuitively meaningful visual processing mechanisms. The work also presents some predictions and ideas regarding biological visual processing.

Learning Challenges in Biogas Production for Sustainability : An activity theoretical study of a network from a swine industry chain

This study is about the challenges of learning in the creation and implementation of new sustainable technologies. The system of biogas production in the Programme of Sustainable Swine Production (3S Programme) conducted by the Sadia food processing company in Santa Catarina State, Brazil, is used as a case example for exploring the challenges, possibilities and obstacles of learning in the use of biogas production as a way to increase the environmental sustainability of swine production. The aim is to contribute to the discussion about the possibilities of developing systems of biogas production for sustainability (BPfS). In the study I develop hypotheses concerning the central challenges and possibilities for developing systems of BPfS in three phases. First, I construct a model of the network of activities involved in the BP for sustainability in the case study. Next, I construct a) an idealised model of the historically evolved concepts of BPfS through an analysis of the development of forms of BP and b) a hypothesis of the current central contradictions within and between the activity systems involved in BP for sustainability in the case study. This hypothesis is further developed through two actual empirical analyses: an analysis of the actors senses in taking part in the system, and an analysis of the disturbance processes in the implementation and operation of the BP system in the 3S Programme. The historical analysis shows that BP for sustainability in the 3S Programme emerged as a feasible solution for the contradiction between environmental protection and concentration, intensification and specialisation in swine production. This contradiction created a threat to the supply of swine to the food processing company. In the food production activity, the contradiction was expressed as a contradiction between the desire of the company to become a sustainable company and the situation in the outsourced farms. For the swine producers the contradiction was expressed between the contradictory rules in which the market exerted pressure which pushed for continual increases in scale, specialisation and concentration to keep the production economically viable, while the environmental rules imposed a limit to this expansion. Although the observed disturbances in the biogas system seemed to be merely technical and localised within the farms, the analysis proposed that these disturbances were formed in and between the activity systems involved in the network of BPfS during the implementation. The disturbances observed could be explained by four contradictions: a) contradictions between the new, more expanded activity of sustainable swine production and the old activity, b) a contradiction between the concept of BP for carbon credits and BP for local use in the BPfS that was implemented, c) contradictions between the new UNFCCC1 methodology for applying for carbon credits and the small size of the farms, and d) between the technologies of biogas use and burning available in the market and the small size of the farms. The main finding of this study relates to the zone of proximal development (ZPD) of the BPfS in Sadia food production chain. The model is first developed as a general model of concepts of BPfS and further developed here to the specific case of the BPfS in the 3S Programme. The model is composed of two developmental dimensions: societal and functional integration. The dimension of societal integration refers to the level of integration with other activities outside the farm. At one extreme, biogas production is self-sufficient and highly independent and the products of BP are consumed within the farm, while at the other extreme BP is highly integrated in markets and networks of collaboration, and BP products are exchanged within the markets. The dimension of functional integration refers to the level of integration between products and production processes so that economies of scope can be achieved by combining several functions using the same utility. At one extreme, BP is specialised in only one product, which allows achieving economies of scale, while at the other extreme there is an integrated production in which several biogas products are produced in order to maximise the outcomes from the BP system. The analysis suggests that BP is moving towards a societal integration, towards the market and towards a functional integration in which several biogas products are combined. The model is a hypothesis to be further tested through interventions by collectively constructing the new proposed concept of BPfS. Another important contribution of this study refers to the concept of the learning challenge. Three central learning challenges for developing a sustainable system of BP in the 3S Programme were identified: 1) the development of cheaper and more practical technologies of burning and measuring the gas, as well as the reduction of costs of the process of certification, 2) the development of new ways of using biogas within farms, and 3) the creation of new local markets and networks for selling BP products. One general learning challenge is to find more varied and synergic ways of using BP products than solely for the production of carbon credits. Both the model of the ZPD of BPfS and the identified learning challenges could be used as learning tools to facilitate the development of biogas production systems. The proposed model of the ZPD could be used to analyse different types of agricultural activities that face a similar contradiction. The findings could be used in interventions to help actors to find their own expansive actions and developmental projects for change. Rather than proposing a standardised best concept of BPfS, the idea of these learning tools is to facilitate the analysis of local situations and to help actors to make their activities more sustainable.

Prototype learning methods for online handwriting recognition

In this paper, we study different methods for prototype selection for recognizing handwritten characters of Tamil script. In the first method, cumulative pairwise- distances of the training samples of a given class are used to select prototypes. In the second method, cumulative distance to allographs of different orientation is used as a criterion to decide if the sample is representative of the group. The latter method is presumed to offset the possible orientation effect. This method still uses fixed number of prototypes for each of the classes. Finally, a prototype set growing algorithm is proposed, with a view to better model the differences in complexity of different character classes. The proposed algorithms are tested and compared for both writer independent and writer adaptation scenarios.

A neural network based automatic generation control design through reinforcement learning

This paper presents the design and implementation of a learning controller for the Automatic Generation Control (AGC) in power systems based on a reinforcement learning (RL) framework. In contrast to the recent RL scheme for AGC proposed by us, the present method permits handling of power system variables such as Area Control Error (ACE) and deviations from scheduled frequency and tie-line flows as continuous variables. (In the earlier scheme, these variables have to be quantized into finitely many levels). The optimal control law is arrived at in the RL framework by making use of Q-learning strategy. Since the state variables are continuous, we propose the use of Radial Basis Function (RBF) neural networks to compute the Q-values for a given input state. Since, in this application we cannot provide training data appropriate for the standard supervised learning framework, a reinforcement learning algorithm is employed to train the RBF network. We also employ a novel exploration strategy, based on a Learning Automata algorithm,for generating training samples during Q-learning. The proposed scheme, in addition to being simple to implement, inherits all the attractive features of an RL scheme such as model independent design, flexibility in control objective specification, robustness etc. Two implementations of the proposed approach are presented. Through simulation studies the attractiveness of this approach is demonstrated.

Learning polyhedral classifiers using logistic function

In this paper we propose a new algorithm for learning polyhedral classifiers. In contrast to existing methods for learning polyhedral classifier which solve a constrained optimization problem, our method solves an unconstrained optimization problem. Our method is based on a logistic function based model for the posterior probability function. We propose an alternating optimization algorithm, namely, SPLA1 (Single Polyhedral Learning Algorithm1) which maximizes the loglikelihood of the training data to learn the parameters. We also extend our method to make it independent of any user specified parameter (e.g., number of hyperplanes required to form a polyhedral set) in SPLA2. We show the effectiveness of our approach with experiments on various synthetic and real world datasets and compare our approach with a standard decision tree method (OC1) and a constrained optimization based method for learning polyhedral sets.

Subject independent human action recognition using spatio-depth information and meta-cognitive RBF network

In this paper, we present a machine learning approach for subject independent human action recognition using depth camera, emphasizing the importance of depth in recognition of actions. The proposed approach uses the flow information of all 3 dimensions to classify an action. In our approach, we have obtained the 2-D optical flow and used it along with the depth image to obtain the depth flow (Z motion vectors). The obtained flow captures the dynamics of the actions in space time. Feature vectors are obtained by averaging the 3-D motion over a grid laid over the silhouette in a hierarchical fashion. These hierarchical fine to coarse windows capture the motion dynamics of the object at various scales. The extracted features are used to train a Meta-cognitive Radial Basis Function Network (McRBFN) that uses a Projection Based Learning (PBL) algorithm, referred to as PBL-McRBFN, henceforth. PBL-McRBFN begins with zero hidden neurons and builds the network based on the best human learning strategy, namely, self-regulated learning in a meta-cognitive environment. When a sample is used for learning, PBLMcRBFN uses the sample overlapping conditions, and a projection based learning algorithm to estimate the parameters of the network. The performance of PBL-McRBFN is compared to that of a Support Vector Machine (SVM) and Extreme Learning Machine (ELM) classifiers with representation of every person and action in the training and testing datasets. Performance study shows that PBL-McRBFN outperforms these classifiers in recognizing actions in 3-D. Further, a subject-independent study is conducted by leave-one-subject-out strategy and its generalization performance is tested. It is observed from the subject-independent study that McRBFN is capable of generalizing actions accurately. The performance of the proposed approach is benchmarked with Video Analytics Lab (VAL) dataset and Berkeley Multimodal Human Action Database (MHAD). (C) 2013 Elsevier Ltd. All rights reserved.

Database Independent Human Emotion Recognition with Meta-Cognitive Neuro-Fuzzy Inference System

Facial emotions are the most expressive way to display emotions. Many algorithms have been proposed which employ a particular set of people (usually a database) to both train and test their model. This paper focuses on the challenging task of database independent emotion recognition, which is a generalized case of subject-independent emotion recognition. The emotion recognition system employed in this work is a Meta-Cognitive Neuro-Fuzzy Inference System (McFIS). McFIS has two components, a neuro-fuzzy inference system, which is the cognitive component and a self-regulatory learning mechanism, which is the meta-cognitive component. The meta-cognitive component, monitors the knowledge in the neuro-fuzzy inference system and decides on what-to-learn, when-to-learn and how-to-learn the training samples, efficiently. For each sample, the McFIS decides whether to delete the sample without being learnt, use it to add/prune or update the network parameter or reserve it for future use. This helps the network avoid over-training and as a result improve its generalization performance over untrained databases. In this study, we extract pixel based emotion features from well-known (Japanese Female Facial Expression) JAFFE and (Taiwanese Female Expression Image) TFEID database. Two sets of experiment are conducted. First, we study the individual performance of both databases on McFIS based on 5-fold cross validation study. Next, in order to study the generalization performance, McFIS trained on JAFFE database is tested on TFEID and vice-versa. The performance The performance comparison in both experiments against SVNI classifier gives promising results.