Abstract: Although backpropagation ANNs generally predict
better than decision trees do for pattern classification problems, they
are often regarded as black boxes, i.e., their predictions cannot be
explained as those of decision trees. In many applications, it is
desirable to extract knowledge from trained ANNs for the users to
gain a better understanding of how the networks solve the problems.
A new rule extraction algorithm, called rule extraction from artificial
neural networks (REANN) is proposed and implemented to extract
symbolic rules from ANNs. A standard three-layer feedforward ANN
is the basis of the algorithm. A four-phase training algorithm is
proposed for backpropagation learning. Explicitness of the extracted
rules is supported by comparing them to the symbolic rules generated
by other methods. Extracted rules are comparable with other methods
in terms of number of rules, average number of conditions for a rule,
and predictive accuracy. Extensive experimental studies on several
benchmarks classification problems, such as breast cancer, iris,
diabetes, and season classification problems, demonstrate the
effectiveness of the proposed approach with good generalization
ability.
Abstract: E-learning is not restricted to the use of new technologies for the online content, but also induces the adoption of new approaches to improve the quality of education. This quality depends on the ability of these approaches (technical and pedagogical) to provide an adaptive learning environment. Thus, the environment should include features that convey intentions and meeting the educational needs of learners by providing a customized learning path to acquiring a competency concerned In our proposal, we believe that an individualized learning path requires knowledge of the learner. Therefore, it must pass through a personalization of diagnosis to identify precisely the competency gaps to fill, and reduce the cognitive load To personalize the diagnosis and pertinently measure the competency gap, we suggest implementing the formative assessment in the e-learning environment and we propose the introduction of a pre-regulation process in the area of formative assessment, involving its individualization and implementation in e-learning.
Abstract: This research presents a system for post processing of
data that takes mined flat rules as input and discovers crisp as well as
fuzzy hierarchical structures using Learning Classifier System
approach. Learning Classifier System (LCS) is basically a machine
learning technique that combines evolutionary computing,
reinforcement learning, supervised or unsupervised learning and
heuristics to produce adaptive systems. A LCS learns by interacting
with an environment from which it receives feedback in the form of
numerical reward. Learning is achieved by trying to maximize the
amount of reward received. Crisp description for a concept usually
cannot represent human knowledge completely and practically. In the
proposed Learning Classifier System initial population is constructed
as a random collection of HPR–trees (related production rules) and
crisp / fuzzy hierarchies are evolved. A fuzzy subsumption relation is
suggested for the proposed system and based on Subsumption Matrix
(SM), a suitable fitness function is proposed. Suitable genetic
operators are proposed for the chosen chromosome representation
method. For implementing reinforcement a suitable reward and
punishment scheme is also proposed. Experimental results are
presented to demonstrate the performance of the proposed system.
Abstract: This paper is an exploration of the conceptual
confusion between E-learning and M-learning particularly in Africa.
Section I provides a background to the development of E-learning
and M-learning. Section II focuses on the conceptual analysis as it
applies to Africa. It is with an investigative and expansive mind that
this paper is elaborated to respond to a profound question of the
suitability of the concepts in a particular era in Africa. The aim of this
paper is therefore to shed light on which concept best suits the unique
situation of Africa in the era of cloud computing.
Abstract: Collaborative problem solving in e-learning can take
in the form of discussion among learner, creating a highly social
learning environment and characterized by participation and
interactivity. This paper, designed a collaborative learning
environment where agent act as co-learner, can play different roles
during interaction. Since different roles have been assigned to the
agent, learner will assume that multiple co-learner exists to help and
guide him all throughout the collaborative problem solving process,
but in fact, alone during the learning process. Specifically, it answers
the questions what roles of the agent should be incorporated to
contribute better learning outcomes, how agent will facilitate the
communication process to provide social learning and interactivity
and what are the specific instructional strategies that facilitate learner
participation, increased skill acquisition and develop critical thinking.
Abstract: The competitive learning is an adaptive process in
which the neurons in a neural network gradually become sensitive to
different input pattern clusters. The basic idea behind the Kohonen-s
Self-Organizing Feature Maps (SOFM) is competitive learning.
SOFM can generate mappings from high-dimensional signal spaces
to lower dimensional topological structures. The main features of this
kind of mappings are topology preserving, feature mappings and
probability distribution approximation of input patterns. To overcome
some limitations of SOFM, e.g., a fixed number of neural units and a
topology of fixed dimensionality, Growing Self-Organizing Neural
Network (GSONN) can be used. GSONN can change its topological
structure during learning. It grows by learning and shrinks by
forgetting. To speed up the training and convergence, a new variant
of GSONN, twin growing cell structures (TGCS) is presented here.
This paper first gives an introduction to competitive learning, SOFM
and its variants. Then, we discuss some GSONN with fixed
dimensionality, which include growing cell structures, its variants
and the author-s model: TGCS. It is ended with some testing results
comparison and conclusions.
Abstract: Obtaining labeled data in supervised learning is often
difficult and expensive, and thus the trained learning algorithm tends
to be overfitting due to small number of training data. As a result,
some researchers have focused on using unlabeled data which may
not necessary to follow the same generative distribution as the labeled
data to construct a high-level feature for improving performance on
supervised learning tasks. In this paper, we investigate the impact of
the relationship between unlabeled and labeled data for classification
performance. Specifically, we will apply difference unlabeled data
which have different degrees of relation to the labeled data for
handwritten digit classification task based on MNIST dataset. Our
experimental results show that the higher the degree of relation
between unlabeled and labeled data, the better the classification
performance. Although the unlabeled data that is completely from
different generative distribution to the labeled data provides the lowest
classification performance, we still achieve high classification performance.
This leads to expanding the applicability of the supervised
learning algorithms using unsupervised learning.
Abstract: The pedagogy project has been proven as an active
learning method, which is used to develop learner-s skills and
knowledge.The use of technology in the learning world, has filed
several gaps in the implementation of teaching methods, and online
evaluation of learners. However, the project methodology presents
challenges in the assessment of learners online.
Indeed, interoperability between E-learning platforms (LMS) is
one of the major challenges of project-based learning assessment.
Firstly, we have reviewed the characteristics of online assessment
in the context of project-based teaching. We addressed the
constraints encountered during the peer evaluation process.
Our approach is to propose a meta-model, which will describe a
language dedicated to the conception of peer assessment scenario in
project-based learning. Then we illustrate our proposal by an
instantiation of the meta-model through a business process in a
scenario of collaborative assessment on line.
Abstract: In the automotive industry test drives are being conducted
during the development of new vehicle models or as a part of
quality assurance of series-production vehicles. The communication
on the in-vehicle network, data from external sensors, or internal
data from the electronic control units is recorded by automotive
data loggers during the test drives. The recordings are used for fault
analysis. Since the resulting data volume is tremendous, manually
analysing each recording in great detail is not feasible.
This paper proposes to use machine learning to support domainexperts
by preventing them from contemplating irrelevant data and
rather pointing them to the relevant parts in the recordings. The
underlying idea is to learn the normal behaviour from available
recordings, i.e. a training set, and then to autonomously detect
unexpected deviations and report them as anomalies.
The one-class support vector machine “support vector data description”
is utilised to calculate distances of feature vectors. SVDDSUBSEQ
is proposed as a novel approach, allowing to classify subsequences
in multivariate time series data. The approach allows to
detect unexpected faults without modelling effort as is shown with
experimental results on recordings from test drives.
Abstract: Text categorization (the assignment of texts in natural language into predefined categories) is an important and extensively studied problem in Machine Learning. Currently, popular techniques developed to deal with this task include many preprocessing and learning algorithms, many of which in turn require tuning nontrivial internal parameters. Although partial studies are available, many authors fail to report values of the parameters they use in their experiments, or reasons why these values were used instead of others. The goal of this work then is to create a more thorough comparison of preprocessing parameters and their mutual influence, and report interesting observations and results.
Abstract: Self-directed learning (SDL) was developed initially
for adult learning. Guglielmino constructed a scale to measure SDL.
Recent researchers have applied this concept to children. Although
there are sufficient theoretical evidences to present the possibility of
applying this concept to children, empirical evidences were not
provided. This study aimed to examine the quality of SDL and
construct a scale to measure SDL among young children. A modified
scale of Guglielmino-s scale was constructed and piloted with 183
subjects of age 9. Findings suggest that the qualities of SDL in young
ages are apparently congruent with that of adults.
Abstract: The main purpose of this study was to determine the predictors of academic achievement of student Information and Communications Technologies (ICT) teachers with different learning styles. Participants were 148 student ICT teachers from Ankara University. Participants were asked to fill out a personal information sheet, the Turkish version of Kolb-s Learning Style Inventory, Weinstein-s Learning and Study Strategies Inventory, Schommer's Epistemological Beliefs Questionnaire, and Eysenck-s Personality Questionnaire. Stepwise regression analyses showed that the statistically significant predictors of the academic achievement of the accommodators were attitudes and high school GPAs; of the divergers was anxiety; of the convergers were gender, epistemological beliefs, and motivation; and of the assimilators were gender, personality, and test strategies. Implications for ICT teaching-learning processes and teacher education are discussed.
Abstract: Business and IT alignment has continued as a
top concern for business and IT executives for almost three
decades. Many researchers have conducted empirical studies on
the relationship between business-IT alignment and performance.
Yet, these approaches, lacking a social perspective, have had little
impact on sustaining performance and competitive advantage. In
addition to the limited alignment literature that explores
organisational learning that is represented in shared understanding,
communication, cognitive maps and experiences.
Hence, this paper proposes an integrated process that enables
social and intellectual dimensions through the concept of
organisational learning. In particular, the feedback and feedforward
process which provide a value creation across dynamic
multilevel of learning. This mechanism enables on-going
effectiveness through development of individuals, groups and
organisations, which improves the quality of business and IT
strategies and drives to performance.
Abstract: Using activity theory, organisational theory and
didactics as theoretical foundations, a comprehensive model of the
organisational dimensions relevant for learning and knowledge
transfer will be developed. In a second step, a Learning Assessment
Guideline will be elaborated. This guideline will be designed to
permit a targeted analysis of organisations to identify the status quo
in those areas crucial to the implementation of learning and
knowledge transfer. In addition, this self-analysis tool will enable
learning managers to select adequate didactic models for e- and
blended learning. As part of the European Integrated Project
"Process-oriented Learning and Information Exchange" (PROLIX),
this model of organisational prerequisites for learning and knowledge
transfer will be empirically tested in four profit and non-profit
organisations in Great Britain, Germany and France (to be finalized
in autumn 2006). The findings concern not only the capability of the
model of organisational dimensions, but also the predominant
perceptions of and obstacles to learning in organisations.
Abstract: An additive fuzzy system comprising m rules with
n inputs and p outputs in each rule has at least t m(2n + 2 p + 1)
parameters needing to be tuned. The system consists of a large
number of if-then fuzzy rules and takes a long time to tune its
parameters especially in the case of a large amount of training data
samples. In this paper, a new learning strategy is investigated to cope
with this obstacle. Parameters that tend toward constant values at the
learning process are initially fixed and they are not tuned till the end
of the learning time. Experiments based on applications of the
additive fuzzy system in function approximation demonstrate that the
proposed approach reduces the learning time and hence improves
convergence speed considerably.
Abstract: Bagging and boosting are among the most popular re-sampling ensemble methods that generate and combine a diversity of regression models using the same learning algorithm as base-learner. Boosting algorithms are considered stronger than bagging on noise-free data. However, there are strong empirical indications that bagging is much more robust than boosting in noisy settings. For this reason, in this work we built an ensemble using an averaging methodology of bagging and boosting ensembles with 10 sub-learners in each one. We performed a comparison with simple bagging and boosting ensembles with 25 sub-learners on standard benchmark datasets and the proposed ensemble gave better accuracy.
Abstract: A cognitive collaborative reinforcement learning
algorithm (CCRL) that incorporates an advisor into the learning
process is developed to improve supervised learning. An autonomous
learner is enabled with a self awareness cognitive skill to decide
when to solicit instructions from the advisor. The learner can also
assess the value of advice, and accept or reject it. The method is
evaluated for robotic motion planning using simulation. Tests are
conducted for advisors with skill levels from expert to novice. The
CCRL algorithm and a combined method integrating its logic with
Clouse-s Introspection Approach, outperformed a base-line fully
autonomous learner, and demonstrated robust performance when
dealing with various advisor skill levels, learning to accept advice
received from an expert, while rejecting that of less skilled
collaborators. Although the CCRL algorithm is based on RL, it fits
other machine learning methods, since advisor-s actions are only
added to the outer layer.
Abstract: A complex valued neural network is a neural network, which consists of complex valued input and/or weights and/or thresholds and/or activation functions. Complex-valued neural networks have been widening the scope of applications not only in electronics and informatics, but also in social systems. One of the most important applications of the complex valued neural network is in image and vision processing. In Neural networks, radial basis functions are often used for interpolation in multidimensional space. A Radial Basis function is a function, which has built into it a distance criterion with respect to a centre. Radial basis functions have often been applied in the area of neural networks where they may be used as a replacement for the sigmoid hidden layer transfer characteristic in multi-layer perceptron. This paper aims to present exhaustive results of using RBF units in a complex-valued neural network model that uses the back-propagation algorithm (called 'Complex-BP') for learning. Our experiments results demonstrate the effectiveness of a Radial basis function in a complex valued neural network in image recognition over a real valued neural network. We have studied and stated various observations like effect of learning rates, ranges of the initial weights randomly selected, error functions used and number of iterations for the convergence of error on a neural network model with RBF units. Some inherent properties of this complex back propagation algorithm are also studied and discussed.
Abstract: Biological evolution has generated a rich variety of
successful solutions; from nature, optimized strategies can be
inspired. One interesting example is the ant colonies, which are able
to exhibit a collective intelligence, still that their dynamic is simple.
The emergence of different patterns depends on the pheromone trail,
leaved by the foragers. It serves as positive feedback mechanism for
sharing information.
In this paper, we use the dynamic of TASEP as a model of
interaction at a low level of the collective environment in the ant-s
traffic flow. This work consists of modifying the movement rules of
particles “ants" belonging to the TASEP model, so that it adopts with
the natural movement of ants. Therefore, as to respect the constraints
of having no more than one particle per a given site, and in order to
avoid collision within a bidirectional circulation, we suggested two
strategies: decease strategy and waiting strategy. As a third work
stage, this is devoted to the study of these two proposed strategies-
stability. As a final work stage, we applied the first strategy to the
whole environment, in order to get to the emergence of traffic flow,
which is a way of learning.
Abstract: This paper discusses the development of a qualitative
simulator (abbreviated QRiOM) for predicting the behaviour of
organic chemical reactions. The simulation technique is based on the
qualitative process theory (QPT) ontology. The modelling constructs
of QPT embody notions of causality which can be used to explain the
behaviour of a chemical system. The major theme of this work is
that, in a qualitative simulation environment, students are able to
articulate his/her knowledge through the inspection of explanations
generated by software. The implementation languages are Java and
Prolog. The software produces explanation in various forms that
stresses on the causal theories in the chemical system which can be
effectively used to support learning.