Luminita Dumitriu

Methods to evaluate open source learning platforms

Open-source e-learning platforms have lately become an obvious choice whenever an e-learning infrastructure is being designed. Depending on the constraints of the e-learning organization, the best suited one should be selected. This decision is a difficult one, due to the diversity of available platforms. The main goal of this paper is to propose a new method for multiple criteria classification of these software systems, based upon Formal Concept Analysis, namely using conceptual lattices. By means of classification, the paper offers a comprehensive analysis of the features provided by several widely used open-source e-learning platforms. This analysis could support educators in choosing the most suitable platform for particular requirements.

Signal selective amplification: A solution for an improved detection of amphetamines with QCL equipped portable GC-IRAS spectrometers

In this article we are presenting a signal selective amplification method that was used for improving the automatic detection of amphetamines based on their infrared absorptions recorded in the narrow spectral window in which the QCL source of a new portable GC-IRAS spectrometer is emitting (1550-1330 cm-1). The expert system developed for class identity assignment is based on Principal Component Analysis (PCA) and distinguishes the amphetamines according to their toxic effect (stimulant and hallucinogenic amphetamines). The detection efficiency and accuracy was evaluated by using Cluster Analysis (CA).

A framework for conceptually modelling the domain knowledge of an instructional system

This paper presents a solution for conceptually modelling the teaching domain knowledge for computer-assisted instructional systems. The model consists of a theoretical framework and a knowledge representation approach. This model might be adopted in order to build a well-defined conceptual structure for the domain knowledge of a CAI system. The paper also presents an authoring system which implements the features of the modelling methods. This approach can offer a better solution to the problem of knowledge structuring and organising for computer-assisted instructional systems.

Domain Knowledge Modelling for Intelligent Instructional Systems

This paper presents a solution for conceptually modeling the training domain knowledge of computer-assisted instructional environments. A stringent requirement for an effective computer-assisted instructional environment is to elaborate a proper answer for the differentiated cognitive demands of users. The different cognitive style of learners imposes different modalities of presenting and structuring the information (the knowledge) to be taught. Conceptual organization of the training domain knowledge, with learning stages phasing, can constitute a better solution to the problem of adapting the instructional system interaction to users with different cognitive style and needs.

On modeling adaptive web-based instructional systems

Web-based instructional systems implement constructive learning models, in which the user has control over the learning stages, as well as over the building stages of its own cognitive structures. This model has obvious advantages, but it may fail for closed cognitive style users. To solve this problem many solutions have been attempted: extension of surfing functions offered by the instructional system, upgrading systems adaptive capabilities or conceptual organization of the teaching knowledge, with learning stages phasing. This paper presents and evaluates several solutions adopted in designing adaptive Web-based instructional environments.

Hierarchical Cluster Analysis: A reliable tool allowing more detailed (regional) traceability investigations

We are presenting an artificial intelligence application designed to perform a reliable recognition of the geographical origin of horticultural products. The system allows more detailed traceability investigations than those required by the present general and specific standards imposed by the European Community legislation. The classification is performed by using an unsupervised pattern recognition technique, i.e. Hierarchical Cluster Analysis. The efficiency of the system is illustrated for dill (Anethum gruveoles), which is one of the most popular spice in Europe. Dill is also used for its digestive, antispasmodic, anti-inflammatory, diuretic and antioxidant properties. The knowledge base includes physico-chemical information about dill samples originating from four neighboring regions of Romania and of the Republic of Moldova. The inference engine assigns the class identity (region of origin) based on agglomerative clustering. The results show that the system is a remarkably reliable tool for in-depth traceability investigations. It clearly discriminates dill samples originating from closely located regions, which are characterized by quite similar pedo-climatic conditions. The human-machine interface is user-friendly, allowing the system to be easily used even by non-specialists. The sensitivity of selectivity of the system is discussed in comparison with those obtained by using Principal Component Analysis.

Simultaneous regional traceability assessments based on Artificial Neural Networks

We are presenting an Artificial Neural Networks (ANN) application designed to perform detailed (regional) authenticity and traceability assessments in the case of herbal spices. Its capacity to correctly assign the class (regional) identity when the properties of a new sample are compared simultaneously with models built for several regions of origin has been evaluated. A case study performed for dill (Anethum graveolens) indicates that ANN is very fit for the purpose, the system providing efficient and cost-effective simultaneous regional traceability assessments.

A Hybrid Approach for Data Preprocessing in the QSAR Problem

One of the approaches in the Knowledge Discovery in Databases (KDD) domain is Predictive Toxicology (PT). Its aim is to discover and represent the relationships between the chemical structure of chemical compounds and biological and toxicological processes. The challenges in real toxicology problems are big amount of the chemical descriptors and imperfect data (means noisy, redundant, incomplete, and irrelevant). The main goals in knowledge discovery field are to detect these undesirable proprieties and to eliminate or correct them. This supposes noise reduction, data cleaning and feature selection because the performance of the applied Machine Learning algorithms is strongly related with the quality of the used data. In this paper, we present some of the issues that can be performed for preparing data before the knowledge discovery process begin.

Professor:e – An IMS Standard Based Adaptive E-learning Platform

Recently, the IMS Global Learning Consortium has provided specifications on modeling several aspects related to e-learning. Some of these aspects are separating the actual educational content from the structure of a learning unit, the learner design, the quiz testing, the learning scenario. Learning Simple Sequencing (LSS) is a scenario modeling principle, published in March 2003, meant to help e-learning platform developers to implement educational content delivery modules. Also, an extension of the Content Packaging XML manifest was provided in order to include the new standard.

Adaptive Learning Scenarios in Intelligent Instructional Environment

The possibility of personalising the educational process is the main demand in building computer-assisted learning systems for any form of education, continuous education and professional re - con- version included. In the case of these education forms, the subject of study frequently consists of groups of users; between these users there can be significant differences concerning the learning level, the age or the internal motives (objective, purpose) pursued through study. Con- sequently, the way the educational process is approached - cognitive style - can be extremely different among the users who form the study groups. When computer-assisted learning systems are used for any of these forms of education, they have to be able to dynamically adapt to the various cognitive necessities and demands of the users, in order to ensure the efficiency of the educational act.