Information2015, 6(3), 564-575; doi:10.3390/info6030564 - published 1 September 2015 Show/Hide Abstract
Abstract: An adapted ant colony algorithm is proposed to adapt e-content to learner’s profile. The pertinence of proposed units keeps learners motivated. A model of categorization of course’s units is presented. Two learning paths are discussed based on a predefined graph. In addition, the ant algorithm is simulated on the proposed model. The adapted algorithm requires a definition of a new pheromone which is a parameter responsible for defining whether the unit is in the right pedagogical sequence or in the wrong one. Moreover, it influences the calculation of quantity of pheromone deposited on each arc. Accordingly, results show that there are positive differences in learner’s passages to propose the suitable units depending on the sequence and the number of successes. The proposed units do not depend on the change of number of units around 10 to 30 units in the algorithm process.
Information2015, 6(3), 550-563; doi:10.3390/info6030550 - published 27 August 2015 Show/Hide Abstract
Abstract: A software product line is a complex system the aim of which is to provide a platform dedicated to large reuse. It necessitates a great investment. Thus, its ability to cope with customers’ ever-changing requirements is among its key success factors. Great effort has been made to deal with the software product line evolution. In our previous works, we carried out a classification of these works to provide an overview of the used techniques. We also identified the following key challenges of software product lines evolution: the ability to predict future changes, the ability to define the impact of a change easily and the improvement in understanding the change. We have already tackled the second and the third challenges. The objective of this paper is to deal with the first challenge. We use the cladistics classification which was used in biology to understand the evolution of organisms sharing the same ancestor and their process of descent at the aim of predicting their future changes. By analogy, we consider a population of applications for media management on mobile devices derived from the same platform and we use cladistics to construct their evolutionary tree. We conducted an analysis to show how to identify the evolution trends of the case study products and to predict future changes.
Information2015, 6(3), 536-549; doi:10.3390/info6030536 - published 24 August 2015 Show/Hide Abstract
Abstract: With the rapid development of M2M wireless network, damages caused by malicious worms are getting more and more serious. The main goal of this paper is to explore the influences of removable devices on the interaction dynamics between malicious worms and benign worms by using a mathematical model. The model takes two important network environment factors into consideration: benign worms and the influences of removable devices. Besides, the model’s basic reproduction number is obtained, along with the correct control conditions of the local and global asymptotical stability of the worm-free equilibrium. Simulation results show that the effectiveness of our proposed model in terms of reflecting the influences of removable devices on the interaction dynamics of an anti-treat model. Based on numerical analyses and simulations, effective methods are proposed to contain the propagation of malicious worms by using anti-worms.
Information2015, 6(3), 522-535; doi:10.3390/info6030522 - published 21 August 2015 Show/Hide Abstract
Abstract: The collection of massive Global Positioning System (GPS) data from travel surveys has increased exponentially worldwide since the 1990s. A number of methods, which range from rule-based to advanced classification approaches, have been applied to detect travel modes from GPS positioning data collected in travel surveys based on GPS-enabled smartphones or dedicated GPS devices. Among these approaches, neural networks (NNs) are widely adopted because they can extract subtle information from training data that cannot be directly obtained by human or other analysis techniques. However, traditional NNs, which are generally trained by back-propagation algorithms, are likely to be trapped in local optimum. Therefore, particle swarm optimization (PSO) is introduced to train the NNs. The resulting PSO-NNs are employed to distinguish among four travel modes (walk, bike, bus, and car) with GPS positioning data collected through a smartphone-based travel survey. As a result, 95.81% of samples are correctly flagged for the training set, while 94.44% are correctly identified for the test set. Results from this study indicate that smartphone-based travel surveys provide an opportunity to supplement traditional travel surveys.
Information2015, 6(3), 505-521; doi:10.3390/info6030505 - published 20 August 2015 Show/Hide Abstract
Abstract: Feature extraction and classification are two key steps for activity recognition in a smart home environment. In this work, we used three methods for feature extraction: Principal Component Analysis (PCA), Independent Component Analysis (ICA), and Linear Discriminant Analysis (LDA). The new features selected by each method are then used as the inputs for a Weighted Support Vector Machines (WSVM) classifier. This classifier is used to handle the problem ofimbalanced activity data from the sensor readings. The experiments were implemented on multiple real-world datasets with Conditional Random Fields (CRF), standard Support Vector Machines (SVM), Weighted SVM, and combined methods PCA+WSVM, ICA+WSVM, and LDA+WSVM showed that LDA+WSVM had a higher recognition rate than other methods for activity recognition.
Information2015, 6(3), 494-504; doi:10.3390/info6030494 - published 14 August 2015 Show/Hide Abstract
Abstract: This paper presents a new biometric score fusion approach in an identification system using the upper integral with respect to Sugeno’s fuzzy measure. First, the proposed method considers each individual matcher as a fuzzy set in order to handle uncertainty and imperfection in matching scores. Then, the corresponding fuzzy entropy estimates the reliability of the information provided by each biometric matcher. Next, the fuzzy densities are generated based on rank information and training accuracy. Finally, the results are aggregated using the upper fuzzy integral. Experimental results compared with other fusion methods demonstrate the good performance of the proposed approach.