Peter Blanchfield

Web-page adaptation framework for PC & mobile device collaboration

Existing Web-content adaptation engines assume that the task being undertaken is the browsing of the Web-pages by a single-user. There are, however, other tasks that necessitate the viewing of Web-pages on mobile devices. One such task, which forms the motivation for this paper, is mobile Web-based collaboration involving the co-browsing of Web-pages from multiple devices. The paper begins by reviewing the considerations taken into account by existing Web-content adaptation engines in handling the single-user browsing task. Next, the differences between single-user browsing and co-browsing are discussed and a content adaptation framework proposed based on the concept of shared viewpoint and personal viewpoint. Finally, the effectiveness of the framework is demonstrated through a perceptual experiment.

Automated detection of obstructive apnea and hypopnea events from oxygen saturation signal

Our objective is to automate the detection of apnea and hypopnea events in obstructive sleep apnea hypopnea (OSAH) syndrome based on analysis of arterial oxygen saturation signal alone. This is the first attempt where wavelet is used to detect OSAH events. Detection of OSAH events through wavelet depends on the fluctuations in the magnitude of the transformed coefficients, thus circumventing the problem of variability in the criteria on the magnitude and duration of the signal. Our work evaluates the performance of the wavelet transform to detect OSAH events against three conventional amplitude and duration algorithms. High performance in the detection of OSAH events can be achieved through the wavelet algorithm (score 96.55%, sensitivity 95.74% and specificity 97.02%) if the threshold on wavelet coefficients is individually tuned for each study. However, this is impossible in clinical practice. It is interesting to observe that the conventional methods based on amplitude and duration are able to attain a performance as close as this. The Nervus algorithm obtains the best result (score 96.66%, sensitivity 95.26% and specificity 97.46%) compared to the amplitude duration algorithm, the drop duration algorithm and the wavelet algorithm with global threshold, in descending order of performance.

Handling diversity in evolutionary multiobjective optimization

In evolutionary multiobjective optimisation (EMO), the diversity of the set of non-dominated solutions used to be handled by the niching and fitness sharing technique. The main downside of this technique is the need to set the niche radius. Quite recently, new techniques have emerged and proved to be more successful. The grid-based density of the adaptive grid algorithm (AGA), the crowding-distance technique of the nondominated sorting genetic algorithm (NSGA-II), and the archive truncation procedure of the strength Pareto evolutionary algorithm (SPEA2) are the latest successful methods that ensure a better diversity than the traditional less effective and computationally expensive niching method. In this work, a crowding-dispersion technique which is based on the Pareto potential regions (PPR), is proposed and compared to three recent techniques.

Involving the user with low or no vision in the design of an audio-haptic learning environment for learning about 3D shapes: The first approach

This report summarises the current research that is being carried out into haptic sense technology for assisting visual impairment. Moreover, it presents a framework of mechanisms blind students apply to identify basic 3D objects into free space. People with visual impairments are still excluded from accessing certain types of information - like graphics - that are easily accessible by the general public. Today, screen reading software and Braille displays or text-to-speech systems have solved many problems of accessing text. For accessing graphics, and specifically digital graphics, there is to date no standardised technology that is accepted and used by blind people. Especially in school education, access to pictures is almost impossible to master; making teamwork very complicated and thus isolating blind students from their sighted peers making it impossible to acquire a shared understanding of school material. Focusing on geometry, which is believed to be the main theory of space, is very crucial for non-sighted students to develop an accurate spatial sense and reasoning. The emergence of haptic technology and the potential to create interfaces for non-visual audio-haptic interaction opens new promising doors to accessing digital graphics.

ERiSA: building emotionally realistic social game-agents companions

We propose an integrated framework for social and emotional game-agents to enhance their believability and quality of interaction, in particular by allowing an agent to forge social relations and make appropriate use of social signals. The framework is modular including sensing, interpretation, behaviour generation, and game components. We propose a generic formulation of action selection rules based on observed social and emotional signals, the agent’s personality, and the social relation between agent and player. The rules are formulated such that its variables can easily be obtained from real data. We illustrate and evaluate our framework using a simple social game called The Smile Game.

A Hybrid Data Mining and Case-Based Reasoning User Modeling System (HDCU) for Monitoring and Predicting of Blood Sugar Level

In this paper we present HDCU, a hybrid data mining and case-based reasoning user modeling system, which is used to monitor and predict the blood sugar level in diabetics. The practical objective for this project is to reduce the cost of direct blood sugar self monitoring by minimizing the number of times that a diabetic needs to measure his or her sugar levels every day. From the technological point of view, the main aim is using the support vector machine as the classifier and implementing a case-based reasoning cycle as the retrieval cycle in order to indirectly determine and predict blood sugar level in diabetics and finally implement this software into a mobile device with wireless sensor networks and link it to a server which houses the relevant knowledgebase.

Playing with Social and Emotional Game Companions

This paper presents the findings of an empirical study that explores player game experience by implementing the ERiSA Framework in games. A study with Action Role-Playing Game (RPG) was designed to evaluate player interactions with game companions, who were imbued with social and emotional skill by the ERiSA Framework. Players had to complete a quest in the Skyrim game, in which players had to use social and emotional skills to obtain a sword. The results clearly show that game companions who are capable of perceiving and exhibit emotions, are perceived to have personality and can forge relationships with the players, enhancing the player experience during the game.

Simulation of the Generation and Processing of Doppler Ultrasound Fetal Heart Signals to obtain Directional Motion Information

In fetal heart monitoring using Doppler ultrasound signals the cardiac information is commonly extracted from non-directional signals. As a consequence often some of the cardiac events cannot be observed clearly which may lead to the incorrect detection of the valve and wall motions. Here, directional signals were simulated to investigate their enhancement of cardiac events, and hence provide clearer information regarding the cardiac activities. First, fetal Doppler ultrasound signals were simulated with signals encoding forward and reverse motion then obtained using a pilot frequency. The simulation results demonstrate that the model has the ability to produce realistic Doppler ultrasound signals and a pilot frequency can be used in the mixing process to produce directional signals that allow the simulated cardiac events to be distinguished clearly and correctly

Study on the Correlation Coefficient of Gene Expression by using a Hybrid Intelligent System

Normal Lung and carcinoid are high relative classes in our “Detection and Prediction of Lung Cancer using the zNose with the Support Vector Machine Classifier” project. The mRNA expression level of these two classes was analyzed by using oligonucleotide microarrays. The correlation coefficient measurement results referred to the 20 subclasses (mRNA expression) of normal lung and carcinoid, which were collected from a total of 203 specimens (186 snap-frozen lung tumors and 17 normal lungs). The distinct subclasses (mRNA expressions) are 31687_f_at hemoglobin (β), 31525_s_at hemoglobin (α2), and 31481_s_at thymosin (β10). The Correlation Coefficient reflected the results at 0.8702, 0.8935 and 0.9105 respectively (SMOreg PolyKernel -E 1.0). This study also showed the best prediction class was the first level class which was reflected from the correlation coefficient, recorded at 0.9409. This result was further verified by the prediction capacity of our proposed system. General Terms Hybrid Intelligent System, Lung Cancer Detection System, Data Mining and Artificial Intelligence Classifiers.

Integrity Check for Printed Binary Document Images

Following the digital revolution, many documents have been converted to the digital format and saved in computers. Documents are usually printed using black and white printers when converted to hardcopy documents. The printed documents may be scanned to computers as a binary text images. However, hard copy documents are subject to forgery. Also, a scanned document cannot be an exact replica of its original version because of the additional noise caused by D.A conversion. Therefore, it has been essential to check the integrity of scanned documents to detect and locate alternations in forged documents and validate unaltered scanned images. It is unreasonable to keep records of original copies for each printed document for authentication purposes. Thus, it is better to embed information about the document contents into the document itself. In this paper, a self-validating binary document images to validate hardcopy black and white text image.