Pattern recognition has become an accessible tool in developing advanced adaptive products. The need for such products is not diminishing but on the contrary, requirements for systems that are more and more aware of their environmental circumstances are constantly growing. Feed-forward neural networks are used to learn patterns in their training data without the need to discover by hand the relationships present in the data. However, the problem of estimating the required size of the neural network is still not solved. If we choose a neural network that is too small for a particular given task, the network is unable to "comprehend" the intricacies of the data. On the other hand if we choose a network size that is too big for the given task, we will observe that there are too many parameters to be tuned for the network, or we can fall in the "Curse of dimensionality" or even worse, the training algorithm can easily be trapped in local minima of the error surface. Therefore, we choose to investigate possible ways to find the 'Goldilocks' size for a feed-forward neural network (which is just right in some sense), being given a training set. Furthermore, we used a common paradigm used by the Roman Empire and employed on a wide scale in computer programming, which is the "Divide-et-Impera" approach, to divide a given dataset in multiple sub-datasets, solve the problem for each of the sub-dataset and fuse the results of all the sub-problems to form the result for the initial problem as a whole. To this effect we investigated modular neural networks and their performance.

The very large dimensionality of real world datasets is a challenging problem for classification algorithms, since often many features are redundant or irrelevant for classification. In addition, a very large number of features leads to a high computational time for classification algorithms. Feature selection methods are used to deal with the large dimensionality of data by selecting a relevant feature subset according to an evaluation criterion. The vast majority of research on feature selection involves conventional single-label classification problems, where each instance is assigned a single class label; but there has been growing research on more complex multi-label classification problems, where each instance can be assigned multiple class labels. This thesis proposes three types of new Multi-Label Correlation-based Feature Selection (ML-CFS) methods, namely: (a) methods based on hill-climbing search, (b) methods that exploit biological knowledge (still using hill-climbing search), and (c) methods based on genetic algorithms as the search method. Firstly, we proposed three versions of ML-CFS methods based on hill climbing search. In essence, these ML-CFS versions extend the original CFS method by extending the merit function (which evaluates candidate feature subsets) to the multi-label classification scenario, as well as modifying the merit function in other ways. A conventional search strategy, hill-climbing, was used to explore the space of candidate solutions (candidate feature subsets) for those three versions of ML-CFS. These ML-CFS versions are described in detail in Chapter 4. Secondly, in order to try to improve the performance of ML-CFS in cancer-related microarray gene expression datasets, we proposed three versions of the ML-CFS method that exploit biological knowledge. These ML-CFS versions are also based on hill-climbing search, but the merit function was modified in a way that favours the selection of genes (features) involved in pre-defined cancer-related pathways, as discussed in detail in Chapter 5. Lastly, we proposed two more sophisticated versions of ML-CFS based on Genetic Algorithms (rather than hill-climbing) as the search method. The first version of GA-based ML-CFS is based on a conventional single-objective GA, where there is only one objective to be optimized; while the second version of GA-based ML-CFS performs lexicographic multi-objective optimization, where there are two objectives to be optimized, as discussed in detail in Chapter 6. In this thesis, all proposed ML-CFS methods for multi-label classification problems were evaluated by measuring the predictive accuracies obtained by two well-known multi-label classification algorithms when using the selected featuresม namely: the Multi-Label K-Nearest neighbours (ML-kNN) algorithm and the Multi-Label Back Propagation Multi-Label Learning Neural Network (BPMLL) algorithm. In general, the results obtained by the best version of the proposed ML-CFS methods, namely a GA-based ML-CFS method, were competitive with the results of other multi-label feature selection methods and baseline approaches. More precisely, one of our GA-based methods achieved the second best predictive accuracy out of all methods being compared (both with ML-kNN and BPMLL used as classifiers), but there was no statistically significant difference between that GA-based ML-CFS and the best method in terms of predictive accuracy. In addition, in the experiment with ML-kNN (the most accurate) method selects about twice as many features as our GA-based ML-CFS; whilst in the experiments with BPMLL the most accurate method was a baseline method that does not perform any feature selection, and runs the classifier once (with all original features) for each of the many class labels, which is a very computationally expensive baseline approach. In summary, one of the proposed GA-based ML-CFS methods managed to achieve substantial data reduction, (selecting a smaller subset of relevant features) without a significant decrease in predictive accuracy with respect to the most accurate method.

It is widely accepted that public policies have an important role in driving science, technology and innovation (STI) initiatives in order to achieve socio-economic and development objectives. Nevertheless, previous research reveals that developing countries still face difficulties in formulating policies to support and promote STI. A possible reason for this is found in the apparent lack of capabilities for policymaking. Capabilities are "a precondition for effective policy formulation in developing countries" (UNIDO, 2005, p.16). However, our knowledge and understanding of what these capabilities are, remain limited. In this thesis I examine the roles that capabilities play in formulating STI policies, the development of these capabilities and their evolution over the years. I group policy capabilities into organisational capabilities – which refers to policy processes and routines – and individual capabilities – which refers to the skills of individual policymakers (Nelson and Winter, 1982; Dosi et al., 2000; Feldman and Pentland, 2003). In order to address the identified gaps in literature, I use the Nigerian Federal Ministry of Science and Technology (FMST) – which in 2012 completed the formulation of a new national STI policy – as an illustrative case for the investigation of these issues. To achieve the aim of the thesis, I address three research questions: (1) What roles do capabilities play in formulating STI policies at FMST and why? (2) How did policy formulation capabilities originally emerge at FMST and why? (3) How have policy formulation capabilities evolved (i.e. changed over the years, from 1986 to 2012) at FMST and why? To collect data, I interviewed key staff at FMST and stakeholder organisations (who participated in the STI policy formulation exercise), in addition to secondary data from relevant policy documents. The data analysis was based on the “explanation-building” technique (Yin, 2009). The findings reveal the various roles that policy capabilities (processes, routines and skills) play in policy formulation; how and why policy capabilities were developed and their evolution over the years at FMST. The results address the aforementioned gaps. The findings should be useful to policymakers, decision-makers and practitioners involved in STI policymaking, research and capability management.

Al Madinah, in Saudi Arabia, is the second most holy city for Muslims throughout the world and has a long and rich heritage. However, most of the historical and traditional buildings, city walls and holy places have been replaced with modern structures. But, there have been several attempts, many by individuals, to preserve the heritage of Al Madinah. This thesis took an in-depth look at the history of Al Madinah, with emphasis on a 3D virtual environment which was produced as part of this project and inspired by a 3D model depicting the historical city of Al Madinah. First, this research examined the documentation of the historical city and identified its limitations by visiting location museums and evaluating the display mediums concerned with the heritage of Al Madinah. To contrast the traditional methods employed in local museums, eight museums in the UK were visited to explore their use of technology and digital devices. After these two initial steps, the main contribution focused on developing an effective installation to present the heritage of Al Madinah using first hand material. The Madinah Virtual Heritage (MVH) installation was developed in two main stages and tested for its usability. MVH provides a virtual reality experience by using an affordable head-mounted VR display, which would be especially beneficial for local museums with limited budgets. This thesis gives an overview of how to create a virtual heritage environment, and the principles can be applied to other fields. The findings show that there are limited resources available to understand the heritage of Al Madinah, especially because local museums are self-funded and use traditional media and redundant displays. The use of 3D is a possible solution to reconstruct the demolished buildings. Virtual reality brings interactivity and engagement to the installation, which could be used in local museums as it is now available in head-mounted format at an affordable cost.