Massey Documents by Type
Permanent URI for this communityhttps://mro.massey.ac.nz/handle/10179/294
Browse
12 results
Search Results
Item The acquisition of knowledge from multiple experts in the domain of sensory evaluation panel training : a thesis presented in partial fulfilment of the requirements for the degree of Master of Philosophy in Social Science (Computer Science) at Massey University, New Zealand(Massey University, 1994) Ewing-Jarvie, SimonKnowledge acquisition is the elicitation and representation of human expertise and is one of the first steps taken in constructing an expert system. It has often been cited as the 'bottleneck' in expert systems development due to the labour intensive processes needed to deal with the expert human. Various researchers have proposed methodologies for improving both the accuracy and the productivity of the process. This has ranged from manual to automated methods as well as examining what the expert might be thinking during a study of the conscious activity. This research has focused on the issues involved in the manual elicitation of knowledge using multiple experts in the same domain. It utilises the transcripts of semi-structured interviews and discourse analysis techniques to construct the domain layer of a knowledge base, employing the KADS methodology. The findings highlight the importance of the relationship between the knowledge engineer, organisation and the human experts. Issues such as motivation, organisational commitment and communication skills feature as key indicators of the likely success of an expert system development project. While automated acquisition assists with productivity, it works against the development of relationships within the project team and the trade-off must be carefully considered by the project manager.Item The AIM methodology: analysis interpretation and modelling : a methodology for the role of the knowledge analyst in knowledge acquisition : a thesis presented in partial fulfilment of the requirements for the degree of Master of Science in Computer Science at Massey University(Massey University, 1994) da Silva, Ana Estela AntunesA methodology for the analysis, interpretation and modelling of knowledge is developed in this thesis - the AIM methodology. This methodology can be used by a knowledge engineer who has had no involvement with the elicitation process. The knowledge engineer who carries out the processes of the AIM methodology is called the knowledge analyst. As a general description, the AIM methodology presents techniques together with their advantages and pitfalls in the processes of analysis, interpretation and modelling. The process of analysis consists of the refinement of the elicited knowledge. During this process, techniques such as the discourse analysis are used. The interpretation and modelling processes aim at representing the domain elements obtained during the analysis process. Models such as the model of expertise of the KADS methodology are used during the interpretation and modelling processes. This research also aims at defining the activities involved in the knowledge acquisition phase in the development of knowledge-based systems. In this thesis, knowledge acquisition is considered as a set of processes which comprises not only the elicitation of knowledge, but also the refinement of the knowledge elicited through its analysis, interpretation and modelling.Item Study and design of security system for AudioGraph multimedia teaching system : a thesis presented in partial fulfillment of the requirements for the degree of Master of Science in Computer Science at Massey University, Palmerston North, New Zealand(Massey University, 2002) Zhang, ShaohuaiThe AudioGraph is a project developed in the Institute of Information Sciences and Technology of Massey University as copyrighted software tools. The AudioGraph courseware files produced by the AudioGraph Recorder may be copyrighted. In order to protect them from being played back or copied by unauthorized parties, a security system to protect the AudioGraph courseware files is required. This thesis presents the study, design and implementation of the AudioGraph security system. The security system proposed in this thesis consists of three parts: Copy Protection Record inside the AudioGraph courseware files; a Key Insertion Tool to detect, extract, insert and update the Copy Protection Record in AudioGraph courseware files; and a scheme of usage control embedded into the AudioGraph Plug-in. The issues covered in this thesis include all relevant aspects. In order to select good encryption algorithms for the AudioGraph security system, this thesis introduces the concept of cryptography and describes some of the most important conventional and public-key encryption algorithms. It also investigates and compares various aspects of some of the conventional cryptography algorithms and chooses very strong, simple and suitable encryption algorithms to be used in the AudioGraph security system. A scheme to protect AudioGraph courseware files is described in this thesis, this scheme meets the requirements of the AudioGraph security system, and it is strong enough to withstand brute-force attack and all known cryptanalysis. The implementation of the AudioGraph security system has also been described in this thesis. The result from system testing demonstrates that this AudioGraph security system works well and had achieved its goal to protect the AudioGraph courseware material.Item Associative access in persistent object stores : a thesis presented in partial fulfilment of the requirements for the degree of Master of Information Sciences in Information Systems at Massey University(Massey University, 2004) Nusdin, WeenaThe overall aim of the thesis is to study associative access in a Persistent Object Store (POS) providing necessary object storage and retrieval capabilities to an Object Oriented Database System (OODBS) (Delis, Kanitkar & Kollios, 1998 cited in Kirchberg & Tretiakov, 2002). Associative access in an OODBS often includes navigational access to referenced or referencing objects of the object being accessed (Kim. Kim. & Dale. 1989). The thesis reviews several existing approaches proposed to support associative and navigational access in an OODBS. It was found that the existing approaches proposed for associative access could not perform well when queries involve multiple paths or inheritance hierarchies. The thesis studies how associative access can be supported in a POS regardless of paths or inheritance hierarchies involved with a query. The thesis proposes extensions to a model of a POS such that approaches that are proposed for navigational access can be used to support associative access in the extended POS. The extensions include (1) approaches to cluster storage objects in a POS on their storage classes or values of attributes, and (2) approaches to distinguish references between storage objects in a POS based on criteria such as reference types - inheritance and association, storage classes of referenced storage objects or referencing storage objects, and reference names. The thesis implements Matrix-Index Coding (MIC) approach with the extended POS by several coding techniques. The implementation demonstrates that (1) a model of a POS extended by proposed extensions is capable of supporting associative access in an OODBS and (2) the MIC implemented with the extended POS can support a query that requires associative access in an OODBS and involves multiple paths or inheritance hierarchies. The implementation also provides proof of the concepts suggested by Kirchberg & Tretiakov (2002) that (1) the MIC can be made independent from a coding technique, and (2) data compression techniques should be considered as appropriate alternatives to implement the MIC because they could reduce the storage size required.Item Detecting live person for the face recognition problem : submitted in partial fulfilment of the requirements for the degree of Master of Information Sciences, Massey University(Massey University, 2016) Alrashed, H HFace recognition has been a challenging problem for computer vision scientists for the last few decades. Hence it was the center of attention for computer vision researchers. The purpose of this research is to improve the security of the face recognition system by identifying the liveness of a person in front of a camera to be recognised. The objective was to detect if the images used to be recognised reflect a real person’s face, i.e., a live person’s face instead of just a static image of the face. This can be achieved by randomly asking the person to carry out certain tasks. Simple tasks such as blinking an eye or smiling can then be repeated randomly according to the instructions given by the new system, so even a video of the target face made previously would not be able to perform the authentication easily. Each component of the system were tested separately. The accuracy of the face detection component was impressively at 98.93%. The eye blinking detection uses a new proposed method with a high accuracy of 91%. Face recognition component was also tested and had a high recognition rate of 96%. Keywords: Face Recognition, Face Detection, Eigenfaces, OpenCV, Face Anti-Spoofing, Eye Detection, Smile Detection, Eye Blinking DetectionItem Conceptual data modelling for geographical information systems : a thesis presented in partial fulfilment of the requirements for the degree of Master of Arts (Information Systems) at Massey University(Massey University, 1994) Bekesi, ErzsebetThis thesis sets out to find an answer to the question: does an appropriate conceptual data model exist for the practitioners of Geographical Information Systems database design? It aims to investigate and answer the question by: • Finding a workable data model to solve a database design problem (Manawatu-Wanganui Regional Council, Palmerston North, Natural Resources Management, Groundwater Section database). • Analysing the user's data requirements and producing a feasible conceptual schema. Usage of Geographical Information Systems applications is a recognised need in a growing number of organisations in New Zealand, but many factors block the way of this relatively new technology. One of these factors is the lack of well-designed databases to support the data needs of these non-traditional applications. One school of thought adopts general data modelling techniques for every database design problem, another group of researchers suggests that specialised data models are necessary to model data in various problem domains. This thesis summarises the "specialities" pertaining to the GIS database domain. The most important are the special data needs of GIS applications and the problem of the placement of spatial data models in the traditional taxonomy of database models. It chooses the objectives of conceptual data modelling as the evaluation criteria which the selected data model must satisfy i.e. to model reality and to form the basis for database schema design. This thesis reviews a group of published papers, selected from proponents of the entity-relationship and the object-oriented data modelling paradigms and the applications of these data modelling techniques in a spatial context. It compares various extensions to the original entity relationship model, and a comparison of the main data modelling paradigms is included. Data modelling shortcomings encountered in the literature are also summarised. The literature reviewed concludes that not appreciating the conceptual data modelling objectives leads to unsatisfactory conceptual database design. The selected data model, the spatially extended entity relationship (SEER) model is described and applied to the database design problem of a local authority to produce conceptual schemas. Findings are summarised and issues for future research are identified. Conclusions reached are: further evaluative work on the applied spatially extended entity relationship (SEER) model would be useful and clear directions are essential for practitioners showing the guiding principles of conceptual data modelling in a spatial context.Item Computational complexity of elitist population-based evolutionary algorithms : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Computer Science at Massey University, Palmerston North, New Zealand(Massey University, 2012) Ter-Sarkisov, AramEvolutionary Algorithms (EAs) are a modern heuristic algorithm that have proven efficiency on a large number of real-life problems. Despite the rich history of applications understanding of both how and why EAs work is lagging far behind. This is especially true for one of the main components of EAs, that is hypothesized by many to underlie their efficiency: population. The first problem considered in this thesis is the introduction of a recombination operator, K-Bit-Swap (KBS) and its comparison to mainstream operators, such as mutation and different types of crossover. A vast amount of statistical evidence is presented that shows that EAs using KBS outperform other algorithms on a whole range of problems. Two problems are selected for a deep theoretical analysis: OneMax and Royal Roads. The main problem of modeling EAs that use both population and a pool of parents is the complexity of the structures that arise from the process of evolution. In most cases either one type of species is considered or certain simple assumptions are made about fitness of the species. The main contribution of this thesis is the development of a new approach to modeling of EAs that is based on approximating the structure of the population and the evolution of subsets thereof. This approach lies at the core of the new tool presented here, the Elitism Levels Traverse Mechanism that was used to derive upper bounds on the runtime of EAs. In addition, lower bounds were found using simpler assumptions of the underlying distribution of species in the population.The second important result of the approach is the derivation of limiting distributions of a subset of the population, a problem well-known in areas such as epidemiology. To the best of the author's knowledge, no such findings have been published in the EA community so far.Item The development of a decision support system for energy cost management, using an expert system shell : a case study in the integrated use of software packages : report presented in fulfilment of the thesis requirements for the degree of Master of Technology (Computing Technology)(Massey University, 1989) Robertson, L JThe theory of minimizing total energy usage is well known (if not well documented), and the techniques are widely practiced. Because of the way in which energy is sold, the total cost incurred may be affected even more strongly by the time-distribution of the energy usage, than by the total quantity used. A major subject of ECM is the MANAGEMENT of this time-distribution of energy usage, with the objective of minimizing of total energy costs to the user. A software package (named ECMES, Energy Cost Management Expert System) has been developed using the Lotus Symphony integrated spreadsheet software package. The ECMES application consists (currently) of three modules offering analyses of several aspects of electrical energy cost management (plus three corresponding modules for gas costs, which are not considered further). The Symphony ECM application modules have been developed over the last few years, largely on a spare time basis, by Professor W Monteith of Massey University's Production Technology Department. The analysis of Energy Cost Management on a PC is one which requires functions supplied by several standard software packages, particularly spreadsheet, graphics, database and expert system. The relatively recent availability of moderately priced and user-friendly expert system development packages has brought an additional set of powerful tools within the reach of the application developer. A Decision Support System (using an Expert System shell) has been developed, which is well integrated with the spreadsheet data, and with a database, to expand the functions of the original spreadsheet ECM analysis tool. Theoretical work on the data requirements and the production rules has opened up possibilities for future work.Item An ant colony simulator : a thesis presented in partial fulfilment of the requirements for the degree of Master of Science in Computer Science at Massey University, Albany, New Zealand(Massey University, 2012) Ren, DelanIn recent years, ant colony algorithms have become more popular research topics in the artificial intelligence area of computer science. This biological modeling algorithm simulates the natural behavior of an ant colony looking for food among the insect kingdom. This algorithm was initially proposed by Marco Dorigo in 1992 in his PhD thesis – the first algorithm was aiming to search for an optimal path in a graph based on the behavior of ants seeking a path between their colony and a source of food. The original idea has since been diversified to solve a wider class of numerical problems, and as a result, several problems have emerged, drawing on various aspects of the behavior of ants (Ant colony optimization, 2010). The famous science journal “Nature” has published articles relating to ant colony algorithms several times, and lots of other publishers around the world have produced quite a few books for ant colony optimization. These days, ant colony algorithms have become a hot topic for the international artificial intelligence computing. The biological modeling optimization algorithm is an important branch in the artificial intelligence research area, which includes simulation biosphere natural selection and heredity mechanism genetic algorithm (Duan, 2005). This thesis continues research on the original ant colony algorithm, and creates a simulator to handle the ant colony’s natural behavior to find food.Item A study of frequent pattern mining in transaction datasets : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Computer Science at Massey University, Palmerston North, New Zealand(Massey University, 2011) Xu, LuofengWithin data mining, the efficient discovery of frequent patterns—sets of items that occur together in a dataset—is an important task, particularly in transaction datasets. This thesis develops effective and efficient algorithms for frequent pattern mining, and considers the related problem of how to learn, and utilise, the characteristics of the particular datasets being investigated. The first problem considered is how to mine frequent closed patterns in dynamic datasets, where updates to the dataset are performed. The standard approach to this problem is to use a standard pattern mining algorithm and simply rerun it on the updated dataset. An alternative method is proposed in this thesis that is significantly more efficient provided that the size of the updates is relatively small. Following this is an investigation of the pattern support distribution of transaction datasets, which measures the numbers of times each pattern appears within the dataset. The evidence for the pattern support distribution of real retail datasets obeying a power law is investigated using qualitative appraisals and statistical goodness-of-fit tests, and the power law is found to be a good model. Based on this, the thesis demonstrates how to efficiently estimate the pattern support distribution based on sampling techniques, reducing the computational cost of finding this distribution. The last major contribution of the thesis is to consider novel ways to set the main user-specified parameters of frequent pattern mining, the minimum support, which defines how many times a pattern needs to be seen before it is ‘frequent’. This is a critical parameter, and very hard to set without a lot of knowledge of the dataset. A method to enable the user to specify rather looser requirements for what they require from the mining is proposed based on the assumption of a power-law-based pattern support distribution and fuzzy logic techniques.