A quantitative view on fuzzy numbers

Since its introduction, fuzzy set theory has become a useful tool in the mathematical modelling of problems in Operations Research and many other fields. The number of applications is growing continuously. In this thesis we investigate a special type of fuzzy set, namely fuzzy numbers. Fuzzy numbers (which will be considered in the thesis as possibility distributions) have been widely used in quantitative analysis in recent decades. In this work two measures of interactivity are defined for fuzzy numbers, the possibilistic correlation and correlation ratio. We focus on both the theoretical and practical applications of these new indices. The approach is based on the level-sets of the fuzzy numbers and on the concept of the joint distribution of marginal possibility distributions. The measures possess similar properties to the corresponding probabilistic correlation and correlation ratio. The connections to real life decision making problems are emphasized focusing on the financial applications. We extend the definitions of possibilistic mean value, variance, covariance and correlation to quasi fuzzy numbers and prove necessary and sufficient conditions for the finiteness of possibilistic mean value and variance. The connection between the concepts of probabilistic and possibilistic correlation is investigated using an exponential distribution. The use of fuzzy numbers in practical applications is demonstrated by the Fuzzy Pay-Off method. This model for real option valuation is based on findings from earlier real option valuation models. We illustrate the use of number of different types of fuzzy numbers and mean value concepts with the method and provide a real life application.

Bankruptcy prediction using fuzzy methods

In this thesis, a classi cation problem in predicting credit worthiness of a customer is tackled. This is done by proposing a reliable classi cation procedure on a given data set. The aim of this thesis is to design a model that gives the best classi cation accuracy to e ectively predict bankruptcy. FRPCA techniques proposed by Yang and Wang have been preferred since they are tolerant to certain type of noise in the data. These include FRPCA1, FRPCA2 and FRPCA3 from which the best method is chosen. Two di erent approaches are used at the classi cation stage: Similarity classi er and FKNN classi er. Algorithms are tested with Australian credit card screening data set. Results obtained indicate a mean classi cation accuracy of 83.22% using FRPCA1 with similarity classi- er. The FKNN approach yields a mean classi cation accuracy of 85.93% when used with FRPCA2, making it a better method for the suitable choices of the number of nearest neighbors and fuzziness parameters. Details on the calibration of the fuzziness parameter and other parameters associated with the similarity classi er are discussed.

Feature selection using Fuzzy Entropy measures with Yu's Similarity measure

In this study, feature selection in classification based problems is highlighted. The role of feature selection methods is to select important features by discarding redundant and irrelevant features in the data set, we investigated this case by using fuzzy entropy measures. We developed fuzzy entropy based feature selection method using Yu's similarity and test this using similarity classifier. As the similarity classifier we used Yu's similarity, we tested our similarity on the real world data set which is dermatological data set. By performing feature selection based on fuzzy entropy measures before classification on our data set the empirical results were very promising, the highest classification accuracy of 98.83% was achieved when testing our similarity measure to the data set. The achieved results were then compared with some other results previously obtained using different similarity classifiers, the obtained results show better accuracy than the one achieved before. The used methods helped to reduce the dimensionality of the used data set, to speed up the computation time of a learning algorithm and therefore have simplified the classification task

"Fuzzy front end" ja sen menestyksekäs johtaminen teollisuusyrityksessä

Työssä käsitellään innovaatioprosessin ensimmäistä ”fuzzy front end” -vaihetta, jota työssä kutsutaan front end -vaiheeksi. Front end -vaihe on innovaatioprosessin alustava tutkimus ja suunnittelu vaihe ennen teknistä kehittämisvaihetta. Front end -vaihetta on tutkittu innovaatioprosessin osista vähiten, sekä se on useimmille yrityksillä sumea ja vaikeasti käsitettävä. Tutkimusten mukaan front end -vaiheen osaaminen on kuitenkin erittäin merkittävä tekijä yrityksen innovatiivisuudelle. Työssä avataan innovaatioprosessin sisältöä ja tavoitteita, sekä vertaillaan käytössä olevia malleja front end -vaiheen rakenteesta. Työssä selvitetään avaintekijöitä front end -vaiheen menestykseen ja tehokkuuteen. Lisäksi käsitellään johtamisen tekijöitä, jotka edesauttavat onnistumaan front end -vaiheessa.

Railways, Airports and Sea Container Operators as Publicly Listed Companies – Financial Performance and Shareholder Value Creation Perspective

Logistics infrastructure and transportation services have been the liability of countries and governments for decades, or these have been under strict regulation policies. One of the first branches opened for competition in EU as well as in other continents, has been air transports (operators, like passenger and freight) and road transports. These have resulted on lower costs, better connectivity and in most of the cases higher service quality. However, quite large amount of other logistics related activities are still directly (or indirectly) under governmental influence, e.g. railway infrastructure, road infrastructure, railway operations, airports, and sea ports. Due to the globalization, governmental influence is not that necessary in this sector, since transportation needs have increased with much more significant phase as compared to economic growth. Also freight transportation needs do not correlate with passenger side, due to the reason that only small number of areas in the world have specialized in the production of particular goods. Therefore, in number of cases public-private partnership, or even privately owned companies operating in these sub-branches have been identified as beneficial for countries, customers and further economic growth. The objective of this research work is to shed more light on these kinds of experiments, especially in the relatively unknown sub-branches of logistics like railways, airports and sea container transports. In this research work we have selected companies having public listed status in some stock exchange, and have needed amount of financial scale to be considered as serious company rather than start-up phase venture. Our research results show that railways and airports usually need high fixed investments, but have showed in the last five years generally good financial performance, both in terms of profitability and cash flow. In contrary to common belief of prosperity in globally growing container transports, sea vessel operators of containers have not shown that impressive financial performance. Generally margins in this business are thin, and profitability has been sacrificed in front of high growth – this also concerns cash flow performance, which has been lower too. However, as we examine these three logistics sub-branches through shareholder value development angle during time period of 2002-2007, we were surprised to find out that all of these three have outperformed general stock market indexes in this period. More surprising is the result that financially a bit less performing sea container transportation sector shows highest shareholder value gain in the examination period. Thus, it should be remembered that provided analysis shows only limited picture, since e.g. dividends were not taken into consideration in this research work. Therefore, e.g. US railway operators have disadvantage to other in the analysis, since they have been able to provide dividends for shareholders in long period of time. Based on this research work we argue that investment on transportation/logistics sector seems to be safe alternative, which yields with relatively low risk high gain. Although global economy would face smaller growth period, this sector seems to provide opportunities in more demanding situation as well.

Semantic and fuzzy modelling for human behaviour recognition in smart spaces : a case study on ambient assisted living.

Human activity recognition in everyday environments is a critical, but challenging task in Ambient Intelligence applications to achieve proper Ambient Assisted Living, and key challenges still remain to be dealt with to realize robust methods. One of the major limitations of the Ambient Intelligence systems today is the lack of semantic models of those activities on the environment, so that the system can recognize the speci c activity being performed by the user(s) and act accordingly. In this context, this thesis addresses the general problem of knowledge representation in Smart Spaces. The main objective is to develop knowledge-based models, equipped with semantics to learn, infer and monitor human behaviours in Smart Spaces. Moreover, it is easy to recognize that some aspects of this problem have a high degree of uncertainty, and therefore, the developed models must be equipped with mechanisms to manage this type of information. A fuzzy ontology and a semantic hybrid system are presented to allow modelling and recognition of a set of complex real-life scenarios where vagueness and uncertainty are inherent to the human nature of the users that perform it. The handling of uncertain, incomplete and vague data (i.e., missing sensor readings and activity execution variations, since human behaviour is non-deterministic) is approached for the rst time through a fuzzy ontology validated on real-time settings within a hybrid data-driven and knowledgebased architecture. The semantics of activities, sub-activities and real-time object interaction are taken into consideration. The proposed framework consists of two main modules: the low-level sub-activity recognizer and the high-level activity recognizer. The rst module detects sub-activities (i.e., actions or basic activities) that take input data directly from a depth sensor (Kinect). The main contribution of this thesis tackles the second component of the hybrid system, which lays on top of the previous one, in a superior level of abstraction, and acquires the input data from the rst module's output, and executes ontological inference to provide users, activities and their in uence in the environment, with semantics. This component is thus knowledge-based, and a fuzzy ontology was designed to model the high-level activities. Since activity recognition requires context-awareness and the ability to discriminate among activities in di erent environments, the semantic framework allows for modelling common-sense knowledge in the form of a rule-based system that supports expressions close to natural language in the form of fuzzy linguistic labels. The framework advantages have been evaluated with a challenging and new public dataset, CAD-120, achieving an accuracy of 90.1% and 91.1% respectively for low and high-level activities. This entails an improvement over both, entirely data-driven approaches, and merely ontology-based approaches. As an added value, for the system to be su ciently simple and exible to be managed by non-expert users, and thus, facilitate the transfer of research to industry, a development framework composed by a programming toolbox, a hybrid crisp and fuzzy architecture, and graphical models to represent and con gure human behaviour in Smart Spaces, were developed in order to provide the framework with more usability in the nal application. As a result, human behaviour recognition can help assisting people with special needs such as in healthcare, independent elderly living, in remote rehabilitation monitoring, industrial process guideline control, and many other cases. This thesis shows use cases in these areas.

Differential evolution based classification with pool of distances and aggregation operators

The objective of this thesis is to develop and generalize further the differential evolution based data classification method. For many years, evolutionary algorithms have been successfully applied to many classification tasks. Evolution algorithms are population based, stochastic search algorithms that mimic natural selection and genetics. Differential evolution is an evolutionary algorithm that has gained popularity because of its simplicity and good observed performance. In this thesis a differential evolution classifier with pool of distances is proposed, demonstrated and initially evaluated. The differential evolution classifier is a nearest prototype vector based classifier that applies a global optimization algorithm, differential evolution, to determine the optimal values for all free parameters of the classifier model during the training phase of the classifier. The differential evolution classifier applies the individually optimized distance measure for each new data set to be classified is generalized to cover a pool of distances. Instead of optimizing a single distance measure for the given data set, the selection of the optimal distance measure from a predefined pool of alternative measures is attempted systematically and automatically. Furthermore, instead of only selecting the optimal distance measure from a set of alternatives, an attempt is made to optimize the values of the possible control parameters related with the selected distance measure. Specifically, a pool of alternative distance measures is first created and then the differential evolution algorithm is applied to select the optimal distance measure that yields the highest classification accuracy with the current data. After determining the optimal distance measures for the given data set together with their optimal parameters, all determined distance measures are aggregated to form a single total distance measure. The total distance measure is applied to the final classification decisions. The actual classification process is still based on the nearest prototype vector principle; a sample belongs to the class represented by the nearest prototype vector when measured with the optimized total distance measure. During the training process the differential evolution algorithm determines the optimal class vectors, selects optimal distance metrics, and determines the optimal values for the free parameters of each selected distance measure. The results obtained with the above method confirm that the choice of distance measure is one of the most crucial factors for obtaining higher classification accuracy. The results also demonstrate that it is possible to build a classifier that is able to select the optimal distance measure for the given data set automatically and systematically. After finding optimal distance measures together with optimal parameters from the particular distance measure results are then aggregated to form a total distance, which will be used to form the deviation between the class vectors and samples and thus classify the samples. This thesis also discusses two types of aggregation operators, namely, ordered weighted averaging (OWA) based multi-distances and generalized ordered weighted averaging (GOWA). These aggregation operators were applied in this work to the aggregation of the normalized distance values. The results demonstrate that a proper combination of aggregation operator and weight generation scheme play an important role in obtaining good classification accuracy. The main outcomes of the work are the six new generalized versions of previous method called differential evolution classifier. All these DE classifier demonstrated good results in the classification tasks.

Fuzzy real option analysis in patent related decision making and patent valuation

The shift towards a knowledge-based economy has inevitably prompted the evolution of patent exploitation. Nowadays, patent is more than just a prevention tool for a company to block its competitors from developing rival technologies, but lies at the very heart of its strategy for value creation and is therefore strategically exploited for economic pro t and competitive advantage. Along with the evolution of patent exploitation, the demand for reliable and systematic patent valuation has also reached an unprecedented level. However, most of the quantitative approaches in use to assess patent could arguably fall into four categories and they are based solely on the conventional discounted cash flow analysis, whose usability and reliability in the context of patent valuation are greatly limited by five practical issues: the market illiquidity, the poor data availability, discriminatory cash-flow estimations, and its incapability to account for changing risk and managerial flexibility. This dissertation attempts to overcome these impeding barriers by rationalizing the use of two techniques, namely fuzzy set theory (aiming at the first three issues) and real option analysis (aiming at the last two). It commences with an investigation into the nature of the uncertainties inherent in patent cash flow estimation and claims that two levels of uncertainties must be properly accounted for. Further investigation reveals that both levels of uncertainties fall under the categorization of subjective uncertainty, which differs from objective uncertainty originating from inherent randomness in that uncertainties labelled as subjective are highly related to the behavioural aspects of decision making and are usually witnessed whenever human judgement, evaluation or reasoning is crucial to the system under consideration and there exists a lack of complete knowledge on its variables. Having clarified their nature, the application of fuzzy set theory in modelling patent-related uncertain quantities is effortlessly justified. The application of real option analysis to patent valuation is prompted by the fact that both patent application process and the subsequent patent exploitation (or commercialization) are subject to a wide range of decisions at multiple successive stages. In other words, both patent applicants and patentees are faced with a large variety of courses of action as to how their patent applications and granted patents can be managed. Since they have the right to run their projects actively, this flexibility has value and thus must be properly accounted for. Accordingly, an explicit identification of the types of managerial flexibility inherent in patent-related decision making problems and in patent valuation, and a discussion on how they could be interpreted in terms of real options are provided in this dissertation. Additionally, the use of the proposed techniques in practical applications is demonstrated by three fuzzy real option analysis based models. In particular, the pay-of method and the extended fuzzy Black-Scholes model are employed to investigate the profitability of a patent application project for a new process for the preparation of a gypsum-fibre composite and to justify the subsequent patent commercialization decision, respectively; a fuzzy binomial model is designed to reveal the economic potential of a patent licensing opportunity.

Modelling Energy Consumption of a Switch using Fuzzy-Rule Classifier

The energy consumption of IT equipments is becoming an issue of increasing importance. In particular, network equipments such as routers and switches are major contributors to the energy consumption of internet. Therefore it is important to understand how the relationship between input parameters such as bandwidth, number of active ports, traffic-load, hibernation-mode and their impact on energy consumption of a switch. In this paper, the energy consumption of a switch is analyzed in extensive experiments. A fuzzy rule-based model of energy consumption of a switch is proposed based on the result of experiments. The model can be used to predict the energy saving when deploying new switches by controlling the parameters to achieve desired energy consumption and subsequent performance. Furthermore, the model can also be used for further researches on energy saving techniques such as energy-efficient routing protocol, dynamic link shutdown, etc.

Omnibus accounts to Finnish investors? – An analysis of the costs and benefits that omnibus account structure brings to Finnish account operators and investors

The Finnish Securities Markets are being harmonized to enable better, more reliable and timely settlement of securities. Omnibus accounts are a common practice in the European securities markets. Finland forbids the use of omnibus accounts from its domestic investors. There is a possibility that the omnibus account usage is allowed for Finnish investors in the future. This study aims to build a comprehensive image to Finnish investors and account operators in determining the costs and benefits that the omnibus account structure would have for them. This study uses qualitative research methods. A literature review provides the framework for this study. Different kinds of research articles, regulatory documents, studies performed by European organisations, and Finnish news reportages are used to analyse the costs and benefits of omnibus accounts. The viewpoint is strictly of account operators and investors, and different effects on them are contemplated. The results of the analysis show that there are a number of costs and benefits that investors and account operators must take into consideration regarding omnibus accounts. The costs are related to development of IT-systems so that participants are able to adapt to the new structure and operate according to its needs. Decrease in the holdings’ transparency is a disadvantage of the structure and needs to be assessed precisely to avoid some problems it might bring. Benefits are mostly related to the increased competition in the securities markets as well as to the possible cost reductions of securities settlement. The costs and benefits were analysed according to the study plan of this thesis and as a result, the significance and impact of omnibus accounts to Finnish investors and account operators depends on the competition level and the decisions that all market participants make when determining if the account structure is beneficial for their operations.