Devido Ãs crescentes ameaÃas inerentes aos sistemas de informaÃÃo, o uso de mecanismos de autenticaÃÃo e autorizaÃÃo baseados em identificador de usuÃrio e senha nÃo à mais suficiente para garantir a seguranÃa das informaÃÃes. Este trabalho propÃe um novo modelo de autenticaÃÃo e autorizaÃÃo para controle de acesso de aplicaÃÃes distribuÃdas, baseado em resumos criptogrÃficos e certificados de atributos. Os resumos criptogrÃficos sÃo utilizados no processo de autenticaÃÃo da aplicaÃÃo, enquanto os certificados de atributos especificam privilÃgios e outras informaÃÃes de autorizaÃÃo associadas ao seu proprietÃrio. Os certificados de atributos sÃo gerenciados pela infra-estrutura de gerenciamento de privilÃgios (IGP). A arquitetura e o funcionamento do modelo bem como os processos de geraÃÃo do certificado de atributos, autenticaÃÃo e autorizaÃÃo da aplicaÃÃo sÃo descritos. O modelo proposto foi especificado em Redes de Petri Coloridas e validado por meio de simulaÃÃes. / Due to increasing threats inherent to the information systems, the use of authentication and authorization mechanisms based in login and password does not enough to assure the information security. This work proposes a new model of authentication and authorization for distributed applications, based in hash and attributes certificates. Hash is used in the application authentication process, while certificates of attributes specify privileges and other authorization information. Its use is managed by the privilege management infrastructure (PMI). In this work, we describe the architecture and the functioning of the model, as well the processes of the attributes certificates generation, authentication and authorization of the application. The proposed model was specified in Coloured Petri Nets and validated by simulation.

Com a intensificação e expansão da produção agrícola brasileira, seguidas da crescente preocupação ambiental, surge a necessidade da adoção de novas tecnologias para o monitoramento e avaliação em larga escala dos recursos naturais, em especial o solo. Nesse sentido a espectroscopia de reflectância, aliada as bibliotecas espectrais, atende estas expectativas por ser uma técnica eficiente, rápida, barata e não poluente, pois não usa reagentes e nem gera resíduos, porém sua aplicação nos solos tropicais brasileiros ainda é incipientes e carece de mais estudos. Sendo assim, o objetivo principal do trabalho foi comparar as informações espectrais obtidas por espectroscopia de reflectância nas faixas do visível, infravermelho próximo e, principalmente, no infravermelho médio com análises convencionais de solo a fim entender as relações existentes entre o comportamento espectral e os atributos físicos, químicos e mineralógicos dos solos estudados, ainda quantificar e caracterizar a variabilidade desses atributos e dos solos com base no comportamento espectral. Para tanto, foram utilizadas 1288 amostras de solo correspondentes a diferentes horizontes de 458 perfis da região central brasileira (GO, MS, MG e SP), sobre as quais foram procedidas análise granulométrica e química, para fins de levantamento de solos, e mineralógica. Os espectros VisNIR foram adquiridos com o sensor FieldSpec Pro, enquanto que espectros Mid-IR com o sensor Nicolet 6700 Fourier Transform Infrared (FT-IR) e, após isso, os dados foram transformados com os seguintes préprocessamentos: remoção do espectro contínuo, absorbância e correção da linha base. As diferentes faixas espectrais também foram combinadas pela outer product analysis. A distribuição dos dados espectrais e a influência de cada atributo nos comportamentos espectrais dos solos foram avaliadas através de análises estatísticas multivariadas, tais como correlação multivariada, análise por componentes principais, seguidas das análises de agrupamento kmédias, para a avaliação entre amostras, e distância taxonômica, para avaliação entre perfis. A predição dos atributos dos solos com base nos dados espectrais foi determinada pelos algoritmos de regressão multivariada: mínimos quadrados parciais (PLSR), árvores de regressão ampliadas (BT) e máquina de vetor suporte (SVM); sendo que a qualidade das predições foi definida pelos R2, RMSE, RPD e RPIQ. O comportamento espectral das principais classes de solos tropicais brasileiros, em ambas as faixas espectrais, foi apresentado juntamente com as principais feições de absorção diagnosticadas (minerais 2:1, caulinita, óxidos de Fe e Al), assim como a contribuição de cada atributo do solo na variação dos espectros. Foi possível estabelecer a formação de grupos de solo com base no grau de intemperismo das amostras, enquanto que os perfis das principais classes de solos puderam ser discriminados com base na variação do comportamento espectral em profundidade. Grande parte dos atributos estudados pode ser quantificada nas duas faixas espectrais pelo menos com um nível de qualidade razoável (RPD > 1,40), sendo que o principal algoritmo de predição foi o SVM com os espectros em valores de absorbância. A espectroscopia de reflectância, principalmente na faixa Mid-IR, apresenta-se como uma tecnologia eficaz de alta potencialidade e aplicabilidade nas análises qualitativa e quantitativas dos solos, principalmente para fins de levantamento e mapeamento. / Considering the intensification and expansion of Brazilian agricultural production followed by the growing environmental concern, there is the necessity of adopting new technologies to monitor on a large scale the natural resources, in particular, the soil. In this sense, reflectance spectroscopy, integrated with spectral libraries, meets that expectation as an alternative for proximal soil sensing because its more efficient than other techniques being cheaper and faster than conventional analyses, besides, it doesnt produce waste. However, its application on Brazilian tropical soils is still incipient and needs more studies. The main aim of this research was to compare spectral information acquired by reflectance spectroscopy in the visible, near infrared and, mainly, in the mid infrared ranges with conventional soil analyses in order to understand the relationships among spectral behavior and physical, chemical, and mineralogical attributes of the studied soils, also, to quantify and characterize the variability of these attributes and soils based on their spectral behavior. For that, a dataset with 1288 soil samples referring to different horizons from 458 soil profiles from the central region of Brazil (States of Goiás, Mato Grosso do Sul, Minas Gerais e São Paulo) was used, and convetional analyses (granulometric, chemical and mineralogical) for soil survey were applied over them. VisNIR reflectance data were acquired by the FiledSpec Pro sensor whereas Mid-IR spectral data were acquired by the Nicolet 6700 (FT-IR) sensor and, after that, both data were transformed by the following pre-processing: continuum removal, absorbance and baseline correction. The different spectral ranges were also combined by outer product analysis. The data distribution and the influence of each attribute on soil spectral behavior were evaluated by multivariate statistical analyses, such as, multivariate correlations, principal component analysis followed by k-means clustering, for assessment among soil samples, and by taxonomic distance, for assessment among soil profiles. Predictions of soil attributes based on spectral data were modeled by following multivariate regression algorithms: partial least square regression (PLSR), boosted regression trees (BT) and support vector machine (SVM) and quality of modeling were evaluated by R2, RMSE, RPD and RPIQ. Spectral behavior of the main soil classes, in both ranges, were showed along with the absorption features of the major tropical minerals (2:1 minerals, kaolinite, Fe and Al oxides), as well as, the contribution of each soil attribute on spectra variations. Sample clustering based on different weathering levels of soils was possible regarding differences in absorbance intensities and features whereas profiles of the main soil classes could be discriminated based on variation of spectral behavior in depth. Much of the studied attributes could be predicted in both spectral range at least in a reasonable level of quality (RPD > 1,40), and SVM was considered the principal prediction algorithm as well as absorbance transformation as the major pre-processing. Reflectance spectroscopy, especially in the Mid-IR range, shows up as a high potential technique for qualitative and quantitative analysis of Brazilian soils, in particular for soil survey or mapping.

Submitted by Mariana Dornelles Vargas (marianadv) on 2015-05-06T18:02:07Z No. of bitstreams: 1 humor_propagandas.pdf: 942498 bytes, checksum: ccceb92f33058e5404e26b45302a2152 (MD5) / Made available in DSpace on 2015-05-06T18:02:07Z (GMT). No. of bitstreams: 1 humor_propagandas.pdf: 942498 bytes, checksum: ccceb92f33058e5404e26b45302a2152 (MD5) Previous issue date: 2010 / Nenhuma / A propaganda cumpre um papel importante em favor da sociedade, economia e mercado das empresas. Seu valor no marketing é fundamental, considerada a principal ferramenta de exposição ao público, capaz de modificar comportamentos e atitudes, encantando e persuadindo os mesmos com apelos variados e que fazem a diferença no contexto propagandístico. Isto é, o humor é um apelo considerado irreverente e de sucesso quando bem usado nas peças publicitárias. É um recurso que pode facilitar o interesse do consumidor pela mensagem. Objetiva-se neste estudo aprofundar o entendimento acerca da percepção dos consumidores sobre o humor presente nas propagandas televisivas. Destaca-se, no entanto, que existem poucos estudos que remetem ao apelo humor no Brasil, especificamente nas propagandas televisivas. Para tanto, a presente dissertação contribui com o tema, apresentando estudos bibliográficos de autores seminais e descrevendo um estudo qualitativo no método de amostragem teórica, especificamente como entrevista em profundidade. Abordando doze mulheres, ditas como donas de casa e profissionais de mercado, que dizem gostar de ver televisão, com características biográficas variadas (idade, estado civil, profissão) e consequentemente com características comportamentais bem distintas. Nas etapas metodológicas, as participantes foram entrevistadas separadamente (uma por uma), totalizando nove perguntas, mas ressaltando, quando necessário, a realização de "subperguntas", a fim de instigar e fazê-las falar e exprimir sentimentos, emoções e valores. O método dessa entrevista expõe tais condições, tendo em vista que sua intenção é levantar todos esses dados comportamentais para uma conclusão satisfatória do estudo. A entrevistadora, dita como investigadora (STRAUSS, 2008), não tendência as respostas, mas faz as entrevistadas responderem com detalhes, buscando envolvê-las para assim perceber e aprofundar nas suas formas de comunicação verbal ou não-verbal o assunto proposto. Depois dessa etapa, fazem-se as análises comparando respostas e avaliando, numa visão sistêmica e num foco aprofundado, o olhar que as mesmas têm em relação ao humor nas propagandas. Diante dessa proposta, o estudo constata que há um comportamento positivo, um olhar notável na aceitação do humor nas propagandas que condiciona e acaba influenciando no quesito lembrança e "no chamar a atenção" do telespectador, embora o atributo humor deva estar unido a outros elementos-chaves que ajudam e favorecem a persuasão, como por exemplo, a criatividade contextual, a necessidade dos produtos na visão do consumidor, o preço, o nome, os personagens, entre outros. O fato de aceitar e ver o humor como um apelo forte e dito como diferencial pode estabelecer uma relação direta junto à intenção da compra, mas nessa condição, são necessários mais estudos e métodos diferenciados para admitir tal hipótese com exatidão. Com o estudo, pode-se constatar que o indivíduo vê a propaganda humorística, quando "bem feita", com "bons olhos". De acordo com as entrevistadas, o comercial que utiliza o humor fica na lembrança do indivíduo, associa o produto à propaganda, e como consequência, serve de auxílio no ato da compra. Quanto às implicações acadêmicas e sugestões a dar continuidade a futuros estudos são expostos no fim da dissertação. / Advertising plays an important role for society, economy and business market. Its value in marketing is crucial, as the leading tool for public exposure, can change behavior and attitudes, charming and persuading them with varied and appeals that make a difference in the context propaganda. The humor is irreverent and considered an appeal of success when properly used in the materials. It is a resource that can facilitate the interests of consumers by the message. The objective of this study was to deepen the understanding of consumers perceptions of the humor present in television advertisements. Stands out, however, that there are few studies that refer the appeal humor in Brazil, specifically in television advertisements. To that end, this paper contributes to the topic, presenting bibliographical studies author seminal and a qualitative study describing the method of theoretical sampling, specifically as in-depth interview, covering twelve women, spoken as housewives and professional market, which say they enjoy watching television, with various biographical characteristics (age, marital status, occupation) and consequently with very different behavioral Characteristics. In methodological steps, participants were interviewed separately (one by one), totaling nine questions, but noting, where necessary the creation of sub-questions in order to instigate and make them talk and express feelings, emotions and values. The method of this interview exposes these conditions, remembering that their intention is to raise all these behavioral data to meet the study. The interviewer said to the researcher (Strauss, 2008) tendency not the answers but does the interviewees respond in detail to involve them trying to understand and thus deepen in its forms of verbal or nonverbal communication the matter proposed. After this step, make up the analysis by comparing and evaluating responses in a systemic view and a focus depth look that they have in relation to humor in advertising. Given this proposal, the study finds that there is a positive attitude, a look at the remarkable acceptance of humor in advertisements that conditions influencing and ends in the question and remembrance in calling the attention of the viewer, although the mood attribute should be attached to other elements Key which help promote and persuasion, such as contextual creativity, the need for vision products in the consumer price, the names, characters, among others. The fact accept and see humor as a strong call said as differential can establish a direct relationship with the intention of buying, but in this condition, further studies are needed and different methods to admit this hypothesis accurately. With the study, one can see that the individual sees the humorous advertisement, when well done with good eyes, according to the interviewees, the shopping is in remembrance of the individual, associate product advertising, and as a consequence serves as an aid in the purchase. As for the academic implications and suggestions for future studies continue to be exposed at the end of the dissertation.

Chocolate is a food that is craved by many in the world and healthy chocolates have become a current topic in the healthy eating trend. The healthiness of a chocolate depends mainly on the polyphenol, fat and sugar contents. Although the literature serves several interventions to reduce the fat content and total calorie content of chocolates, it does not provide many interventions that consider both improving the polyphenol and reducing the fat content simultaneously. Considering this gap in the literature, this research project sought to develop a chocolate that is both low in fat and high in polyphenol content, without sacrificing the taste that consumers would expect in a good dark chocolate. The research resulted in three chocolate formulas that consists of 60% cocoa and 30% fat (formula A, formula B and formula C). They differed from each other in the polyphenol content due to the changes in the cocoa powder type and the presence or absence of freeze-dried blueberry powder. Formula A and formula B mainly differed in the percentage of each type of cocoa powder, both alkalized and non- alkalized. Formula C differed from the other two formulas because there was no freeze-dried blueberry powder. The resulted chocolates were evaluated for the sensory characteristics using paired preference tests and consumer-oriented attribute diagnostic tests (color, flavor, melting properties, overall preference, likelihood to buy). They were analyzed for polyphenol content using Folin-Ciocalteu assay. The chocolates were compared with a well-established commercial chocolate (formula D) of a 60% cocoa. Sensory evaluation tests revealed that formula C was superior and comparable to the commercial chocolate D in all the tested attributes except for color. Polyphenol analysis confirmed that all three developed formulas had a higher polyphenol estimate than formula D. Consumers preferred to buy each chocolate type regardless of the differences in their preference for the other attributes, if the chocolates were known to be healthier. Formula C was found to be the best prototype formula among the developed formulas and it can further be developed to enhance the color and other attributes in future work.

Over the past thirty years, clinical research has benefited substantially from the adoption of electronic medical record systems. As deployment has increased, so too has the number of researchers seeking to improve the overall analytical environment by way of tools and models. Although much work has been done, there are still many uninvestigated areas; two of which are explored in this dissertation. The first pertains to the physical storage of the data itself. There are two generally accepted storage models: relational and entity-attribute-value (EAV). For clinical data, EAV systems are preferred due to their natural way of managing many-to-many relationships, sparse attributes, and dynamic processes along with minimal conversion effort and reduction in federation complexities. However, the relational database management systems on which they are implemented, are not intended to organize and retrieve data in this format; eroding their performance gains. To combat this effect, we present the foundation for an EAV Database Management System (EDBMS). We discuss data conversion methodologies, formulate the requisite metadata and partitioned type-sensing index structures, and provide detailed runtime and experimental analysis with five extant methods. Our results show that the prototype, EAVDB, reduces space and conversion requirements while enhancing overall query performance. The second topic concerns query performance in a federated environment. One method used to decrease query execution time, is to pre-compute and store "beneficial" queries (views). The View Selection Problem (VSP) identifies these views subject to resource constraints. A federated model, however, has yet to be developed. In this dissertation, we submit three advances in view materialization. First, a more robust optimization function, the Minimum-Maintenance View Selection Problem (MMVSP), is derived by combining existing approaches. Second, the Federated View Selection Problem (FVSP), built upon the MMVSP, and federated data cube lattice are formalized. The FVSP allows for multiple querying nodes, partial and full materialization, and data propagation constriction. The latter two are shown to greatly reduce the overall number of valid solutions within the solution space and thus a novel, multi-tiered approach is given. Lastly, EAV materialization, which is introduced in this dissertation, is incorporated into an expanded, multi-modal variant of the FVSP. As models and heuristics for both the federated and EAV VSP, to the best of our knowledge, do not exist, this research defines two new branches of data warehouse optimization. Coupled with our EDBMS design, this dissertation confronts two main challenges associated with clinical data warehousing and federation.

本文撰寫目的在於探討並解決斯賓諾莎形上學體系內的兩項「一多問題」，即實體與屬性的同一問題，以及實體的單一性與有限樣態的多樣性之間的調和問題。 為達上述目的，本文以斯賓諾莎原始文獻之英譯本為主，並以國內外著名學者之研究論著為輔，期能為上述兩項問題之研究提供充分的文獻基礎。 在研究方法的選擇上，本文乃採歷史研究法與哲學之純理推演法並兼，資料分 析法與結論歸納法並重的研究方式，俾能在適當的主題範圍內，提供有利 的研究工具，進而完成研究目的。 經由上述諸般方法之運用，吾人不唯從斯氏未嘗是一笛卡兒學說信徒之確認當中，探知實體與無限多屬性全體間「一而多」的整然存在狀況及特殊的同一關係；又從個別屬性與實體間等同之特定條件之確認當中，將實體與屬性的同一關係做出更精密而全面的闡釋，從而也因此解決了斯氏形上學體系內兩項「一多問題」之第一項。為解決斯氏形上學體系內的第二項「一多問題」，吾人一面先敷陳斯氏理智認識下的實在之一般真貌，一面又點出蘊藏在一般真貌之下的矛盾暗流；在斯氏關鍵命題給出的當口，激流化暗為明，存在於實體單一性與有限樣態多樣性間的形式矛盾浮現檯面。在重重的資 料分析與結論的最終歸納後，觀念論者以違背斯氏實在之一般真貌而無法 為吾人苟同；實在論者則能俱足諸項條件，從而清晰一貫而合理地為斯氏 第二項「一多問題」，提供令人滿意的解決。 本文主要關鍵字有五：1.實體(Substance)；2.屬性(Attribute)；3.樣 態(Mode)；單一性(Unity)；5.多樣性(Multiplicity)。

<p>It is not unusual that software systems are redeveloped when the systems environment changes or new requirements to functionality are found. The reason for this is that system evolution is usually not considered during the initial development of a system.</p><p>In this report we propose a development process focusing on modifiability in security-safety critical systems. The process is based on the standard IEC 61508 Functional safety of electrical/electronic/programmable electronic safety-related systems and the CORAS framework for model-based risk assessment. We focus on integrating modifiability as a non-functionality and extend the development process with an architecture and design phase. Modifiability is addressed using Architecture Tradeoff Analysis Method (ATAM) and Attribute Driven Design (ADD). An evaluation of this process is given by redeveloping a prototype for an already existing security-critical system using the approaches given by the process. The main result from the evaluation showed that we where able to obtain a higher level of modifiability than the original prototype, without compromising the functionality and safety level required.</p>

Die Arbeit untersucht die Annahme einer unterschiedlichen Gewichtung von distinktiven enzyklopädischen, funktionalen und sensorischen Merkmalen innerhalb der Repräsentationen von Objekten der belebten und unbelebten semantischen Domäne. Hierzu wurde ein Reaktionszeitexperiment zur Merkmalsverifikation durchgeführt. Vorab wurden deutsche Normen über das geschätzte Erwerbsalter für 244 Stimuli aus dem Korpus von Snodgrass & Vanderwart (1980) erhoben. Weiterhin wurde eine Datenbank von Merkmalsnormen für 80 konkrete Objektbegriffe erstellt. Insgesamt wurden zwei Reaktionszeitexperimente durchgeführt, die sich lediglich durch die Darbietungsdauer des Konzeptbegriffes unterschieden. Der Konzeptbegriff wurde entweder 1000 ms (lange Darbietung) oder 250 ms (kurze Darbietung) präsentiert, bevor das zu verifizierende semantische Merkmal erschien. Bei langer Präsentationszeit des Objektbegriffes zeigten sich für Objekte der unbelebten Domäne schnellere Reaktionszeiten beim Verifizieren von distinktiven funktionalen Merkmalen als beim Verifizieren von distinktiven enzyklopädischen Merkmalen. Dieser Effekt wurde bei kurzer Darbietungsdauer des Konzeptbegriffes repliziert. Bei kurzer Darbietung konnten für Objekte der unbelebten Domäne zusätzlich kürzere Reaktionszeiten beim Verifizieren distinktiver funktionaler Merkmale als beim Verifizieren distinktiver sensorischer Merkmale beobachtet werden. Für Objekte der belebten Domäne lagen weder nach kurzer noch nach langer Präsentation des Objektbegriffes Unterschiede in den Reaktionszeiten beim Verifizieren der semantischen Merkmale vor. Die Ergebnisse werden vor dem Hintergrund aktueller neurolinguistischer Modelle zur Organisation des semantischen Gedächtnisses diskutiert. Die Ergebnisse deuten darauf hin, dass innerhalb der Objektrepräsentationen belebter Objekte alle drei Merkmalstypen interkorrelieren. Für Objekte der unbelebten Domäne werden starke Interkorrelationen zwischen funktionalen und sensorischen Merkmalen angenommen. Zusätzlich wird davon ausgegangen, dass distinktive funktionale Merkmale innerhalb der Repräsentationen unbelebter Objekte besonders stark gewichtet sind. / The study investigates the assumption of a differential weighting of distinctive encyclopaedic, functional and sensory semantic attributes within the concept representations of living and nonliving objects. A speeded feature verification task was conducted. In order to control the stimuli for various aspects, German norms were collected of estimated age of acquistion for 244 stimuli of the Snodgrass and Vanderwart (1980) pictures. In addition, a feature listing study was carried out for 80 concrete living and nonliving concepts. Two reaction time experiments were conducted with different duration times of the presented object name. In Experiment 1 (long duration time), the object name was presented for 1000 ms, whereas in Experiment 2 (short duration time), the object name was presented for 250 ms before the semantic feature appeared. With long duration time, distinctive functional features of nonliving objects were verified faster than distinctive encyclopaedic features. This effect was replicated in Experiment 2 after short presentation of the object names. In addition, in Experiment 2, shorter reaction times were observed for functional features than for sensory features of nonliving objects. For living objects, no differences were observed in verifying the different feature types neither in Experiment 1 nor in Experiment 2. The results are discussed within the framework of different models of semantic memory. The results suggest that within the category of living objects, distinctive encyclopaedic, functional and sensory features are highly intercorrelated. For nonliving objects, it is assumed that distinctive sensory and functional but not encyclopaedic features are highly intercorrelated. In addition, distinctive functional features of nonliving objects seem to be weighted more strongly than distinctive sensory features within the object representations of nonliving objects.

It is not unusual that software systems are redeveloped when the systems environment changes or new requirements to functionality are found. The reason for this is that system evolution is usually not considered during the initial development of a system. In this report we propose a development process focusing on modifiability in security-safety critical systems. The process is based on the standard IEC 61508 Functional safety of electrical/electronic/programmable electronic safety-related systems and the CORAS framework for model-based risk assessment. We focus on integrating modifiability as a non-functionality and extend the development process with an architecture and design phase. Modifiability is addressed using Architecture Tradeoff Analysis Method (ATAM) and Attribute Driven Design (ADD). An evaluation of this process is given by redeveloping a prototype for an already existing security-critical system using the approaches given by the process. The main result from the evaluation showed that we where able to obtain a higher level of modifiability than the original prototype, without compromising the functionality and safety level required.

This work focuses on a cognitive science inspired solution to an event detection problem in a video domain. The thesis raises the question whether video sequences that are taken in highway tunnels can be used to create meaningful data in terms of symbolic representation, and whether these symbolic representations can be used as sequences to be parsed by attribute grammars into abnormal and normal events. The main motivation of the research was to develop a novel algorithm that parses sequences of primitive events created by the image processing algorithms. The domain of the research is video detection and the special application purpose is for highway tunnels, which are critical places for abnormality detection. The method used is attribute grammars to parse the sequences. The symbolic sequences are created from a cascade of image processing algorithms such as / background subtracting, shadow reduction and object tracking. The system parses the sequences and creates alarms if a car stops, moves backwards, changes lanes, or if a person walks into the road or is in the vicinity when a car is moving along the road. These critical situations are detected using Earley&rsquo / s parser, and the system achieves real-time performance while processing the video input. This approach substantially lowers the number of false alarms created by the lower level image processing algorithms by preserving the number of detected events at a maximum. The system also achieves a high compression rate from primitive events while keeping the lost information at minimum. The output of the algorithm is measured against SVM and observed to be performing better in terms of detection and false alarm performance.