Design and evaluation of web interface signs to improve web usability : a semiotic framework

Technological innovations, the development of the internet, and globalization have increased the number and complexity of web applications. As a result, keeping web user interfaces understandable and usable (in terms of ease-of-use, effectiveness, and satisfaction) is a challenge. As part of this, designing userintuitive interface signs (i.e., the small elements of web user interface, e.g., navigational link, command buttons, icons, small images, thumbnails, etc.) is an issue for designers. Interface signs are key elements of web user interfaces because ‘interface signs’ act as a communication artefact to convey web content and system functionality, and because users interact with systems by means of interface signs. In the light of the above, applying semiotic (i.e., the study of signs) concepts on web interface signs will contribute to discover new and important perspectives on web user interface design and evaluation. The thesis mainly focuses on web interface signs and uses the theory of semiotic as a background theory. The underlying aim of this thesis is to provide valuable insights to design and evaluate web user interfaces from a semiotic perspective in order to improve overall web usability. The fundamental research question is formulated as What do practitioners and researchers need to be aware of from a semiotic perspective when designing or evaluating web user interfaces to improve web usability? From a methodological perspective, the thesis follows a design science research (DSR) approach. A systematic literature review and six empirical studies are carried out in this thesis. The empirical studies are carried out with a total of 74 participants in Finland. The steps of a design science research process are followed while the studies were designed and conducted; that includes (a) problem identification and motivation, (b) definition of objectives of a solution, (c) design and development, (d) demonstration, (e) evaluation, and (f) communication. The data is collected using observations in a usability testing lab, by analytical (expert) inspection, with questionnaires, and in structured and semi-structured interviews. User behaviour analysis, qualitative analysis and statistics are used to analyze the study data. The results are summarized as follows and have lead to the following contributions. Firstly, the results present the current status of semiotic research in UI design and evaluation and highlight the importance of considering semiotic concepts in UI design and evaluation. Secondly, the thesis explores interface sign ontologies (i.e., sets of concepts and skills that a user should know to interpret the meaning of interface signs) by providing a set of ontologies used to interpret the meaning of interface signs, and by providing a set of features related to ontology mapping in interpreting the meaning of interface signs. Thirdly, the thesis explores the value of integrating semiotic concepts in usability testing. Fourthly, the thesis proposes a semiotic framework (Semiotic Interface sign Design and Evaluation – SIDE) for interface sign design and evaluation in order to make them intuitive for end users and to improve web usability. The SIDE framework includes a set of determinants and attributes of user-intuitive interface signs, and a set of semiotic heuristics to design and evaluate interface signs. Finally, the thesis assesses (a) the quality of the SIDE framework in terms of performance metrics (e.g., thoroughness, validity, effectiveness, reliability, etc.) and (b) the contributions of the SIDE framework from the evaluators’ perspective.

Feature Selection and Classification Using Age Layered Population Structure Genetic Programming

The curse of dimensionality is a major problem in the fields of machine learning, data mining and knowledge discovery. Exhaustive search for the most optimal subset of relevant features from a high dimensional dataset is NP hard. Sub–optimal population based stochastic algorithms such as GP and GA are good choices for searching through large search spaces, and are usually more feasible than exhaustive and deterministic search algorithms. On the other hand, population based stochastic algorithms often suffer from premature convergence on mediocre sub–optimal solutions. The Age Layered Population Structure (ALPS) is a novel metaheuristic for overcoming the problem of premature convergence in evolutionary algorithms, and for improving search in the fitness landscape. The ALPS paradigm uses an age–measure to control breeding and competition between individuals in the population. This thesis uses a modification of the ALPS GP strategy called Feature Selection ALPS (FSALPS) for feature subset selection and classification of varied supervised learning tasks. FSALPS uses a novel frequency count system to rank features in the GP population based on evolved feature frequencies. The ranked features are translated into probabilities, which are used to control evolutionary processes such as terminal–symbol selection for the construction of GP trees/sub-trees. The FSALPS metaheuristic continuously refines the feature subset selection process whiles simultaneously evolving efficient classifiers through a non–converging evolutionary process that favors selection of features with high discrimination of class labels. We investigated and compared the performance of canonical GP, ALPS and FSALPS on high–dimensional benchmark classification datasets, including a hyperspectral image. Using Tukey’s HSD ANOVA test at a 95% confidence interval, ALPS and FSALPS dominated canonical GP in evolving smaller but efficient trees with less bloat expressions. FSALPS significantly outperformed canonical GP and ALPS and some reported feature selection strategies in related literature on dimensionality reduction.

Feature Selection and Classification Using Age Layered Population Structure Genetic Programming

The curse of dimensionality is a major problem in the fields of machine learning, data mining and knowledge discovery. Exhaustive search for the most optimal subset of relevant features from a high dimensional dataset is NP hard. Sub–optimal population based stochastic algorithms such as GP and GA are good choices for searching through large search spaces, and are usually more feasible than exhaustive and determinis- tic search algorithms. On the other hand, population based stochastic algorithms often suffer from premature convergence on mediocre sub–optimal solutions. The Age Layered Population Structure (ALPS) is a novel meta–heuristic for overcoming the problem of premature convergence in evolutionary algorithms, and for improving search in the fitness landscape. The ALPS paradigm uses an age–measure to control breeding and competition between individuals in the population. This thesis uses a modification of the ALPS GP strategy called Feature Selection ALPS (FSALPS) for feature subset selection and classification of varied supervised learning tasks. FSALPS uses a novel frequency count system to rank features in the GP population based on evolved feature frequencies. The ranked features are translated into probabilities, which are used to control evolutionary processes such as terminal–symbol selection for the construction of GP trees/sub-trees. The FSALPS meta–heuristic continuously refines the feature subset selection process whiles simultaneously evolving efficient classifiers through a non–converging evolutionary process that favors selection of features with high discrimination of class labels. We investigated and compared the performance of canonical GP, ALPS and FSALPS on high–dimensional benchmark classification datasets, including a hyperspectral image. Using Tukey’s HSD ANOVA test at a 95% confidence interval, ALPS and FSALPS dominated canonical GP in evolving smaller but efficient trees with less bloat expressions. FSALPS significantly outperformed canonical GP and ALPS and some reported feature selection strategies in related literature on dimensionality reduction.

FEATURE SELECTION AND COMPARISON OF TWO NAÏVE BAYES CLASSIFICATION METHODS IN THE CONTEXT OF SPAM FILTERING

Treating e-mail filtering as a binary text classification problem, researchers have applied several statistical learning algorithms to email corpora with promising results. This paper examines the performance of a Naive Bayes classifier using different approaches to feature selection and tokenization on different email corpora

Experience in introduction of English terminology in engineering lessons: methodology and evaluation

This communication explains a experience for the introduction of English terminology in a technical degree of higher education. We present the methodology and assessment procedures used to evaluate the way the students perceived the introduction of terminology in English in two different subjects from 3rd and 5th year courses of a Computer Science degree in which English was not the vehicular language. We propose a strategy based on two main pillars, namely: 1) The design of materials, explanations, and exams, paying particular attention to the way in which the specific terminology was exposed to the students, and 2) The assessment of the impact in the students by means of the analysis of the feedback trough a set of enquiries. Our experience showed that the students responded very positively to the introduction of English terminology, and presented an affirmative feedback about the impact that an improvement of their linguistic abilities would have in their future work. Further, we present statistics regarding the use of English as the vehicular language for technical reports, which is envisaged as very useful by the students. Finally, we propose a set of questions for further debate which are centered in the role that English terminology should pay in technical degrees, and about the way in which universities should deploy resources in English languages within the different Syllabus

The automation and evaluation of nested clade phylogeographic analysis

Nested clade phylogeographic analysis (NCPA) is a popular method for reconstructing the demographic history of spatially distributed populations from genetic data. Although some parts of the analysis are automated, there is no unique and widely followed algorithm for doing this in its entirety, beginning with the data, and ending with the inferences drawn from the data. This article describes a method that automates NCPA, thereby providing a framework for replicating analyses in an objective way. To do so, a number of decisions need to be made so that the automated implementation is representative of previous analyses. We review how the NCPA procedure has evolved since its inception and conclude that there is scope for some variability in the manual application of NCPA. We apply the automated software to three published datasets previously analyzed manually and replicate many details of the manual analyses, suggesting that the current algorithm is representative of how a typical user will perform NCPA. We simulate a large number of replicate datasets for geographically distributed, but entirely random-mating, populations. These are then analyzed using the automated NCPA algorithm. Results indicate that NCPA tends to give a high frequency of false positives. In our simulations we observe that 14% of the clades give a conclusive inference that a demographic event has occurred, and that 75% of the datasets have at least one clade that gives such an inference. This is mainly due to the generation of multiple statistics per clade, of which only one is required to be significant to apply the inference key. We survey the inferences that have been made in recent publications and show that the most commonly inferred processes (restricted gene flow with isolation by distance and contiguous range expansion) are those that are commonly inferred in our simulations. However, published datasets typically yield a richer set of inferences with NCPA than obtained in our random-mating simulations, and further testing of NCPA with models of structured populations is necessary to examine its accuracy.

The road not taken: observing a process of issue selection and model conceptualization

This article describes a case study involving information technology managers and their new programmer recruitment policy, but the primary interest is methodological. The processes of issue generation and selection and model conceptualization are described. Early use of “magnetic hexagons” allowed the generation of a range of issues, most of which would not have emerged if system dynamics elicitation techniques had been employed. With the selection of a specific issue, flow diagraming was used to conceptualize a model, computer implementation and scenario generation following naturally. Observations are made on the processes of system dynamics modeling, particularly on the need to employ general techniques of knowledge elicitation in the early stages of interventions. It is proposed that flexible approaches should be used to generate, select, and study the issues, since these reduce any biasing of the elicitation toward system dynamics problems and also allow the participants to take up the most appropriate problem- structuring approach.

Calibration and evaluation of individual-based models using Approximate Bayesian Computation

This paper investigates the feasibility of using approximate Bayesian computation (ABC) to calibrate and evaluate complex individual-based models (IBMs). As ABC evolves, various versions are emerging, but here we only explore the most accessible version, rejection-ABC. Rejection-ABC involves running models a large number of times, with parameters drawn randomly from their prior distributions, and then retaining the simulations closest to the observations. Although well-established in some fields, whether ABC will work with ecological IBMs is still uncertain. Rejection-ABC was applied to an existing 14-parameter earthworm energy budget IBM for which the available data consist of body mass growth and cocoon production in four experiments. ABC was able to narrow the posterior distributions of seven parameters, estimating credible intervals for each. ABC’s accepted values produced slightly better fits than literature values do. The accuracy of the analysis was assessed using cross-validation and coverage, currently the best available tests. Of the seven unnarrowed parameters, ABC revealed that three were correlated with other parameters, while the remaining four were found to be not estimable given the data available. It is often desirable to compare models to see whether all component modules are necessary. Here we used ABC model selection to compare the full model with a simplified version which removed the earthworm’s movement and much of the energy budget. We are able to show that inclusion of the energy budget is necessary for a good fit to the data. We show how our methodology can inform future modelling cycles, and briefly discuss how more advanced versions of ABC may be applicable to IBMs. We conclude that ABC has the potential to represent uncertainty in model structure, parameters and predictions, and to embed the often complex process of optimizing an IBM’s structure and parameters within an established statistical framework, thereby making the process more transparent and objective.

A situation awareness interface for a bi-wheeled industrial hovercraft: design, development and evaluation

The intention of this thesis is to develop a prototype interface that enables an operator to control a bi-wheeled industrial hovercraft that will work within a fusion power plant if the automation system fails. This fusion power plant is part of the ITER project a conjoint effort of various industrialized countries to develop cleaner sources of energy. The development of the interface prototype will be based on situation awareness concepts, which provide a means to understand how human operators perceive the world around, then process that information and make decisions based on the knowledge that they already have and the projected knowledge of the reactions that will occur in the world in response to the actions the operator makes. Two major situation awareness methods will be used, GDTA as a means to discover the requirements the interface needs to solve, and SAGAT to conduct the evaluation on the three interfaces. This technique can isolate the differences an operator has in situation awareness when presented with relevant information given by each of the three interfaces that were built for this thesis. Where the first interface presents the information within the operator’s focal point of view in a pictorial style, the second interface shows the same information within the same point of view has the first interface but only shows it in a textual manner. While the third interface shows the relevant information in the operator’s peripheral field of view. Also SAGAT can provide insight on the question to know if providing the operator with feed-forward information about the stoppage distances of the bi-wheeled industrial hovercraft has any effect on the operator’s decision making.

Interactive music visualization: implementation, realization and evaluation

This thesis describes all process of the development of music visualization, starting with the implementation, followed by realization and then evaluation. The main goal is to have to knowledge of how the audience live performance experience can be enhanced through music visualization. With music visualization is possible to give a better understanding about the music feelings constructing an intensive atmosphere in the live music performance, which enhances the connection between the live music and the audience through visuals. These visuals have to be related to the live music, furthermore has to quickly respond to live music changes and introduce novelty into the visuals. The mapping between music and visuals is the focus of this project, in order to improve the relationship between the live performance and the spectators. The implementation of music visualization is based on the translation of music into graphic visualizations, therefore at the beginning the project was based on the existent works. Later on, it was decided to introduce new ways of conveying music into visuals. Several attempts were made in order to discover the most efficient mapping between music and visualization so people can fully connect with the performance. Throughout this project, those attempts resulted in several music visualizations created for four live music performances, afterwards it was produced an online survey to evaluate those live performances with music visualization. In the end, all conclusions are presented based on the results of the online survey, and also is explained which music elements should be depicted in the visuals, plus how those visuals should respond to the selected music elements.

Optimal conductor size selection and reconductoring in radial distribution systems using a mixed-integer LP approach

This paper presents a mixed-integer linear programming model to solve the conductor size selection and reconductoring problem in radial distribution systems. In the proposed model, the steady-state operation of the radial distribution system is modeled through linear expressions. The use of a mixed-integer linear model guarantees convergence to optimality using existing optimization software. The proposed model and a heuristic are used to obtain the Pareto front of the conductor size selection and reconductoring problem considering two different objective functions. The results of one test system and two real distribution systems are presented in order to show the accuracy as well as the efficiency of the proposed solution technique. © 1969-2012 IEEE.

Use of cone-beam tomography and digital subtraction radiography for diagnosis and evaluation of traumatized teeth treated with endodontic surgery and MTA. A case report

The aim of this paper was to report the diagnosis and evaluation of periradicular bone repair by using computed tomography and digital subtraction radiography in an endodontic surgery case treated with mineral trioxide aggregate (MTA). The patient had local swelling and fistula 7years after trauma in the upper jaw. Periapical lesion on tooth #12 and root resorption of tooth #11 were detected radiographically. Endodontic therapy and placement of intracanal medication were carried out. After 3months, with no improvement in the clinical signs and symptoms, computed tomography was performed, showing extensive apical bone resorption on tooth #12 and dental resorption promoting communication of the root canal with the periodontium of tooth #11. The patient was referred to endodontic surgery. After surgery and postoperative periods of 15, 30, 60, 90, and 120days, standardized radiographs were taken, digitized, and subjected to digital subtraction of the images using Adobe Photoshop CS software. Four years later, a cone-beam computed tomography was performed, showing bone repair and absence of root resorption at tooth #11. In this clinical case, digital subtraction radiography was effective for early detection of new bone formation and evolution of repair. © 2011 John Wiley & Sons A/S.

Collecting and evaluation of germplasm of spiked pepper from Brazilian Amazon

Piper aduncum L. é uma planta que ocorre na Amazônia Brasileira com elevado teor de óleo essencial e que apresenta propriedades biológicas utilizáveis na agricultura e saúde humana. Com o objetivo de avaliar germoplasma visando ao melhoramento genético e cultivo econômico, realizaram-se coletas (inflorescências, estacas, folhas e ramos finos) em dez municípios da Amazônia Brasileira (Manaus, Marabá, Goianésia, Moju, Belém, Santa Izabel, Americano, Bonito, Santarém Novo e Aveiro). Tomaram-se dados do ambiente, populações e de doze caracteres morfoagronômicos (número de folhas por ramo, comprimento da folha, largura da folha, circunferência do ramo mais velho, altura da planta, número de ramos ortotrópicos, número de ramos plagiotrópicos, comprimento do entrenó, número de espigas por ramo, rendimento de óleo, teor e produção de dilapiol). As inflorescências e estacas foram identificadas e encaminhadas para a UFRA em Belém-PA e, as folhas e ramos finos, para o Museu Paraense Emílio Goeldi-MPEG, para extração do óleo essencial (hidrodestilação). Utilizaram-se estimadores de média, desvio padrão, coeficiente de variação e amplitude total para estudo da variabilidade fenotípica. As matrizes prevalenceram em ambientes antropizados, solos argilosos, condições de drenagem variáveis, terrenos planos e clima Ami, como também predominaram populações definíveis pela agregação dos indivíduos, em terra alta e a pleno sol, serrapilheira, tamanho das populações e presença de plântulas no chão muito variáveis. Os caracteres de maior variabilidade foram número de ramos ortotrópicos, número de espigas por ramo, circunferência do ramo mais velho (morfológicos), teor e produção de dilapiol (agronômicos). Concluiu-se que a espécie apresenta adaptação a diferentes ambientes com relação à vegetação, solo, clima, relevo e drenagem, facilitando o cultivo e domesticação. Há variabilidade morfoagronômica favorecendo a seleção e fitomelhoramento.