In model-driven engineering, powerful query/view languages exist to compute result sets/views from underlying models. However, to use these languages effectively, one must understand the query/view language concepts as well as the underlying models and metamodels structures. Consequently, it is a challenge for domain experts to create queries/views due to the lack of knowledge about the computer-internal abstract representation of models and metamodels. To better support domain experts in the query/view creation, the goal of this paper is the presentation of a generic concept to specify queries/views on models without requiring deep knowledge on the realization of modeling languages. The proposed concept is agnostic to specific modeling languages and allows the query/view generation by-example with a simple mechanism for filtering model elements. Based on this generic concept, a generic query/view language is proposed that uses role-oriented modeling for its non-intrusive application for specific modeling languages. The proposed language is demonstrated based on the role-based single underlying model (RSUM) approach for AutomationML to create queries/views by-example, and subsequently, associated viewtypes to modify the result set or view.

Changes in the workplace and impending shortages of organizational leaders make it imperative that HRD professionals develop a better understanding of the developmental processes of emergent leaders entering the workplace. While leader development research within the field of HRD has typically focused on established workers, the research in this study assumes a lifespan approach to leader development. This study contributes to the development of the field by examining the leadership experiences of 18 to 20 year olds who were leaders of organizations in high school and how these experiences shaped the identities of these emergent leaders. Themes that emerged related to their experiences included their relationships with others, how they led by example, the development of authentic leadership qualities, and their motivation to lead in new venues. Implications for practice and future research are identified.

In our attempt to construct a semantic filter for satellite image content, we have built a software that allows user to indicate a few number of image regions that contains a specific geographical object, such as, a bridge, and to retrieve similar objects on the same satellite image. We are particularly interested in performing a data analysis approach based on user interaction. User can guide the classification procedure by interaction and visual observation of the results. We have applied a two step procedure for this and preliminary results show that we eliminate many true negatives while keeping most of the true positives.

As digital images become ubiquitous in many applications, the need for efficient and effective retrieval techniques is more demanding than ever. Query by Example (QBE) and Query by Concept (QBC) are among the most popular query models. The former model accepts example images as queries and searches for similar ones based on low-level features such as colors and textures. The latter model allows queries to be expressed in the form of high-level semantics or concept words, such as "boat" or "car," and finds images that match the specified concepts. Recent research has focused on the connections between these two models and attempts to close the semantic-gap between them. This research involves finding the best method that maps a set of low-level features into high-level concepts. Automatic annotation techniques are investigated in this dissertation to facilitate QBC. In this approach, sets of training images are used to discover the relationship between low-level features and predetermined high-level concepts. The best mapping with respect to the training sets is proposed and used to analyze images, annotating them with the matched concept words. One principal difference between QBE and QBC is that, while similarity matching in QBE must be done at the query time, QBC performs concept exploration off-line. This difference allows QBC techniques to shift the time-consuming task of determining similarity away from the query time, thus facilitating the additional processing time required for increasingly accurate matching. Consequently, QBC's primary design objective is to achieve accurate annotation within a reasonable processing time. This objective is the guiding principle in the design of the following proposed methods which facilitate image annotation: 1.A novel dynamic similarity function. This technique allows users to query with multiple examples: relevant, irrelevant or neutral. It uses the range distance in each group to automatically determine weights in the distance function. Among the advantages of this technique are higher precision and recall rates with fast matching time. 2.Object recognition based on skeletal graphs. The topologies of objects' skeletal graphs are captured and compared at the node level. Such graph representation allows preservation of the skeletal graph's coherence without sacrificing the flexibility of matching similar portions of graphs across different levels. The technique is robust to translation, scaling, and rotation invariants at object level. This technique achieves high precision and recall rates with reasonable matching time and storage space. 3.ASIA (Automatic Sampling-based Image Annotation) is a technique based on a new sampling-based matching framework allowing users to identify their area of interest. ASIA eliminates noise, or irrelevant areas of the image. ASIA is robust to translation, scaling, and rotation invariants at the object level. This technique also achieves high precision and recall rates. While the above techniques may not be the fastest when contrasted with some other recent QBE techniques, they very effectively perform image annotation. The results of applying these processes are accurately annotated database images to which QBC may then be applied. The results of extensive experiments are presented to substantiate the performance advantages of the proposed techniques and allow them to be compared with other recent high-performance techniques. Additionally, a discussion on merging the proposed techniques into a highly effective annotation system is also detailed.

Ease, comfort and efficiency are assumed desirables; they form the established norm of unquestioned values in commercial product design. The norm shapes our everyday. Those daily things considered mundane and commonplace, form how we go about our daily doings. How we actually live our lives. Design needs to be there to defend human interest; to approach the user differently than a passive consumer, to create space for human qualities in contemporary everyday life.  I have been exploring the potential of designing for effort, and argue for its value by giving examples. I have developed a series of three products under the shared name Friction Intended. The proposals evoke effort of different kinds, each creating space for other alternative values. Object A is a light concept working with reflections. Reflecting from one surface to another the light can be followed and its behavior studied. The reflective elements are tools for exploration and active learning; to actively perceive the daily phenomenon of light. Object B is a backpack to be assembled from a large sheet of textile and a set of straps. Over time, the usage of the bag can become a personal ritual. Wearing the bag can also be a statement; how will people  react when the large cloth is dramatically folded open in a public environment? Object C is a cup with rounded base. The cup moves; never fully finding its balance it sways back and forth ever differently depending on the amount of liquid inside and the qualities of the gestures it has been handled with. The attention is drawn to the moment by giving careful attention to a simple daily ritual. Designing for effort in everyday products creates space to design for an engaging and stimulating environment. Once deciding simple things are worth more time, strain and patience there is the opportunity to enrich those activities. Effort has the potential to create space for the development of contemporary rituals, active engagement and everyday curiosity. The design space of the potential of effort is a rich and varied. The examples given by the Friction Intended series, are representatives for a field where much more potential still lays. This is a call, especially on the design field, to question the given, to challenge the norm and to reflect on its impact.

Ce travail de thèse porte sur l'étude et le développement de stratégies innovantes pour la résolution, basée sur l'utilisation de la simulation et de la théorie de l'apprentissage statistique, de problèmes inverses dans le domaine contrôle non destructif (CND) par méthodes électromagnétiques. L’approche générale adoptée consiste à estimer un ensemble des paramètres inconnus, constituant un sous-ensemble des paramètres décrivant le scénario de contrôle étudié. Dans les cas de CND, les trois applications classiquement visées sont la détection, la localisation et la caractérisation de défauts localisés dans le matériau inspecté. Ce travail concerne d’une part la localisation et la caractérisation des fissures et d’autre part l'estimation de certains paramètres de sonde difficiles à maîtriser ou inconnus. Dans la littérature, de nombreuses méthodes permettant de remonter aux paramètres inconnus ont été étudiées. Les approches d'optimisation standard sont basées sur la minimisation d'une fonction de coût, décrivant l'écart entre les mesures et les données simulées avec un solveur numérique. Les algorithmes les plus répandus se fondent sur des approches itératives déterministes ou stochastiques. Cette thèse considère le problème de l'estimation de paramètres inconnus dans une perspective d'apprentissage statistique/automatique. L’approche supervisée adoptée est connue sous le nom de d’apprentissage par l'exemple (LBE en anglais). Elle se compose d’une première phase, dite hors ligne, pendant laquelle un « modèle inverse » est construit sur la base de la connaissance d’un ensemble de couples entrée/sortie connu, appelé ensemble d’entraînement. Une fois la phase d’apprentissage terminée et le modèle généré, le modèle est utilisé dans une phase dite en ligne pour prédire des sorties inconnues (les paramètres d'intérêt) en fonction de nouvelles entrées (signaux CND mesurés appartenant à un second ensemble dit de test) en temps quasi-réel. Lorsqu’on considère des situations pratiques d'inspection, en raison du grand nombre de variables impliquées, la création d'un modèle précis et robuste n’est pas une tâche triviale (problème connu comme la malédiction de la dimensionnalité). Grâce à une étude approfondie et systématique, l’approche développée dans cette thèse a conduit à la mise en place de différentes solutions capables d’atteindre une bonne précision dans l’estimation des paramètres inversés tout en conservant de très bonnes performances en temps de calcul. Le schéma LBE proposé dans cette thèse a été testé avec succès sur un ensemble des cas réels, en utilisant à la fois des données synthétiques bruitées et des mesures expérimentales. / The research activity of the PhD thesis focuses on the study and development of innovative strategies for the solution of inverse problems arising in the field of Non-Destructive Testing and Evaluation (NDT-NDE), based on the use of statistical learning theory. Generally speaking, the objective of the optimization stage is the retrieval of the unknown parameters within the studied electromagnetic scenario. In the case of NDT-NDE, the optimization problem, in terms of parameters to estimate, is divided into three stages, namely detection, localization and characterization. This work mainly addresses localization and characterization of crack(s) and/or estimation of probe(s) parameters. Unknown parameters, constituting a subset of the parameters set describing the electromagnetic scenario, are robustly estimated using several approaches. Standard optimization approaches are based on the minimization, by means of iterative approaches like stochastic and/or deterministic algorithms, of a cost function describing the discrepancy between measurements and prediction. This thesis considers the estimation problem in a machine learning perspective, adopting well known Learning-By-Example (LBE) paradigm. In a so-called offline phase, a surrogate inverse model is first fitted on a set of known input/output couples, generated through numerical simulations. Then, in a so-called online phase, the model predicts unknown outputs (the parameters of interest) based on new inputs (measured NDT signals) in quasi-real time. When considering practical inspection situations, due to the large number of variables involved (known as curse of dimensionality), obtaining an accurate and robust model is not a trivial task. This thesis carries out a deep and systematic study of different strategies and solutions to achieve simultaneously good accuracy and computational time efficiency in the parameters estimation. Moreover, a particular emphasis is put on the different approaches adopted for mitigating the curse of dimensionality issue. The proposed LBE schema has been tested with success on a wide set of practical problems, using both synthetic noisy data and experimental measurements.

In 2013, 35% of the workforce was not engaged, which results in lack of productivity and loss of profitability for small business enterprises (SBEs). The purpose of this qualitative case study was to explore successful strategies that frontline leaders in a 4 generation, family-owned excavating business used to engage their frontline employees. The excavating business was started in 1947 by the father of the current business owners. William Kahn's employee engagement theory was the conceptual framework for this study. Data were collected through a focus group and direct observations of engagement during meetings and frontline areas from a population of 8 frontline leaders from construction work at an excavating business in Stephens City, Virginia. Data from the focus group and direct observations were thematically analyzed and then triangulated to ensure the trustworthiness of the interpretations. The 5 themes that emerged included: investing in sustainability, leading by example, providing clear and open communication, implementing a system of measurement, and developing a professional image. These themes could provide the basis for the area frontline leaders to improve the employee engagement level of their frontline employees. These findings could prompt what has been a missing dialogue of communication that could bridge the employee engagement gap between the area employees and employers. Social change implications of these findings could lead to productivity improvement that could contribute to the survival of SBEs and to the employment status of the community.

This thesis proposes machine learning algorithms for processing geometry by example. Each algorithm takes as input a collection of shapes along with exemplar values of target properties related to shape processing tasks. The goal of the algorithms is to output a function that maps from the shape data to the target properties. The learned functions can be applied to novel input shape data in order to synthesize the target properties with style similar to the training examples. Learning such functions is particularly useful for two different types of geometry processing problems. The first type of problems involves learning functions that map to target properties required for shape interpretation and understanding. The second type of problems involves learning functions that map to geometric attributes of animated shapes required for real-time rendering of dynamic scenes. With respect to the first type of problems involving shape interpretation and understanding, I demonstrate learning for shape segmentation and line illustration. For shape segmentation, the algorithms learn functions of shape data in order to perform segmentation and recognition of parts in 3D meshes simultaneously. This is in contrast to existing mesh segmentation methods that attempt segmentation without recognition based only on low-level geometric cues. The proposed method does not require any manual parameter tuning and achieves significant improvements in results over the state-of-the-art. For line illustration, the algorithms learn functions from shape and shading data to hatching properties, given a single exemplar line illustration of a shape. Learning models of such artistic-based properties is extremely challenging, since hatching exhibits significant complexity as a network of overlapping curves of varying orientation, thickness, density, as well as considerable stylistic variation. In contrast to existing algorithms that are hand-tuned or hand-designed from insight and intuition, the proposed technique offers a largely automated and potentially natural workflow for artists. With respect to the second type of problems involving fast computations of geometric attributes in dynamic scenes, I demonstrate algorithms for learning functions of shape animation parameters that specifically aim at taking advantage of the spatial and temporal coherence in the attribute data. As a result, the learned mappings can be evaluated very efficiently during runtime. This is especially useful when traditional geometric computations are too expensive to re-estimate the shape attributes at each frame. I apply such algorithms to efficiently compute curvature and high-order derivatives of animated surfaces. As a result, curvature-dependent tasks, such as line drawing, which could be previously performed only offline for animated scenes, can now be executed in real-time on modern CPU hardware.

This thesis proposes machine learning algorithms for processing geometry by example. Each algorithm takes as input a collection of shapes along with exemplar values of target properties related to shape processing tasks. The goal of the algorithms is to output a function that maps from the shape data to the target properties. The learned functions can be applied to novel input shape data in order to synthesize the target properties with style similar to the training examples. Learning such functions is particularly useful for two different types of geometry processing problems. The first type of problems involves learning functions that map to target properties required for shape interpretation and understanding. The second type of problems involves learning functions that map to geometric attributes of animated shapes required for real-time rendering of dynamic scenes. With respect to the first type of problems involving shape interpretation and understanding, I demonstrate learning for shape segmentation and line illustration. For shape segmentation, the algorithms learn functions of shape data in order to perform segmentation and recognition of parts in 3D meshes simultaneously. This is in contrast to existing mesh segmentation methods that attempt segmentation without recognition based only on low-level geometric cues. The proposed method does not require any manual parameter tuning and achieves significant improvements in results over the state-of-the-art. For line illustration, the algorithms learn functions from shape and shading data to hatching properties, given a single exemplar line illustration of a shape. Learning models of such artistic-based properties is extremely challenging, since hatching exhibits significant complexity as a network of overlapping curves of varying orientation, thickness, density, as well as considerable stylistic variation. In contrast to existing algorithms that are hand-tuned or hand-designed from insight and intuition, the proposed technique offers a largely automated and potentially natural workflow for artists. With respect to the second type of problems involving fast computations of geometric attributes in dynamic scenes, I demonstrate algorithms for learning functions of shape animation parameters that specifically aim at taking advantage of the spatial and temporal coherence in the attribute data. As a result, the learned mappings can be evaluated very efficiently during runtime. This is especially useful when traditional geometric computations are too expensive to re-estimate the shape attributes at each frame. I apply such algorithms to efficiently compute curvature and high-order derivatives of animated surfaces. As a result, curvature-dependent tasks, such as line drawing, which could be previously performed only offline for animated scenes, can now be executed in real-time on modern CPU hardware.

Voice user interfaces have been growing in popularity and with them an interest for open vocabulary keyword spotting. In this thesis we focus on one particular approach to open vocabulary keyword spotting, query by example keyword spotting. Three types of query by example keyword spotting approaches are described and evaluated: sequence distances, speech to phonemes and deep distance learning. Evaluation is done on a series of custom tasks designed to measure a variety of aspects. The Google Speech Commands benchmark is used for evaluation as well, this to make it more comparable to existing works. From the results, the deep distance learning approach seem most promising in most environments except when memory is very constrained; in which sequence distances might be considered. The speech to phonemes methods is lacking in the usability evaluation. / Röstgränssnitt har växt i populäritet och med dem ett intresse för öppenvokabulärnyckelordsigenkänning. I den här uppsatsen fokuserar vi på en specifik form av öppenvokabulärnyckelordsigenkänning, den s.k nyckelordsigenkänning- genom- exempel. Tre typer av nyckelordsigenkänning- genom- exempel metoder beskrivs och utvärderas: sekvensavstånd, tal till fonem samt djupavståndsinlärning. Utvärdering görs på konstruerade uppgifter designade att mäta en mängd olika aspekter hos metoderna. Google Speech Commands data används för utvärderingen också, detta för att göra det mer jämförbart mot existerade arbeten. Från resultaten framgår det att djupavståndsinlärning verkar mest lovande förutom i miljöer där resurser är väldigt begränsade; i dessa kan sekvensavstånd vara av intresse. Tal till fonem metoderna visar brister i användningsuvärderingen.