Many studies on feature-based search have found that color is preferentially used as a guiding feature. Most of these studies have utilized simple stimuli and the several studies that utilized real-world objects have been limited by the subjective color classification of objects. The current set of studies examined how search was guided by color and color proportions in real-world objects using color histograms to objectively classify color. To bridge the gap between simple stimuli and real-world objects, the first experiment examined how different color compositions in color conjunctions affected search. In the second experiment, real-world objects were objectively classified according to their primary color as determined by a color histogram, which was used to predict behavior. Finally, in the third experiment, real-world objects were classified by both their primary and secondary colors, which again, was used to predict behavior. Across three experiments the results demonstrated that search was guided by both colors of the target and there was more guidance to objects that matched the target's primary color than the target's secondary color. The results showed that theories derived from studying simple stimuli can be applied to more complex stimuli and tools developed in computer vision research can be used as an objective measure.

[pt] Este trabalho está inserido no contexto de ambientes de autoria e execução hipermídia, sendo as linguagens declarativas para autoria de documentos o seu foco principal. Tendo-se como objetivo aumentar a expressividade e o reuso na especificação de documentos hipermídia, este trabalho introduz as linguagens XSMIL e NCL - Nested Context Language - versão 2.1. Utilizando- se o conceito de templates, X-SMIL permite a definição de novas semânticas para composições SMIL, além dos tradicionais elementos seq, par e excl. Templates, em X-SMIL, são especificados em um perfil de XTemplate, que estende a idéia original da linguagem XTemplate de NCL. Com base nas novas facilidades para definição de templates, esse perfil foi usado para especificar a linguagem NCL 2.1. X-SMIL também permite a especificação de conectores hipermídia, tratando relações hipermídia como entidades de primeira classe - funcionalidade incorporada em XSMIL pelo uso do módulo XConnector de NCL. Outro objetivo deste trabalho é o de apresentar um framework para o processamento de documentos XML. Utilizando-se esse framework, diversos compiladores foram implementados, o que possibilitou, entre outras funcionalidades, a conversão de documentos NCL em especificações SMIL ou X-SMIL e vice-versa. / [en] This work is related to hypermedia authoring and execution environments, and its main focus is declarative document authoring. Aiming at improving the expressiveness and reuse in the specification of hypermedia documents, this work introduces the hypermedia authoring languages X-SMIL and NCL - Nested Context Language - version 2.1. Exploiting the concept of templates, X-SMIL allows the definition of new semantics for SMIL compositions, besides its usual seq, par and excl elements. X-SMIL templates are specified using an XTemplate profile, which extends the original idea of the NCL XTemplate language. Bringing new facilities for template definitions, this new profile is used to further improve the NCL language. X-SMIL also offers support for handling hypermedia relations as first-class entities, through the use of hypermedia connectors - brought to X-SMIL via the NCL XConnector module. Another important goal of this work is to present a framework to facilitate the development of XML documents parsing and processing tools. Based on this framework, several compilers were implemented, permitting, among other features, the conversion of NCL documents into SMIL or X-SMIL specifications and vice- versa.

[pt] A abordagem orientada a objetos do padrão MPEG-4, para a codificação de conteúdo audiovisual, é similar às utilizadas em vários modelos e linguagens de especificação de documentos multimídia/hipermídia. Entre essas linguagens, a NCL (Nested Context Language), utilizada no sistema HyperProp, introduz uma série de novos conceitos que podem ser integrados ao padrão, com vantagens. Esta dissertação propõe, inicialmente, a conversão de documentos especificados em NCL para MPEG-4 (XMT-O) e vice-versa, permitindo que ferramentas de autoria e formatação possam ser utilizadas na especificação e exibição de documentos de ambas as linguagens. Este trabalho também propõe a incorporação de cenas MPEG-4 tanto como objetos de mídia quanto composições da linguagem NCL, permitindo o estabelecimento de relacionamentos entre cenas. Para permitir a exibição desses novos objetos NCL, é incorporado ao Formatador HyperProp um exibidor MPEG-4 capaz de reportar ao controlador a ocorrência de eventos que, entre outras coisas, permite o sincronismo entre cenas MPEG-4 e outros objetos NCL, incluindo outras cenas MPEG-4. Por fim, explorando o conceito de templates introduzido pela linguagem NCL, a capacidade de autoria no MPEG-4 é estendida, através da definição de novas semânticas para as composições da linguagem XMT-O e da concepção de compiladores para essa linguagem. / [en] The MPEG-4 standard object-oriented approach, employed to the encoding of audiovisual content, is similar to those used on many models and languages for multimedia/hypermedia document specification. Among those languages, the NCL (Nested Context Language), used in the HyperProp system, introduces a series of new concepts that can be integrated to the standard, with advantages. Initially, the proposal of this work is to convert NCL to MPEG-4 (XMT-O) documents and vice versa, allowing authoring and formatting tools to be used in the specification and presentation of documents in both languages. This work also proposes both the placing of MPEG-4 scenes as media objects and NCL language compositions, allowing the establishment of relationships among scenes. In order to allow displaying these new NCL objects, an MPEG-4 player is incorporated to the HyperProp Formatter. The MPEG-4 player is able to report to the controller the occurrence of events that, among other things, allows the synchronization between MPEG-4 scenes and othe r NCL objects, including other MPEG-4 scenes. Finally, exploring the concept of templates, introduced by the NCL language, the authoring in the MPEG-4 is improved, by means of the definition of new semantics for XMT-O language compositions and the design of compilers for this language.

L'Apprentissage Supervisé est un domaine populaire de l'Apprentissage Automatique en progrès constant depuis plusieurs années. De nombreuses techniques ont été développées pour résoudre le problème de classification, mais, dans la plupart des cas, ces méthodes se basent sur la présence et le nombre de points d'une classe donnée dans des zones de l'espace que doit définir le classifieur. Á cause de cela la construction de ce classifieur est dépendante de la densité du nuage de points des données de départ. Dans cette thèse, nous montrons qu'utiliser la topologie des données peut être une bonne alternative lors de la construction des classifieurs. Pour cela, nous proposons d'utiliser les graphes topologiques comme le Graphe de Gabriel (GG) ou le Graphes des Voisins Relatifs (RNG). Ces dernier représentent la topologie de données car ils sont basées sur la notion de voisinages et ne sont pas dépendant de la densité. Pour appliquer ce concept, nous créons une nouvelle méthode appelée Classification aléatoire par Voisinages (Random Neighborhood Classification (RNC)). Cette méthode utilise des graphes topologiques pour construire des classifieurs. De plus, comme une Méthodes Ensemble (EM), elle utilise plusieurs classifieurs pour extraire toutes les informations pertinentes des données. Les EM sont bien connues dans l'Apprentissage Automatique. Elles génèrent de nombreux classifieurs à partir des données, puis agrègent ces classifieurs en un seul. Le classifieur global obtenu est reconnu pour être très eficace, ce qui a été montré dans de nombreuses études. Cela est possible car il s'appuie sur des informations obtenues auprès de chaque classifieur qui le compose. Nous avons comparé RNC à d'autres méthodes de classification supervisées connues sur des données issues du référentiel UCI Irvine. Nous constatons que RNC fonctionne bien par rapport aux meilleurs d'entre elles, telles que les Forêts Aléatoires (RF) et Support Vector Machines (SVM). La plupart du temps, RNC se classe parmi les trois premières méthodes en terme d'eficacité. Ce résultat nous a encouragé à étudier RNC sur des données réelles comme les tweets. Twitter est un réseau social de micro-blogging. Il est particulièrement utile pour étudier l'opinion à propos de l'actualité et sur tout sujet, en particulier la politique. Cependant, l'extraction de l'opinion politique depuis Twitter pose des défis particuliers. En effet, la taille des messages, le niveau de langage utilisé et ambiguïté des messages rend très diffcile d'utiliser les outils classiques d'analyse de texte basés sur des calculs de fréquence de mots ou des analyses en profondeur de phrases. C'est cela qui a motivé cette étude. Nous proposons d'étudier les couples auteur/sujet pour classer le tweet en fonction de l'opinion de son auteur à propos d'un politicien (un sujet du tweet). Nous proposons une procédure qui porte sur l'identification de ces opinions. Nous pensons que les tweets expriment rarement une opinion objective sur telle ou telle action d'un homme politique mais plus souvent une conviction profonde de son auteur à propos d'un mouvement politique. Détecter l'opinion de quelques auteurs nous permet ensuite d'utiliser la similitude dans les termes employés par les autres pour retrouver ces convictions à plus grande échelle. Cette procédure à 2 étapes, tout d'abord identifier l'opinion de quelques couples de manière semi-automatique afin de constituer un référentiel, puis ensuite d'utiliser l'ensemble des tweets d'un couple (tous les tweets d'un auteur mentionnant un politicien) pour les comparer avec ceux du référentiel. L'Apprentissage Topologique semble être un domaine très intéressant à étudier, en particulier pour résoudre les problèmes de classification...... / Supervised Learning is a popular field of Machine Learning that has made recent progress. In particular, many methods and procedures have been developed to solve the classification problem. Most classical methods in Supervised Learning use the density estimation of data to construct their classifiers.In this dissertation, we show that the topology of data can be a good alternative in constructing classifiers. We propose using topological graphs like Gabriel graphs (GG) and Relative Neighborhood Graphs (RNG) that can build the topology of data based on its neighborhood structure. To apply this concept, we create a new method called Random Neighborhood Classification (RNC).In this method, we use topological graphs to construct classifiers and then apply Ensemble Methods (EM) to get all relevant information from the data. EM is well known in Machine Learning, generates many classifiers from data and then aggregates these classifiers into one. Aggregate classifiers have been shown to be very efficient in many studies, because it leverages relevant and effective information from each generated classifier. We first compare RNC to other known classification methods using data from the UCI Irvine repository. We find that RNC works very well compared to very efficient methods such as Random Forests and Support Vector Machines. Most of the time, it ranks in the top three methods in efficiency. This result has encouraged us to study the efficiency of RNC on real data like tweets. Twitter, a microblogging Social Network, is especially useful to mine opinion on current affairs and topics that span the range of human interest, including politics. Mining political opinion from Twitter poses peculiar challenges such as the versatility of the authors when they express their political view, that motivate this study. We define a new attribute, called couple, that will be very helpful in the process to study the tweets opinion. A couple is an author that talk about a politician. We propose a new procedure that focuses on identifying the opinion on tweet using couples. We think that focusing on the couples's opinion expressed by several tweets can overcome the problems of analysing each single tweet. This approach can be useful to avoid the versatility, language ambiguity and many other artifacts that are easy to understand for a human being but not automatically for a machine.We use classical Machine Learning techniques like KNN, Random Forests (RF) and also our method RNC. We proceed in two steps : First, we build a reference set of classified couples using Naive Bayes. We also apply a second alternative method to Naive method, sampling plan procedure, to compare and evaluate the results of Naive method. Second, we evaluate the performance of this approach using proximity measures in order to use RNC, RF and KNN. The expirements used are based on real data of tweets from the French presidential election in 2012. The results show that this approach works well and that RNC performs very good in order to classify opinion in tweets.Topological Learning seems to be very intersting field to study, in particular to address the classification problem. Many concepts to get informations from topological graphs need to analyse like the ones described by Aupetit, M. in his work (2005). Our work show that Topological Learning can be an effective way to perform classification problem.

Optical properties of nanoscopic materials have been intensively pursued over last couple of decades due to their tunable optical properties. Recent interests in this field have been mainly focused on the preparation of ordered arrays of nano materials and study of their optical properties. These interests have been motivated by the applications of such systems for nano photonic devices. Theoretical predictions from such systems reveal complex absorption and emission properties, different from individual ones mainly because of energy transfer between them. These properties can be controlled further by preparing hybrid arrays of nanostructures, including nano crystals of different types. Hybrid arrays with semiconductor quantum dots and metallic nanoparticles are an example of such system. Optical properties of such a system can be tuned by controlling the interaction between excitons and plasmons. This thesis presents the experimental studies on optical properties of polymer capped nanoparticles, quantum dot arrays and hybrid arrays with semi conducting quantum dot and metal nanoparticles. A brief summary of the experimental methods and results have been highlighted below. In this thesis, we study the controlling decay dynamics of CdSe quantum dots by 2D photonic-plasmonic and metamaterial templates. In Chapter 1 we provide a detailed background on the theoretical methods of Light-Matter interaction at nano scale. We also have given the detailed information on both weak and strong coupling region in the light-matter interaction. This chapter includes the discussion controlling light-matter interaction with both photonic crystals and plasmonic materials with some appropriate examples from the literature. In this chapter we have also explained the relevance of our work in this area and organization of the chapters and there importance has given. In chapter 2 we provide details about various experimental methods used in this thesis. A brief introduction is given on the materials used, their synthesis and the preparation of different type of self assembled plasmonic-photonic templates. This chapter starts with an explanation of the materials used along with the justification; moves on to the preparation of different 2D wire metamaterial. The characterization techniques for these different types of templates like spectroscopic ellipsometer, atomic force spectroscopy, scanning electron microscopy and transmission electron microscopy are discussed. We also discussed optical spectroscopic techniques like confocal optical microscopy and near field optical microscopy techniques. The first two chapters form the basis of all the experiments discussed in the forth coming chapters. In chapter 3 Finite difference time domain (FDTD) simulations were performed on two different plasmonic sub wavelength photonic templates embedded with CdSe quantum dots. Tunable loading of these templates with plasmonic nano antenna allowed control of the emission from the embedded quantum dots. We discuss how large loading of nano antenna can effectively control the optical density of states for the quantum dots leading to enhancement of their radiative decay rates as observed in experiments. On the other hand, at low level of loading, while FDTD fails to capture the observed enhancement of decay rates in experiment, an alternative mechanism is suggested to exist in such cases. Thus, subtle interplay of multiple mechanisms engineered by appropriate placement and loading of plasmonic nano antenna in such templates is demonstrated as an effective method to control optical density of states and hence spontaneous emission of embedded quantum dots. In Chapter 4 we report results of controlled tuning of the local density of states (LDOS) in versatile, flexible and hierarchical self assembled plasmonic templates. Using 5 nm diameter gold (Au) spherical nano antenna within a polymer template randomly dispersed with quantum dots, we show how the photo-luminescence intensity and lifetime anisotropy of these dots can be significantly enhanced through LDOS tuning. Finite difference time domain simulations corroborate the experimental observations and extend the regime of enhancement to a wider range of geometric and spectral parameters bringing out the versatility of these functional plasmonic templates. It is also demonstrated how the templates act as plasmonic resonators for effectively engineer giant enhancement of the scattering efficiency of these nano antenna embedded in the templates. Our work provides an alternative method to achieve spontaneous emission intensity and anisotropy enhancement with true nanoscale plasmon resonators. In chapter 5 we reported enhancement optical properties of quantum dot monolayers on top of the functional, flexible and hierarchical self-assembled plasmonic template using extremely small gold (Au) nanoparticles of diameter 5 nm. We reported how the LODS changes with different polarizations for CdSe quantum dot present on top of the template. We observed the enhanced radiative LDOS from the nano antenna filled pores indicating plasmonic enhanced emission from these templates. The difference in spectral and spatial profile of LDOS and Pur-cells with polarization of quantum dot emission results in the anisotropic emission in these templates. In chapter 6 we reported the emergence of strong coupling between quantum emitters and 2D hyperbolic metamaterials (HMM). We studied both spectral dependence and effect of filling fraction of the HMM on strong interaction. We also show the controlling of the transition from weak coupling region to strong coupling region by changing the distance between QD monolayer and HMM. By using FDTD simulation we are able to calculate both spectral function S(!) and coupling efficiency. In chapter 7 as a conclusion we concluded the work done in this thesis. We also indicated the future directions in this field and possible application.

Monomorphization and dynamic dispatch are two common techniques for implementing polymorphism in statically typed programming languages. Function templates in C++ use the former technique to enable algorithms written as generic functions to be efficiently reused with multiple different data types by producing a separate function instantiation for each invocation that uses a unique permutation of argument types. This avoids the overhead of indirection associated with dynamic dispatch and allows the generated code of each instantiation to be optimized by the compiler for its specific concrete types, which typically yields great improvements in runtime performance over any dynamic approach. The disadvantage of this implementation, compared to the type-erased generics found in many other programming languages, is that careless over-use of templates with many different argument types can lead to an excessive amount of redundant code being generated for the same function. This increase in code size may increase the binary size of the final program and reduce the amount of useful code that can fit into the processor's instruction cache during execution, reducing code locality and thereby potentially reducing performance. Monomorphization can also increase compilation time due to the increase in generated code that needs to be compiled and optimized. This thesis presents a heuristic-based approach to generic programming that allows function templates to be automatically converted to use dynamic dispatch in scenarios where the resulting negative impact on runtime performance is predicted to be low. The thesis project includes the development of a proof of concept plugin for the Clang compiler frontend that can be used to compile existing C++ projects with the conversions applied. The design of a heuristic function for determining whether a given function template should use monomorphization or dynamic dispatch based on statically known metrics is proposed based on the results of an experiment. This heuristic is shown to achieve a small general improvement in program size across a set of open-source C++ projects when they are compiled using the plugin. The key findings from the experiment and from the development of the plugin are summarized with a general strategy for how the approach can be integrated into the design of future programming languages to promote more extensive use of generic programming in performance-sensitive code while avoiding regressions in program size and compilation time.

Målet med studien är att förstå hur och vad som beskrivs och bedöms genom de samtalsmallar som används inför utvecklingssamtal på förskolan. Det är en kvalitativ studie baserad på innehållsanalys av insamlade dokument. Innehållsanalysen utfördes genom kodscheman som tas upp av Bergström och Boréus (2008). I analysarbetet har vi inspirerats av diskursanalysen. Studien är baserad på åtta samtalsmallar inför utvecklingssamtal från olika förskolor i en kommun i Mellansverige. Studien visar att samtalsmallarna innehöll flera olika sätt att beskriva och bedöma barn. Genom vårt analysarbete fann vi framför allt fem tydliga teman som utgör vårt resultat; Det lärande barnet, där barnet genom samtalsmallarna beskrivs och bedöms inom olika områden och ses som lärande eller under utveckling. Barnet som person, baserat på punkterna i samtalsmallarna som utgörs av bedömning och beskrivning av barnets personliga attribut. Det bristande barnet, med grund i samtalsmallarnas fokus på barnets tillkortakommanden och brister. Barnet i verksamhetens kontext, utifrån punkter i samtalsmallarna där verksamheten beskrivs och bedöms och Det positivt framställda barnet, grundat i till exempel uppmaningar i samtalsmallarna om att hålla samtalen i positiv anda. Tillika visar studien att flera olika barnsyner ligger till grund för hur barn ska beskrivas och bedömas i samtalsmallarna, samt att det råder motsägelser och bristande direktiv i förskolans styrdokument för utvecklingssamtalens syfte och innehåll. / The aim of this study is to understand how and what is described and assessed in the templates used for parent-teacher conferences in pre-school. This study is qualitative in nature and the chosen method of examination was content analysis of documents through coding schemes inspired by those described in Bergström and Boréus (2008). Discourse analysis inspired us when analyzing our data. The documents upon which the study is based consist of eight templates for parent-teacher conferences from different pre-schools in a municipality of Sweden. The study shows that templates for parent-teacher conferences consist of several different ways of describing and making assessments of the children. In our analysis we found five distinguishing themes as described in our result; The learning child, where the child is looked upon and described as under development, within different subjects. The child as a person, as the child’s personality was to be assessed and described thru the templates. The child of flaws, based on the conversational topics in the templates focused on flaws and problems. The child in the context of pre-school, due to the focus on describing and assessing the pre-school in the templates and finally The positively described child, based on the directives in some of the templates about describing the child thru positive outlook. The study also shows a variety of teachers’ perspectives on children based on the ways in which the children are to be described and assessed in the templates. Additionally, the study reveals a lack of and somewhat contradicting directives in the regulatory documents for pre-schools, considering the purpose and content of parent-teacher conferences.

We report here the development of very efficient sulfonate based leaving groups, termed Nucleophile Assisting Leaving Groups (NALGs), to accelerate the rate of nucleophilic substitution reactions involving poor nucleophiles and/or substrates traditionally considered too hindered to undergo nucleophilic attack. Indeed NALGs have shown exceptional ability in improving rate of nucleophilic substitution reactions. New very mild stereoretentive halogenations and azidation reactions have also been developed for secondary cyclic alcohols using NALGs involving titanium(IV) reagents. This reaction is particularly significant since the carbon-halogen bond is found widely in natural products and is used extensively as a synthesis intermediate. Azide is also a synthetically important functional group from which a variety of biologically important functional groups are conveniently obtained. Though stereoretentive chlorination and bromination reactions are known, we have developed, for the first time, a stereoretentive azidation reaction using titanium(IV) azide, a reagent not previously used in organic synthesis. During our development of stereoretentive reactions, we eventually developed very efficient, mild, two-step one-pot stereoretentive halogenations (chlorination and bromination) using titanium(IV) halides as catalysts or stoichiometric reagents. These reactions were found to be particularly efficient for cyclic alcohols. An efficient one pot stereoretentive amidation reaction for secondary cyclic alcohols is also reported. The important features of this reaction are that, for the first time, chlorosulfite (prepared in situ from alcohol using thionylchloride) has been used as a leaving group and titanium(IV) fluoride as an activator. / Utilization of those two reagents is unique as thionylchloride has never been used for nucleophilic substitution reactions except in chlorination procedures. In addition, this work has found new and creative applications for titanium (IV) fluoride, a reactant rarely used in organic synthesis. Further exploiting the unique reactivity of titanium(IV), reactions of alkenes with various nucleophiles have been developed with this reagent in both catalytic and stoichiometric quantities. It was observed that a-substituted aromatic conjugated alkenes dimerize to generate important indan class of compounds which are very important in the polymer industry. In addition, non conjugated unactivated alkenes react with various nucleophiles to yield the adduct. / by Deboprosad Mondal. / Thesis (Ph.D.)--Florida Atlantic University, 2010. / Includes bibliography. / Electronic reproduction. Boca Raton, Fla., 2010. Mode of access: World Wide Web.

The recently discovered cucurbit[n]uril are a range of macrocyclic hosts which have enormous potential in industrial, medical and academic applications. Cucurbit[n]uril have a rigid repeating structure of methylene bridged glycouril, which give cucurbit[n]uril their gourd like shape of a cavity with two carbonyl fringed portals. In this thesis the host-guest binding abilities of three cucurbit[n]uril (n = 6, 7, 8) have been examined for a range of potential guests. These guests ranged from simple alkyl amines through globular alkyl and carboranyl amines to bipyridyl systems. In total 45 guest molecules where examined. Most of the guests examined where either cationically charged, capable of hydrogen binding, contained a substantial molecular dipole, or a combination of these. Furthermore, all of the potential guests examined had some solubility in an acidified aqueous sodium sulfate solution within which the host-guest properties were examined. It was generally found that the larger guests did have selectivity for the larger hosts. However, when the host became too large weaker complexes would form and for the range of materials examined here cucurbit[7]uril was found to be the 'best' host system. In one example, p-xylene diamine, a 2:1 complex with cucurbit[8]uril was observed. While not the focus of this work a new rapid purification method was developed for the cucurbit[n]uril using different metal ions to either solubilise or precipitate the different cucurbit[n]uril. In the second part of this work these same guest molecules where used as potential templates in the synthesis of cucurbit[n]uril. Surprisingly the guests that bound strongly to an individual host did not seem to template the cucurbit[n]uril synthesis at all. Rather these strong binders inhibited the reaction such that little or no cucurbit[n]uril formed under the reaction conditions studied. However, several examples provided excellent template results. Indeed the results indicate that guests which bound with intermediate rates of exchange are the best templates and using templates under these conditions we have been able to produce cucurbit[7]uril as 46% by mass of the total cucurbit[n]uril product. This is the highest yield ever recorded for cucurbit[7]uril and it is the first example of cucurbit[7]uril being the major product of this condensation reaction. In an another example cucurbit[8]uril formed 18% of the product an increase of 150% over the standard reaction conditions. While studying both the template reactions and the host-guest binding properties of the cucurbit[n]uril a new supramolecular form, an 'inverse rotaxane' was discovered. Inverse rotaxanes are not held in place by large blocking groups, rather the molecular structure encapsulated by the cucurbit[n]uril host prevents decomplexation of the axle.

Object-oriented application frameworks allow the reuse of both software design and code and are one of the most effective reuse technologies available today. Frameworks provide domain-specific concepts, which are generic units of functionality. Framework-based applications are constructed by writing completion code that instantiates these concepts. The instantiation of such concepts requires implementation steps in the completion code, such as subclassing framework-provided classes, implementing interfaces, and calling appropriate framework services. Unfortunately, many existing frameworks are difficult to use because of their large and complex APIs and often incomplete user documentation. To cope with this problem, application developers often use existing framework applications as a guide. While existing applications contain valuable examples of concept implementation steps, locating them in the application code is often challenging. To address this issue, this dissertation introduces the notion of concept implementation templates, which summarize the necessary concept implementation steps, and a technique named FUDA (Framework API Understanding through Dynamic Analysis) which automatically extracts such templates from runtime information collected when that concept is invoked in two or more different contexts in one or more sample applications. The experimental evaluation of FUDA with twelve realistic concepts on top of four widely-used frameworks suggests that the technique is effective in producing quality implementation templates for a given concept with high precision and recall from only two sample sample applications and execution scenarios. Moreover, it was observed in a user study with twelve subjects that the choice of templates vs. documentation had much less impact on development time than the concept complexity.