Music listening is considered one of the most ubiquitous activities in everyday life, and one of the main reasons why people listen is to affect and regulate their mood. The vast availability and unlimited access of music has made it difficult to find relevant music that fits both the context and the preferences of the music listener. The aim of this project was to investigate the personalized relationship between music and mood using everyday technologies, focusing on how a listening experience could be adapted to the desired affect of a music listener while also taking the user’s individual listening history into account. In large, the project concentrated on the possibility of using context-aware music recommendation to generate personalized playlists by focusing on the audio features and corresponding mood of the music. A web-based application was developed to act as a prototype for the study, where the application allowed users to connect to Spotify, pick a desired mood and generate a playlist. By allowing people to access music in this personalized way, a user study could be conducted in order to investigate their music listening while incorporating this recommendation tool. The findings showed that the users’ found the experience to be engaging in that they could use the application as a companion to everyday tasks in addition to it being a tool for getting new, personalized music recommendations. Overall, the participants also found the generated playlists to be accurate to their music preferences and desired affective state.

Natural Language Generation and generating believable verbal communication are critical components in the development of social robots. The work presented in this paper is based on the sequence-to-sequence model and is focused on improving context and repeatability through the inclusion of task- specific information. The data set on which this study was conducted was collected through a Wizard of Oz framework using a social robot. The generated dialogue was evaluated through a survey designed to measure the adherence to the game context and perceived human qualities. The human qualities were measured using attributes from two well-known attribute scales intended for evaluating Human-Robot Interaction. The evaluation results indicate that the quality of the generated dialogue is on par with examples of actual dialogue spoken during the experiments. This paper also highlights interesting aspects regarding the usefulness of transfer learning in narrow contextual applications. The results presented in this paper show that it is possible to improve the contextual nature of generated dialogue by including additional task-specific information. / Generering av naturligt språk och uppgiften att skapa trovärdig verbal kommunikation är kritiska komponenter i utvecklingen av sociala robotar. Arbetet som presenteras i denna uppsats är baserat på sekvens-till-sekvens-modellen och fokuserar på att förbättra sammanhang och repeterbarhet genom att inkludera uppgiftspecifik information. Datauppsättningen som denna studie genomförde samlades in via ett Wizard of Oz-ramverk med hjälp av en social robot. Den genererade dialogen utvärderades genom en onlineundersökning utformad för att mäta efterlevnaden av spelskontexten och upplevda mänskliga egenskaper. Dessa mänskliga egenskaper mättes med attribut från två välkända attributskalor avsedda för utvärdering av människa-robot-interaktion. Utvärderingsresultaten visar att kvaliteten på den genererade dialogen är i nivå med exempel på faktisk dialog som talats under experimenten. Denna uppsats belyser också intressanta aspekter beträffande nyttan av överföringsinlärning i smala kontextuella applikationer. Resultaten som presenteras i denna uppsats visar att det är möjligt att förbättra den kontextuella karaktären hos genererad dialog genom att inkludera ytterligare uppgiftspecifik information.

Made available in DSpace on 2016-06-02T19:05:58Z (GMT). No. of bitstreams: 1 4486.pdf: 3511223 bytes, checksum: 3f8c09a3cb87230a2ac0f6706ea07944 (MD5) Previous issue date: 2012-08-08 / Financiadora de Estudos e Projetos / The mining association rules are an important task in data mining. Traditional algorithms of mining association rules are based only on the database items, providing a very specific knowledge. This specificity may not be advantageous, because the users normally need more general, interesting and understandable knowledge. In this sense, there are approaches working in order to obtain association rules with items belonging to any level of a taxonomic structure. In the crisp contexts taxonomies are used in different steps of the mining process. When the objective is the generalization they are used, mainly, in the pre-processing or post-processing stages. On the other hand, in the fuzzy context, fuzzy taxonomies are used, mainly, in the pre-processing step, during the generating extended transactions. A great problem of these transactions is related to the huge amount of candidates and rules. Beyond that, the inclusion of ancestors ends up generating redundancy problems. Besides, it is possible to see that many works have directed efforts for the question of mining fuzzy rules, exploring linguistic terms, but few approaches have been proposed for explore new steps of mining process. In this sense, this paper proposes the Context FOntGAR algorithm, a new algorithm for mining generalized association rules under all levels of fuzzy ontologies composed by specialization/generalization degrees varying in the interval [0,1]. In order to obtain more semantic enrichment, the rules may be composed by similarity relations, which are represented at the fuzzy ontologies in different contexts. In this work the generalization is done during the post-processing step. Other relevant points of this paper are the specification of a new approach of generalization; including a new grouping rules treatment, and a new and efficient way for calculating both support and confidence of generalized rules. / Algoritmos tradicionais de associação se caracterizam por utilizar apenas itens contidos na base de dados, proporcionando um conhecimento muito específico. No entanto, essa especificidade nem sempre é vantajosa, pois normalmente os usuários finais necessitam de padrões mais gerais, e de fácil compreensão. Nesse sentido, existem abordagens que não se limitam somente aos itens da base, e trabalham com o objetivo de minerar regras (generalizadas) com itens presentes em qualquer nível de estruturas taxonômicas. Taxonomias podem ser utilizadas em diferentes etapas do processo de mineração. A literatura mostra que, em contextos crisp, essas estruturas são utilizadas tanto em etapa de pré-processamento, quanto em etapa de pós-processamento, e que em domínios fuzzy, a utilização ocorre somente na etapa de pré-processamento, durante a geração de transações estendidas. Além do viés de utilização de transações estendidas, que podem levar a geração de um volume de regras superior ao caso tradicional, é possível notar que, em domínios nebulosos, as pesquisas dão enfoque apenas à mineração de regras fuzzy, deixando de lado a exploração de diferentes graus de especialização/generalização em taxonomias. Nesse sentido, este trabalho propõem o algoritmo FOntGAR, um novo algoritmo para mineração de regras de associação generalizadas com itens presentes em qualquer nível de ontologias compostas por graus de especialização/generalização variando no intervalo [0,1] (ontologias de conceitos fuzzy), em etapa de pós-processamento. Objetivando obter maior enriquecimento semântico, as regras geradas pelo algoritmo também podem possuir relações de similaridade, de acordo com contextos pré-definidos. Outros pontos relevantes são a especificação de uma nova abordagem de generalização (incluindo um novo tratamento de agrupamento das regras), e um novo e eficiente método para calcular o suporte estendido das regras generalizadas durante a etapa mencionada.

Avec l’arrivée des objets connectés, la reconnaissance de l’activité physique connait une nouvelle ère. De nouvelles considérations sont à prendre en compte afin d’aboutir à un meilleur processus de traitement. Dans cette thèse, nous avons exploré le processus de traitement pour la reconnaissance de l’activité physique dans un environnement non contrôlé. Les activités physiques reconnues, avec seulement une centrale inertielle (accéléromètre, gyroscope et magnétomètre), sont dites élémentaires. Les autres types d’activités dépendantes d’un contexte sont dites « basés sur le contexte ». Nous avons extrait la transformée en cosinus discrète (DCT) comme principal descripteur pour la reconnaissance des activités élémentaires. Afin de reconnaitre les activités physiques basées sur le contexte, nous avons défini trois niveaux de granularité : un premier niveau dépendant des objets connectés embarqués (smartphone, smartwatch et samrt TV). Un deuxième niveau concerne l’étude des comportements des participants en interaction avec l’écran de la smart TV. Le troisième niveau concerne l’étude de l’attention des participants envers la TV. Nous avons pris en considération l’aspect imperfection des données en fusionnant les données multi capteurs avec le modèle de Dempster-Shafer. A ce titre, nous avons proposé différentes approches pour calculer et approximer les fonctions de masse. Afin d’éviter de calculer et sélectionner les différents descripteurs, nous avons proposé une approche basée sur l’utilisation d’algorithmes d’apprentissage en profondeur (DNN). Nous avons proposé deux modèles : un premier modèle consiste à reconnaitre les activités élémentaires en sélectionnant la DCT comme principal descripteur (DNN-DCT). Le deuxième modèle consiste à apprendre les données brutes des activités basées sur le contexte (CNN-brutes). L’inconvénient du modèle DNN-DCT est qu’il est rapide mais moins précis, alors que le modèle CNN-brutes est plus précis mais très lent. Nous avons proposé une étude empirique permettant de comparer les différentes méthodes pouvant accélérer l’apprentissage tout en gardant un niveau élevé de précision. Nous avons ainsi exploré la méthode d’optimisation par essaim particulaires (PSO). Les résultats sont très satisfaisants (97%) par rapport à l’apprentissage d’un réseau de neurones profond avec les méthodes d’optimisation classiques telles que la descente de Gradient Stochastique et l’optimisation par Gradient accéléré de Nesterov. Les résultats de nos travaux suggèrent le recours à de bons descripteurs dans le cas où le contexte n’importe peu, la prise en compte de l’imperfection des données capteurs quand le domaine sous-jacent l’exige, l’utilisation de l’apprentissage profond avec un optimiseur permettant d’avoir des modèles très précis et plus rapides. / With the arrival of connected objects, the recognition of physical activity is experiencing a new era. New considerations need to be taken into account in order to achieve a better treatment process. In this thesis, we explored the treatment process for recognizing physical activity in an uncontrolled environment. The recognized physical activities, with only one inertial unit (accelerometer, gyroscope and magnetometer), are called elementary. Other types of context-dependent activities are called "context-based". We extracted the DCT as the main descriptor for the recognition of elementary activities. In order to recognize the physical activities based on the context, we defined three levels of granularity: a first level depending on embedded connected objects (smartphone, smartwatch and samrt TV . A second level concerns the study of participants' behaviors interacting with the smart TV screen. The third level concerns the study of participants' attention to TV. We took into consideration the imperfection aspect of the data by merging the multi sensor data with the Dempster-Shafer model. As such, we have proposed different approaches for calculating and approximating mass functions. In order to avoid calculating and selecting the different descriptors, we proposed an approach based on the use of deep learning algorithms (DNN). We proposed two models: a first model consisting of recognizing the elementary activities by selecting the DCT as the main descriptor (DNN-DCT). The second model is to learn raw data from context-based activities (CNN-raw). The disadvantage of the DNN-DCT model is that it is fast but less accurate, while the CNN-raw model is more accurate but very slow. We have proposed an empirical study to compare different methods that can accelerate learning while maintaining a high level of accuracy. We thus explored the method of optimization by particle swarm (PSO). The results are very satisfactory (97%) compared to deep neural network with stochastic gradients descent and Nesterov accelerated Gradient optimization. The results of our work suggest the use of good descriptors in the case where the context matters little, the taking into account of the imperfection of the sensor data requires that it be used and faster models.

In this study, the aim has been to investigate how different educational materials in the nature science subjects are designed with the intention of how many and what types of laboratory activities the pupils will encounter to contextualize the abstract nature of science to develop their understanding of the content and working methods of science. In order for the pupils to have the opportunity to develop a scientific understanding, it is necessary that they have the opportunity to use their senses to interpret the text that is presented to them. Through various sensual, experimental, and practical activities, pupils can gain an understanding of the abstract phenomena, concepts, and explanatory models of natural science. In today's schools, there are many teachers who use educational materials as their point of departure when teaching science. Educational materials have an important role in science teaching, but depending on how the educational materials are designed, they provide different conditions for learning. In this study, the number of laboratory activities, that appear in different educational materials for science teaching, is investigated, as well as whether these laboratory activities are designed in terms of their degree of freedom and openness. The following questions have been addressed in order to achieve the purpose of the study: - How many laboratory activities can be found in the various educational materials?-  Are there more laboratory activities in any of the three nature-oriented subjects, biology, physics, or chemistry? - What types of laboratory activities do the pupils encounter in the educational materials and what degree of openness and freedom do they include? The investigation has been carried out through a qualitative and quantitative text and content analysis where various educational materials have been analyzed based on the context-based learning (CBL) theory and the theory of laboratory classification. The results of the qualitative and quantitative text and content analysis showed that certain educational materials give the pupils more opportunities to connect the scientific content to real experiences. The results also showed that some educational materials have a greater variety of different laboratory forms where the pupils have the opportunity to develop an understanding of the different working methods and content of the natural sciences. The conclusion that can be drawn after this study is therefore that it is important that teachers carefully review the various2educational materials to gain insight into their design. Through insight into the design of the educational materials, the teacher can then supplement the educational materials with, for example, laboratories of varying degrees of openness and freedom to give the pupils the opportunity to fulfill the goals in the curriculum.

The aim of this study was to investigate the effect of context-based problem solving instruction (CBPSI) on the problem solving performance of Grade 10 learners, who performed poorly in mathematics. A cognitive load theory (CLT) was used to frame the study. In addition, CLT was used to: 1) facilitate the interpretation and explanation of participants‟ problem solving performance; and, 2) influence the design of CBPSI to hone participants‟ problem solving skills. The study was conducted in the Gauteng province of South Africa and involved a two-week intervention program in each of the nine participating high schools. Participants consisted of 783 learners and four Grade 10 mathematics teachers. A non-equivalent control group design was employed, consisting of a pre- and post- measure. In addition, classroom observations and semi-structured interviews were conducted with teachers and learners. Teachers employed conventional problem solving instructions in four control schools while the researcher implemented CBPSI in five experimental schools. Instruction in experimental schools entailed several worked-out context-based problem solving examples given to participants in worksheets. The main aspects of CBPSI embraced elements of the effects of self-explanation and split-attention, as advocated by CLT. Due to the design of CBPSI participants in experimental schools became familiar with the basic context-based problem solving tasks that were presented to them through the worked-out example samples. In turn, the associated cognitive load of problem solving tasks was gradually reduced. The principal instrument for data collection was a standardized Functional Mathematics Achievement Test. The pre-test determined participants‟ initial problem solving status before intervention. A post-test was given at the end of intervention to benchmark change in the functionality of CBPSI over a two-week period. Using one-way analysis of covariance (ANCOVA), Analysis of Variance (ANOVA), and other statistical techniques the study found that participants in experimental schools performed significantly better than participants in control schools on certain aspects of problem solving performance. In addition, semi-structured interviews and classroom observations revealed that participants rated CBPSI highly. On the whole, the study showed that CBPSI is an effective instructional tool to enhance the problem solving performance of Grade 10 mathematics learners. / Mathematics Education / D. Phil. (Mathematics, Science and Technology Education)

The aim of this study was to investigate the effect of context-based problem solving instruction (CBPSI) on the problem solving performance of Grade 10 learners, who performed poorly in mathematics. A cognitive load theory (CLT) was used to frame the study. In addition, CLT was used to: 1) facilitate the interpretation and explanation of participants‟ problem solving performance; and, 2) influence the design of CBPSI to hone participants‟ problem solving skills. The study was conducted in the Gauteng province of South Africa and involved a two-week intervention program in each of the nine participating high schools. Participants consisted of 783 learners and four Grade 10 mathematics teachers. A non-equivalent control group design was employed, consisting of a pre- and post- measure. In addition, classroom observations and semi-structured interviews were conducted with teachers and learners. Teachers employed conventional problem solving instructions in four control schools while the researcher implemented CBPSI in five experimental schools. Instruction in experimental schools entailed several worked-out context-based problem solving examples given to participants in worksheets. The main aspects of CBPSI embraced elements of the effects of self-explanation and split-attention, as advocated by CLT. Due to the design of CBPSI participants in experimental schools became familiar with the basic context-based problem solving tasks that were presented to them through the worked-out example samples. In turn, the associated cognitive load of problem solving tasks was gradually reduced. The principal instrument for data collection was a standardized Functional Mathematics Achievement Test. The pre-test determined participants‟ initial problem solving status before intervention. A post-test was given at the end of intervention to benchmark change in the functionality of CBPSI over a two-week period. Using one-way analysis of covariance (ANCOVA), Analysis of Variance (ANOVA), and other statistical techniques the study found that participants in experimental schools performed significantly better than participants in control schools on certain aspects of problem solving performance. In addition, semi-structured interviews and classroom observations revealed that participants rated CBPSI highly. On the whole, the study showed that CBPSI is an effective instructional tool to enhance the problem solving performance of Grade 10 mathematics learners. / Mathematics Education / D. Phil. (Mathematics, Science and Technology Education)

The thesis shows soft power as one of the most effective strategies of global leadership. Through different methodological concepts author tries to classify instruments of soft power and surmount conceptual contradictions. The research of soft power’s political, cultural and economic instruments is based on the integrate analysis of South Korean foreign policy. Author tries to find the way to increase Republic of Korea geopolitical influence through the most powerful “soft” instruments. / В работе рассматривается комплекс методологических подходов к анализу концепта «мягкой силы» как одного из наиболее эффективных инструментов глобального влияния современных государств. Автор предлагает оригинальную классификацию ресурсов «мягкой силы» и новые пути преодоления концептуальных противоречий. На основе анализа внешней политики Республики Корея автор раскрывает специфику политических, экономических и культурных стратегий «мягкого» влияния указанной страны. Автор выявляет сильные и слабые стороны внешней политики Республики Корея, предлагает методы увеличения влияния южнокорейской «мягкой силы» на мировой арене.