Multi-sensitive microgel particles are covalently crosslinked polymeric chains with a colloidal dimension that can rapidly change their volume through various external stimuli such as pH, ionic strength, temperature, and magnetic and electric field. Due to these characteristics, increasing attention has been focus on the development of multi-sensitive microgels, mainly for application as drug delivery carriers. This study aimed to synthesise pH- and temperature-sensitive microgel particles, based on precipitation polymerisation of N-isopropylacrylamide and acrylic acid, with a defined size, narrow size distribution, spherical morphology and colloidal stability to be used as an oral drug carrier. In order to provide better biocompatibility and biodegradability in the synthesised material, some assays were performed, adding the chitosan biopolymer during the polymerisation of the monomers previously mentioned. This thesis also intended to contribute to a better understanding of the relationship between the microgel synthesis and their properties using spectroscopic techniques to monitor the microgel formation. A new approach was proposed to monitoring the process variables, monomer conversion and average particle size of the microgel particles, during precipitation polymerisation using NIR and UV-VIS-NIR high-resolution spectrophotometers coupled with a probe. Besides, the influence of reaction conditions in the physicochemical characteristics of microgel particles was extensively investigated as well as their potential as an oral drug carrier for insulin. The monitoring results pointed out the enormous potential of these spectroscopy techniques to monitor the precipitation polymerisation process, allowing control over the polymerisation reaction with quickly and directly acquisition of data in real-time. In general, pH- and temperature-sensitive microgels were successfully synthesised, and many formulations showed to be suitable for application as oral drug carriers. However, the preliminary in vitro release results were not satisfactory, and a more in-depth study between the interaction of the drug with the particles as well as the method of release is recommended. In conclusion, with a proper understanding of the influence of the process conditions (e.g., reagent concentrations) on the physicochemical properties of the microgels, it is possible to tailor the multi-sensitive microgels for the desired application. / Microgéis multisensíveis são partículas poliméricas covalentemente reticuladas, com uma dimensão coloidal, e com capacidade para mudar o seu volume através de vários estímulos externos, como por exemplo, pH, força iônica, temperatura e campos magnéticos ou elétricos. Devido a essas características, estas partículas têm recebido grande atenção, principalmente na aplicação como carreadores de fármacos. O objetivo deste estudo foi sintetizar partículas de microgéis sensíveis ao pH e à temperatura através da polimerização por precipitação dos monômeros N-isoproprilacrilamida e ácido acrílico, que tivessem tamanho definido, distribuição de tamanho estreita, morfologia esférica e estabilidade coloidal, para serem utilizadas como carreadores oral de fármacos. Com o intuito de proporcionar melhor biocompatibilidade e biodegradabilidade no material sintetizado, foram realizados alguns ensaios acrescentando o biopolímero quitosana durante a polimerização dos monômeros citados anteriormente. O presente trabalho também teve como objetivo contribuir para uma melhor compreensão da relação entre a síntese de microgéis e suas propriedades utilizando técnicas espectroscópicas para monitorar a formação dos microgéis. Uma nova abordagem foi proposta para monitorar as variáveis de processo, conversão de monômero e tamanho médio das partículas de microgel, durante a polimerização por precipitação através de espectrofotômetros NIR e UV-VIS-NIR acoplados a uma sonda. Além disso, a influência das condições da reação nas características físico-químicas das partículas de microgel foi extensivamente investigada, bem como o seu potencial como carreador oral de insulina. Os resultados do monitoramento indicaram o enorme potencial das técnicas espectroscópicas utilizadas, permitindo o controle da polimerização com aquisição rápida e direta dos dados em tempo real. Em geral, os microgéis sensíveis ao pH e à temperatura foram sintetizados com sucesso, e algumas formulações mostraram-se adequadas para aplicação como veículos de fármaco através da via oral. No entanto, os resultados preliminares da liberação in vitro não foram satisfatórios, e um estudo mais aprofundado entre a interação do fármaco com as partículas e o método de liberação é recomendado. Em conclusão, com uma compreensão adequada da influência das condições do processo (nesse caso, da concentração dos reagentes) nas propriedades físico-químicas dos microgéis, é possível ajustá-las para obtenção de microgéis multissensíveis com características adequadas para a aplicação desejada.

Online supervised learning from fast-evolving imbalanced data streams has applications in many areas. That is, the development of techniques that are able to handle highly skewed class distributions (or 'class imbalance') is an important area of research in domains such as manufacturing, the environment, and health. Solutions should be able to analyze large repositories in near real-time and provide accurate models to describe rare classes that may appear infrequently or in bursts while continuously accommodating new instances. Although numerous online learning methods have been proposed to handle binary class imbalance, solutions suitable for multi-class streams with varying degrees of imbalance in evolving streams have received limited attention. To address this knowledge gap, the first contribution of this thesis introduces the Online Learning from Imbalanced Multi-Class Streams through Dynamic Sampling (DynaQ) algorithm for learning in such multi-class imbalanced settings. Our approach utilizes a queue-based learning method that dynamically creates an instance queue for each class. The number of instances is balanced by maintaining a queue threshold and removing older samples during training. In addition, new and rare classes are dynamically added to the training process as they appear. Our experimental results confirm a noticeable improvement in minority-class detection and classification performance. A comparative evaluation shows that the DynaQ algorithm outperforms the state-of-the-art approaches. Our second contribution in this thesis focuses on fairness-aware learning from imbalanced streams. Our work is motivated by the observation that the decisions made by online learning algorithms may negatively impact individuals or communities. Indeed, the development of approaches to handle these concerns is an active area of research in the machine learning community. However, most existing methods process the data in offline settings and are not directly suitable for online learning from evolving data streams. Further, these techniques fail to take the effects of class imbalance, on fairness-aware supervised learning into account. In addition, recent fairness-aware online learning supervised learning approaches focus on one sensitive attribute only, which may lead to subgroup discrimination. In a fair classification, the equality of fairness metrics across multiple overlapping groups must be considered simultaneously. In our second contribution, we thus address the combined problem of fairness-aware online learning from imbalanced evolving streams, while considering multiple sensitive attributes. To this end, we introduce the Multi-Sensitive Queue-based Online Fair Learning (MQ-OFL) algorithm, an online fairness-aware approach, which maintains valid and fair models over evolving streams. MQ-OFL changes the training distribution in an online fashion based on both stream imbalance and discriminatory behavior of the model evaluated over the historical stream. We compare our MQ-OFL method with state-of-art studies on real-world datasets and present comparative insights on the performance. Our final contribution focuses on explainability and interpretability in fairness-aware online learning. This research is guided by the concerns raised due to the black-box nature of models, concealing internal logic from users. This lack of transparency poses practical and ethical challenges, particularly when these algorithms make decisions in finance, healthcare, and marketing domains. These systems may introduce biases and prejudices during the learning phase by utilizing complex machine learning algorithms and sensitive data. Consequently, decision models trained on such data may make unfair decisions and it is important to realize such issues before deploying the models. To address this issue, we introduce techniques for interpreting the outcomes of fairness-aware online learning. Through a case study predicting income based on features such as ethnicity, biological sex, age, and education level, we demonstrate how our fairness-aware learning process (MQ-OFL) maintains a balance between accuracy and discrimination trade-off using global and local surrogate models.

Im Mittelpunkt dieser Arbeit stand die Darstellung sensitiver Blockcopolymere und deren Gele, die als Ausgangsmaterialien in Sensor- und Aktorsystemen einsetzbar sind. Die Vereinigung verschiedener Ansprechparameter stellt erhöhte Anforderung an die Synthese. Geringe Ansprechzeiten lassen sich mit einer Gelgröße im µm-Bereich erreichen. Hydrogele dieser Größenordnungen können durch nachträgliche Vernetzung funktioneller linearer Polymere ermöglicht werden. Die Makroinitiatormethode ermöglichte den Aufbau verschiedener linearer photovernetzbarer Blockcopolymere. Zum Einen wurde das temperatursensitive P(n-BuAc)-block-P(PNIPAAm-co-DMIAAm) erhalten, des Weiteren gelang die Darstellung der multi-sensitiven Blockcopolymere P2VP-block-P(NIPAAm-co-DMIAAm) und P4VP-block-P(NIPAAm-co-DMIAAm). Die Blockcopolymere wurden mit variierenden Blocklängen und Verhältnissen sowie mit unterschiedlichem Vernetzergehalt dargestellt. Die Charakterisierung der Blockcopolymere erfolgte mittels 1H-NMR-Spektroskopie, GPC-Messungen (Zusammensetzung) und DSC-Messungen (thermische Eigenschaften). Das Löslichkeitsverhalten in wässrigen Medien wurde durch Dynamische Lichtstreuung bestimmt. Die Beschreibung des Quellverhaltens der vernetzten Schichten erfolgte durch vornehmlich durch optische Methoden (SPR/OWS, WAMS, Ellipsometrie). Die Veränderung des E-Moduls in Abhängigkeit äußerer Parameter konnte mittels AFM untersucht werden. Die Reaktion der Schichten wurde gegenüber Temperatur, pH-Wert und Salzkonzentrationen getestet. Die charakterisierten Filme konnten im Anschluss als sensitive Schichten in piezoresistiven Sensorsystemen verwendetet werden.

Im Mittelpunkt dieser Arbeit stand die Darstellung sensitiver Blockcopolymere und deren Gele, die als Ausgangsmaterialien in Sensor- und Aktorsystemen einsetzbar sind. Die Vereinigung verschiedener Ansprechparameter stellt erhöhte Anforderung an die Synthese. Geringe Ansprechzeiten lassen sich mit einer Gelgröße im µm-Bereich erreichen. Hydrogele dieser Größenordnungen können durch nachträgliche Vernetzung funktioneller linearer Polymere ermöglicht werden. Die Makroinitiatormethode ermöglichte den Aufbau verschiedener linearer photovernetzbarer Blockcopolymere. Zum Einen wurde das temperatursensitive P(n-BuAc)-block-P(PNIPAAm-co-DMIAAm) erhalten, des Weiteren gelang die Darstellung der multi-sensitiven Blockcopolymere P2VP-block-P(NIPAAm-co-DMIAAm) und P4VP-block-P(NIPAAm-co-DMIAAm). Die Blockcopolymere wurden mit variierenden Blocklängen und Verhältnissen sowie mit unterschiedlichem Vernetzergehalt dargestellt. Die Charakterisierung der Blockcopolymere erfolgte mittels 1H-NMR-Spektroskopie, GPC-Messungen (Zusammensetzung) und DSC-Messungen (thermische Eigenschaften). Das Löslichkeitsverhalten in wässrigen Medien wurde durch Dynamische Lichtstreuung bestimmt. Die Beschreibung des Quellverhaltens der vernetzten Schichten erfolgte durch vornehmlich durch optische Methoden (SPR/OWS, WAMS, Ellipsometrie). Die Veränderung des E-Moduls in Abhängigkeit äußerer Parameter konnte mittels AFM untersucht werden. Die Reaktion der Schichten wurde gegenüber Temperatur, pH-Wert und Salzkonzentrationen getestet. Die charakterisierten Filme konnten im Anschluss als sensitive Schichten in piezoresistiven Sensorsystemen verwendetet werden.