Machine learning (ML) is transforming a wide range of applications, promising to bring immense economic and social benefits. However, it also raises substantial security, privacy, and transparency challenges. ML workloads indeed push companies toward aggressive data collection and loose data access policies, placing troves of sensitive user information at risk if the company is hacked. ML also introduces new attack vectors, such as adversarial example attacks, which can completely nullify models’ accuracy under attack. Finally, ML models make complex data-driven decisions, which are opaque to the end-users, and difficult to inspect for programmers. In this dissertation we describe three systems we developed. Each system addresses a dimension of the previous challenges, by combining new practical systems techniques with rigorous theory to achieve a guaranteed level of protection, and make systems easier to understand. First we present Sage, a differentially private ML platform that enforces a meaningful protection semantic for the troves of personal information amassed by today’s companies. Second we describe PixelDP, a defense against adversarial examples that leverages differential privacy theory to provide a guaranteed level of accuracy under attack. Third we introduce Sunlight, a tool to enhance the transparency of opaque targeting services, using rigorous causal inference theory to explain targeting decisions to end-users.

O objetivo deste trabalho é refletir sobre critérios para a organização do conhecimento do acervo do Museu Afro-Brasileiro da Universidade Federal da Bahia, o MAFRO, recorrendo a instrumentos de gestão, dentre os quais os meios e instrumentos para uma padronização da documentação. Considera a dimensão social do acervo e discute estratégias para implementar um sistema de documentação na perspectiva da informação enquanto conhecimento, recorrendo a procedimentos que visam representar o acervo, bem como a recuperação e socialização das informações, sem esquecer a garantia da segurança e preservação. É também seu objetivo adequar parâmetros para um melhor desempenho da plataforma de disponibilização virtual das coleções mantidas pela instituição. A pesquisa foi desenvolvida com base nas observações realizadas durante a etapa de pré-produção da sistematização do acervo, bem como nas reflexões realizadas a partir da literatura. Parte do entendimento de que a gestão de acervo envolve estabelecer critérios sobre aquilo que deve ser memorizado, enquanto uma ação política que permeia toda e qualquer atividade relacionada à guarda, proteção, conservação, documentação e disseminação da informação. Realiza um relato sobre uma experiência concreta de tratamento da informação de coleções do acervo MAFRO, e incorpora inquietações sobre a contextualização de seus objetos, além de observações sobre as recomendações internacionais relativas à documentação em museus. Do ponto de vista metodológico, a pesquisa é exploratória e descritiva, com uma abordagem etnográfica, em consonância com o caráter das coleções abrigadas no MAFRO. Aborda o objeto de museu, sua conceituação e aspectos de sua documentação e gestão utilizando contribuições da Museologia e da Ciência da Informação. Observa aspectos da organização do conhecimento - a catalogação, a classificação e a indexação - enfatizando a importância da padronização. Aborda o ambiente experimental com a apresentação do MAFRO e destaca a importância de Pierre Verger no que toca à aquisição do acervo e a expografia e, em seguida, as ações práticas desenvolvidas na reserva técnica do MAFRO. A discussão é adensada com a reflexão desenvolvida no Mestrado Profissional, quer a partir da exploração de bibliografia, como do exercício de sistematização crítica de uma experiência concreta. / The objective of this work is to reflect on criteria for the organization of knowledge of the collection of the Afro-Brazilian Museum of the Federal University of Bahia, MAFRO / UFBA (MAFRO), using management tools, among them the means and instruments for a standardization of documentation. It considers the social dimension of the collection and discusses strategies to implement a documentation system in the perspective of information as knowledge, using procedures that represent the collection, as well as the recovery and socialization of information, not forgetting guaranteeing the security and preservation. It is also its objective to adjust parameters for a better performance of the platform of virtual availability of the collections maintained by the institution. The research was developed based on the observations made during the pre-production stage of the systematization of the collection, as well as in the reflections made from the literature. It is part of the understanding that the collection management involves establishing criteria about what should be memorized, as a political action that permeates all and any activity related to the safeguarding, protection, conservation, documentation and dissemination of information. It reports on a concrete experience of processing information from collections of the MAFRO collection, and incorporates concerns about the contextualization of its objects, as well as observations on international recommendations regarding documentation in museums. From the methodological point of view, the research is exploratory and descriptive, with an ethnographic approach, in consonance with the character of the collections housed in MAFRO. It approaches the museum\'s object, its conceptualization and aspects of its documentation and management using contributions from Museology and Information Science. It observes aspects of the organization of knowledge - cataloging, classification and indexing - emphasizing the importance of standardization. It approachesthe experimental environment with the presentation of MAFRO and highlights the importance of Pierre Verger in acquiring the collection and expography and then the practical actions developed in the MAFRO technical reserve. The discussion is thickened with the reflection developed in the Professional Master, both from the exploration of bibliography and from the exercise of critical systematization of a concrete experience.

The household has always been a place of information consumption. While much is known about the modern household as a consumer of information and adopter of information-related technologies, virtually nothing is known about how the household -as a collectivity - processes and manages its information. This research represents the first systematic study of the household as a human information system and presents an understandingtype theory of household information practices. Using dimensional analysis (including theoretical sampling) as proposed by Schatzman (1991), the information practices of eleven households (28 householders) were explored. Each householder completed a questionnaire to identify the information and information-related devices and services used in the household. This was followed by an unstructured group interview that explored the information practices in the household. Analysis of the questionnaires and interviews revealed that within a household it is individual householders who process and manage information. They do so because information has a role in their life, that is, particular information is meaningful for them. Each householder thus devises their own individual information practices, representing the unique way in which a householder comes to live with information. Household information practices are the sum of these individual information practices. They are mediated by two enabling processes - taking charge and negotiating - and nine dimensions of action. Six of the dimensions affect the information practices of individual householders - which are themselves a dimension - while the final two represent, respectively, the consequences of living with and without information. All processes and dimensions operate against, and interact with, a changing structural context of information, technology and society. As a result, each household has a characteristic mode for dealing with information. The centrality of negotiation in developing household information practices indicates that the practices are socially constructed and represent an emergent phenomenon. This, in turn, suggests that the household as an information system is a negotiated order. Insights from the research can be applied to other types of organisations and other aspects of the Information Systems discipline. Areas of further work are identified to expand upon the exploratory nature of this research.

The information age has brought with it a dual problem. In the first place, the ready access to mechanisms to capture and store vast amounts of data in all forms (text, audio, image and video), has resulted in a continued demand for ever more efficient means to store and transmit this data. In the second, the rapidly increasing store demands effective means to structure and access the data in an efficient and meaningful manner. In terms of audio data, the first challenge has traditionally been the realm of audio compression research that has focused on statistical, unstructured audio representations that obfuscate the inherent structure and semantic content of the underlying data. This has only served to further complicate the resolution of the second challenge resulting in access mechanisms that are either impractical to implement, too inflexible for general application or too low level for the average user. Thus, an artificial dichotomy has been created from what is in essence a dual problem. The founding motivation of this thesis is that, although the hypermedia model has been identified as the ideal, cognitively justified method for organising data, existing audio data representations and coding models provide little, if any, support for, or resemblance to, this model. It is the contention of the author that any successful attempt to create hyperaudio must resolve this schism, addressing both storage and information management issues simultaneously. In order to achieve this aim, an audio representation must be designed that provides compact data storage while, at the same time, revealing the inherent structure of the underlying data. Thus it is the aim of this thesis to present a representation designed with these factors in mind. Perhaps the most difficult hurdle in the way of achieving the aims of content-based audio coding and information management is that of auditory source separation. The MPEG committee has noted this requirement during the development of its MPEG-7 standard, however, the mechanics of "how" to achieve auditory source separation were left as an open research question. This same committee proposed that MPEG-7 would "support descriptors that can act as handles referring directly to the data, to allow manipulation of the multimedia material." While meta-data tags are a part solution to this problem, these cannot allow manipulation of audio material down to the level of individual sources when several simultaneous sources exist in a recording. In order to achieve this aim, the data themselves must be encoded in such a manner that allows these descriptors to be formed. Thus, content-based coding is obviously required. In the case of audio, this is impossible to achieve without effecting auditory source separation. Auditory source separation is the concern of computational auditory scene analysis (CASA). However, the findings of CASA research have traditionally been restricted to a limited domain. To date, the only real application of CASA research to what could loosely be classified as information management has been in the area of signal enhancement for automatic speech recognition systems. In these systems, a CASA front end serves as a means of separating the target speech from the background "noise". As such, the design of a CASA-based approach, as presented in this thesis, to one of the most significant challenges facing audio information management research represents a significant contribution to the field of information management. Thus, this thesis unifies research from three distinct fields in an attempt to resolve some specific and general challenges faced by all three. It describes an audio representation that is based on a sinusoidal model from which low-level auditory primitive elements are extracted. The use of a sinusoidal representation is somewhat contentious with the modern trend in CASA research tending toward more complex approaches in order to resolve issues relating to co-incident partials. However, the choice of a sinusoidal representation has been validated by the demonstration of a method to resolve many of these issues. The majority of the thesis contributes several algorithms to organise the low-level primitives into low-level auditory objects that may form the basis of nodes or link anchor points in a hyperaudio structure. Finally, preliminary investigations in the representation’s suitability for coding and information management tasks are outlined as directions for future research.

This thesis presents an extension of the contemporaryengineering design theory towards a unified view onsimultaneous development of products and manufacturing systems,i.e. concurrent engineering. The traditional engineering design theory explains therealization of a product design as a development of productstructure from four perspectives: technical process, function,technical solution, and physical embodiment. This thesisextends the engineering design theory with a set of definitionsand universal statements. These definitions and universalstatements describe manufacturing systems from same fourperspectives. In that context they also describe therelationship between a product and its manufacturing system.The thesis contributes to the creation of a single theoreticalsystem based on an integration of theories from two engineeringdesign schools, the WDK and the Axiomatic Design. WDKtheoriesare in this new context utilized for qualitative synthesis ofthe developed artifacts, while the Axiomatic Design is utilizedfor structuring and analyzing the corresponding quantitativeparameters. The definitions and universal statements describe thedevelopment structures for productsand manufacturing systems.This description is utilized for definition of a system fordevelopment of these structures, i.e. (i) a stage-gate-basedmanufacturing system development process, (ii) a developmentmethodology toolbox, and (iii) an information managementframework consisted of an information model harmonized with thesystems engineering data management standard STEP AP 233. The research has been carried out in a close collaborationwith Swedish manufacturing industry. The utilized researchmethodology is the hypothetic- deductive method, with casestudy as an observation method. Keywords:Concurrent Engineering, Engineering Design,Development Methods and Tools, Manufacturing System,Information Management.

The study investigates knowledge management (KM) practices in the public sector in Botswana. The underlying premise is that good KM leads to efficiency and effectiveness. The study assesses the KM practices in the government departments by means of a questionnaire survey of senior managers. The assumption is that the corporate manager/directors will know what knowledge is there, how knowledge is created, shared and flow in the organization. The main question of this study is whether the Botswana public sector is practicing KM. The study explores the problem and questions by means of a questionnaire survey amongst 43 departmental directors of the Government of Botswana. The overall finding is that information management rather than KM is being practiced. The respondents, senior public service managers, certainly recognize the value of and the need for KM. But, they themselves identify certain weaknesses, such as lack of knowledge of KM among their staff, weak communication inside and across the departments, lack of policy and lack of good KM systems.

<p>The study investigates knowledge management (KM) practices in the public sector in Botswana. The underlying premise is that good KM leads to efficiency and effectiveness. The study assesses the KM practices in the government departments by means of a questionnaire survey of senior managers. The assumption is that the corporate manager/directors will know what knowledge is there, how knowledge is created, shared and flow in the organization. The main question of this study is whether the Botswana public sector is practicing KM. Related questions are:&nbsp / What are the views of public service managers/ directors on the benefits that can be reaped from KM practices?&nbsp / What evidence is there that the public service has a culture of sharing information and knowledge? And how are staff members encouraged to internalize and use new knowledge? How are creativity and new ideas encouraged?&nbsp / Are there appropriate technological resources to facilitate effective KM, for example central knowledge repositories and social networking?The study explores the problem and questions by means of a questionnaire survey amongst 43 departmental directors of the Government of Botswana. The overall finding is that information management rather than KM is being practiced. The respondents, senior public service managers, certainly recognize the value of and the need for KM. But, they themselves identify certain weaknesses, such as lack of knowledge of KM among their staff, weak communication inside and across the departments, lack of policy and lack of good KM systems.<br /> .</p>

Many Library and Information Science (LIS) scholars have long articulated the importance of physical and social settings—the environment—when examining how individuals acquire, store, organize, maintain, dispose of, and use information in one of their home or work lives. Yet, few have raised the question of how these information practices are altered and affected in home office spaces, fused living and working environments that lie at the intersection of the personal and the professional. This thesis resulted from an exploratory, ethnographic research study centred upon describing and analyzing the habits of information management and information use that characterize home office settings—specifically, professional home offices that each serve as their user’s only workplace. It argues that the professional home office differs from both traditional professional offices in corporate or institutional settings and from personal home offices used for non-professional tasks and pursuits. The professional home offices of four printing company account managers provided the field from which data was gathered, collected by way of guided tours, diagramming, photography, interviews, and observation. Findings suggest that information practices in professional home offices are a continual negotiation between the two spheres of household and organization, but that this will not necessarily imply a compromise of one for the other.

Many Library and Information Science (LIS) scholars have long articulated the importance of physical and social settings—the environment—when examining how individuals acquire, store, organize, maintain, dispose of, and use information in one of their home or work lives. Yet, few have raised the question of how these information practices are altered and affected in home office spaces, fused living and working environments that lie at the intersection of the personal and the professional. This thesis resulted from an exploratory, ethnographic research study centred upon describing and analyzing the habits of information management and information use that characterize home office settings—specifically, professional home offices that each serve as their user’s only workplace. It argues that the professional home office differs from both traditional professional offices in corporate or institutional settings and from personal home offices used for non-professional tasks and pursuits. The professional home offices of four printing company account managers provided the field from which data was gathered, collected by way of guided tours, diagramming, photography, interviews, and observation. Findings suggest that information practices in professional home offices are a continual negotiation between the two spheres of household and organization, but that this will not necessarily imply a compromise of one for the other.

This research takes steps towards developing a new theory of organizational information management based on the idea that information creates ordering effects in transactions and on the idea that there are multiple centers of authority in organizations. The rationale for developing this theory is the empirical observation that hospitals have great difficulty in managing information relating to transactions with patients. The research illustrates the detailed workings of an initial conceptual framework based on an action research project into the revenue cycle of a hospital. The framework facilitates a deeper understanding of how information technology can help transform information management practices in complex organizations, such as hospitals. At the same time, this research adds to the literature on Polycentricity Theory by linking its two core concepts – multiple nested centers of decision-making and context-dependent governance – with Transaction Cost Theory and information management theories to establish a new foundation for understanding the role of information technology in organizational contexts.