Prédire les performances des requêtes et expliquer les résultats pour assister la consommation de données liées. Notre objectif est d'aider les utilisateurs à comprendre les performances d'interrogation SPARQL, les résultats de la requête, et dérivations sur les données liées. Pour aider les utilisateurs à comprendre les performances des requêtes, nous fournissons des prévisions de performances des requêtes sur la base de d’historique de requêtes et d'apprentissage symbolique. Nous n'utilisons pas de statistiques sur les données sous-jacentes à nos prévisions. Ce qui rend notre approche appropriée au Linked Data où les statistiques sont souvent absentes. Pour aider les utilisateurs des résultats de la requête dans leur compréhension, nous fournissons des explications de provenance. Nous présentons une approche sans annotation pour expliquer le “pourquoi” des résultats de la requête. Notre approche ne nécessite pas de reconception du processeur de requêtes, du modèle de données, ou du langage de requête. Nous utilisons SPARQL 1.1 pour générer la provenance en interrogeant les données, ce qui rend notre approche appropriée pour les données liées. Nous présentons également une étude sur les utilisateurs montrant l'impact des explications. Enfin, pour aider les utilisateurs à comprendre les dérivations sur les données liées, nous introduisons le concept d’explications liées. Nous publions les métadonnées d’explication comme des données liées. Cela permet d'expliquer les résultats en suivant les liens des données utilisées dans le calcul et les liens des explications. Nous présentons une extension de l'ontologie PROV W3C pour décrire les métadonnées d’explication. Nous présentons également une approche pour résumer ces explications et aider les utilisateurs à filtrer les explications. / Our goal is to assist users in understanding SPARQL query performance, query results, and derivations on Linked Data. To help users in understanding query performance, we provide query performance predictions based on the query execution history. We present a machine learning approach to predict query performances. We do not use statistics about the underlying data for our predictions. This makes our approach suitable for the Linked Data scenario where statistics about the underlying data is often missing such as when the data is controlled by external parties. To help users in understanding query results, we provide provenance-based query result explanations. We present a non-annotation-based approach to generate why-provenance for SPARQL query results. Our approach does not require any re-engineering of the query processor, the data model, or the query language. We use the existing SPARQL 1.1 constructs to generate provenance by querying the data. This makes our approach suitable for Linked Data. We also present a user study to examine the impact of query result explanations. Finally to help users in understanding derivations on Linked Data, we introduce the concept of Linked Explanations. We publish explanation metadata as Linked Data. This allows explaining derived data in Linked Data by following the links of the data used in the derivation and the links of their explanation metadata. We present an extension of the W3C PROV ontology to describe explanation metadata. We also present an approach to summarize these explanations to help users filter information in the explanation, and have an understanding of what important information was used in the derivation.

In this thesis, I aim to motivate a particular philosophy of mathematics characterised by the following three claims. First, mathematical sentences are generally speaking false because mathematical objects do not exist. Second, people typically use mathematical sentences to communicate content the truth of which does not require mathematical objects to exist. Finally, in using mathematical language in this way, speakers are not doing anything out of the ordinary: they are performing straightforward assertions. In Part I, I argue that the role played by mathematics in our scientific explanations is a purely expressive one, merely allowing us to say more than we otherwise would be able to about, or yielding a greater understanding of, the physical world. Mathematical objects to not need to exist for mathematical language to play this role. This proposal puts a normative constraint on our use of mathematical language: we ought to use mathematically presented theories to express belief only in the consequences they have for non-mathematical things. In Part II, I will argue that what the normative proposal recommends is in fact what people generally do in both pure and applied mathematical contexts. I motivate this claim by showing that it is predicted by our best general means of analysing natural language. I provide a semantic theory of applied arithmetical sentences and show that they do not purport to refer to numbers, as well as a pragmatic theory for pure mathematical language use which shows that pure mathematical utterances do not typically communicate content that implies the existence of mathematical objects. In conclusion, I show the hermeneutic fictionalist position that emerges is preferable to any alternative which interprets mathematical discourse as aimed at describing a domain of independently existing abstract mathematical objects.

This essay discusses and criticizes Richard Swinburne's inductive argument for the existence of God. In his The Existence of God, Swinburne aims at showing that the existence of God is more probable than not. This is an argument taking into consideration the premises of all traditional arguments for the existence of God. Swinburne uses the phenomena and events that constitute the premises of these arguments as evidence in an attempt to show that his hypothesis is more probably true than nor. Swinburne pursues this task by way of applying Bayes' theorem. The aim of this essay is normative - to judge the strength of Swinburne's argument for the existence of God. My primary objections towards Swinburne is that he professes a subjective concept of probability, that he relies too heavily on simplicity as a virtue of plausible and probable hypotheses and that his concept of God involves an incoherent picture of God's nature. I question not only the actual success of Swinburne's project but what his argument, if it had been successful, would have been able to establish. / Denna uppsats diskuterar och kritiserar Richard Swinburne's induktiva argument för Guds existens. I sin The Existence of God, försöker Swinburne visa att Guds existens är mer trolig än inte. Detta argument tar alla traditionella argument för Guds existens i beräknande.. Swinburne  använder de fenomen och händelser som utgör premisser för dessa argument som bevis i ett försök att visa att hans hypotes är mer trolig än inte.  Han genomför detta genom att använda sig av Bayes teorem. Syftet med denna uppsats är normativt - att bedöma styrkan I Swinburnes argument för Guds existens.  Mina primära invändningar mot Swinburne är att han  använder sig av ett subjektivt sannolikhetsbegrepp, att han förlitar sig allt för starkt på enkelhet som en förklaringsmässig dygd och att hans Gudsbegrepp inbegriper en inkoherent bild av Guds natur. Jag ifrågasätter den faktiska framgången i Swinburnes  projekt, och även vad Swinburne hade kunnat fastställa om hans projekt hade varit framgångsrikt.

No description available.

京都大学 / 0048 / 新制・課程博士 / 博士(情報学) / 甲第22806号 / 情博第736号 / 新制||情||126(附属図書館) / 京都大学大学院情報学研究科社会情報学専攻 / (主査)教授 吉川 正俊, 教授 森 信介, 教授 田島 敬史 / 学位規則第4条第1項該当 / Doctor of Informatics / Kyoto University / DFAM

This thesis aims to analyze the legal relevance and forms of instrumentalization of the right to explanation of automated decisions in the context of European Union Law. Under the prisms of trade secret protection, privacy and data protection, competition and consumer law, the investigation identifies legal provisions of primary and secondary European law, as well as complementary sources, which support the existence of a right to explanation. Additionally, the proportionality in the weighing of fundamental rights can corroborate the legal and technical implementation of this right, for which this thesis proposes practical suggestions that are consistent with the Google Search platform, chosen as the object of study through which the analyzes would be carried out. Revisiting Lawrence Lessig's idea that (computational) code would shape society's behavior, accommodated by Shoshana Zuboff's recent contributions on surveillance capitalism, I propose that law (norms) can also be translated into (programming) code. Many steps in this direction have already been taken and continue to be in recently proposed legislation. Robust laws in the areas of data protection, trade secrets protection, competition and consumer law currently subsidize claims for explanation that can be filed with different administrative bodies and be examined by European courts. Competition authorities have already proven that there can be real consequences in the reformulation of some of these practices in the private sector, as seen in previous cases of Google within the scope of the European Commission. Since there is a large asymmetry of information between automated decision-making platforms and its users, by disclosing a portion of an algorithm’s inner functioning (purposes, reasoning, inputs and deciding parameters taken into consideration etc.), in an appropriate fashion to the average user for whom the explanation is aimed, it is possible to better enforce consumer welfare and safeguard competition standards. The substance of this study recognizes the importance of a right to explanation as a stepping stone for algorithmic governance, especially with regard to Google’s search engine and its applications. / Doctorat en Sciences juridiques / info:eu-repo/semantics/nonPublished

Despite the strong efficacy of couple therapy, many couples still do not benefit from treatment. Marriage and family therapy scholars have argued that therapists play a crucial role in the delivery of successful couple therapy, yet little research has documented that the therapist in couple therapy has a significant impact on outcomes. Known as the study of therapist effects, this study sought to assess the amount of variance attributed to the therapist in couple therapy outcomes. Using dropout as the outcome variable, this study analyzed data from 1192 couples treated by 90 therapists at a university-based training clinic. Results from multilevel analysis indicated that therapists in the sample accounted for 9.5% of the variance in couple dropout while controlling for initial couple impairment. Therapist gender and therapist experience did not significantly predict the effectiveness of therapists. These findings give promise to future research on therapist effects in couple therapy and encourage exploration into which therapist characteristics and behaviors contribute to successful clinical outcomes.

In the last decades, the EU’s policy on education has become increasingly geared towards serving the economy. While some have been quick to label these changes as neoliberal, this is not such an obvious analysis considering that European education systems are still well-known for their welfare governance. This research attempts to clarify the relation between the EU educational policy and neoliberalism. Inspired by post-foundational discourse theory, we conducted a discourse analysis of three key European Union educational policy documents. Extending on a frame of 20 years, our analysis identifies an incremental neoliberalization of the policy as well as a number of conclusive similarities with neoliberal rationality. We contribute in showing how the economic dimension of the EU educational policy has not so much replaced the other objectives than it has incorporated them. We also show how contemporary educational policy continues the dynamic of neoliberalization, hinting a further dismantlement of the welfare states system in Europe. Our study stands to reinforce the conceptual link between European education and neoliberalism and as such contributes to the understudied ideational perspectives on the topic, complementing some shortcomings left by the mainstream theories, like neo-functionalism and liberal intergovernmentalism.

Le concept de “Business Rule Management System” (BRMS) a été introduit pour faciliter la création, la vérification, le déploiement et l'exécution des politiques commerciales propres à chaque compagnie. Basée sur une approche symbolique, l'idée générale est de permettre aux utilisateurs métier de gérer les changements des règles métier dans le système sans avoir besoin de recourir à des compétences techniques. Il s'agit donc de fournir à ces derniers la possibilité de formuler des politiques commerciales et d'automatiser leur traitement tout en restant proche du langage naturel. De nos jours, avec l'expansion de ce type de systèmes, il faut faire face à des logiques de décision de plus en plus complexes et à de larges volumes de données. Il n'est pas toujours facile d'identifier les causes conduisant à une décision. On constate ainsi un besoin grandissant de justifier et d'optimiser les décisions dans de courts délais qui induit l'intégration à ses systèmes d'une composante d'explication évoluée. Le principal enjeu de ces recherches est de fournir une approche industrialisable de l'explication des processus de décision d'un BRMS et plus largement d'un système à base de règles. Cette approche devra être en mesure d'apporter les informations nécessaires à la compréhension générale de la décision, de faire office de justification auprès d'entités internes et externes ainsi que de permettre l'amélioration des moteurs de règles existants. La réflexion se portera tant sur la génération des explications en elles-mêmes que sur la manière et la forme sous lesquelles elles seront délivrées. / The concept of “Business Rule Management System” (BRMS) has been introduced in order to facilitate the design, the management and the execution of company-specific business policies. Based on a symbolic approach, the main idea behind these tools is to enable the business users to manage the business rule changes in the system without requiring programming skills. It is therefore a question of providing them with tools that enable to formulate their business policies in a near natural language form and automate their processing. Nowadays, with the expansion of intelligent systems, we have to cope with more and more complex decision logic and large volumes of data. It is not straightforward to identify the causes leading to a decision. There is a growing need to justify and optimize automated decisions in a short time frame, which motivates the integration of advanced explanatory component into its systems. Thus, the main challenge of this research is to provide an industrializable approach for explaining the decision-making processes of business rules applications and more broadly rule-based systems. This approach should be able to provide the necessary information for enabling a general understanding of the decision, to serve as a justification for internal and external entities as well as to enable the improvement of existing rule engines. To this end, the focus will be on the generation of the explanations in themselves as well as on the manner and the form in which they will be delivered.

No description available.