Suchergebnisse

Quantification is defined as the pairing of elements of two sets. This definition and the characteristics of the sets in question is expanded, and the varieties and methods for quantifying in man-machine systems are described. Special attention is given to a brief overview of psychophysics and to probability theory as they relate to the problems of quantification in this area. Finally, guidelines are presented to aid in the search for adequate quantification.

In recent years there has been a growing interest in text quantification, a supervised learning task where the goal is to accurately estimate, in an unlabelled set of items, the prevalence (or "relative frequency") of each class c in a predefined set C. Text quantification has several applications, and is a dominant concern in fields such as market research, the social sciences, political science, and epidemiology. In this paper we tackle, for the first time, the problem of ordinal text quantification, defined as the task of performing text quantification when a total order is defined on the set of classes; estimating the prevalence of "five stars" reviews in a set of reviews of a given product, and monitoring this prevalence across time, is an example application. We present OQT, a novel tree-based OQ algorithm, and discuss experimental results obtained on a dataset of tweets classified according to sentiment strength.

The ethical dimensions of quantification are seldom analysed. We examine three ethical features that are characteristic of quantification – its capacity to express or mediate power, focus attention, and shape opportunity structures. We do so in the context of three recent examples of new types of quantification : university rankings, the racial classification of Asians in the US, and facial recognition algorithms. Our examples highlight the importance of understanding the varied and complex ways that quantification creates and organizes social relations, and the effect of this on multiple forms of inequality.

L'objectif central de ce travail est d'analyser la mesure du stéréotype. Il s'agit de montrer que les mesures du stéréotype obéissent à des évolutions parallèles aux diverses définitions du concept. Dans la littérature, le concept de stéréotype est utilisé par les auteurs selon au moins quatre acceptions. D'abord, la notion était liée à celle d'attitude, puis elle a référé aux relations entre les groupes. La notion s'est ensuite rapportée au prototype et enfin aux TIP (théories implicites de la personnalité). La quantification du stéréotype a suivi l'évolution de ces travaux, allant de l'analyse socio-historique réalisée à l'aide de la «check-list », de tests, d'échelles et inventaires aux différentes mesures probabilistes s'inspirant de la règle de Bayes.

The need for a target vulnerability quantification process arises from the U.S. Navys requirement to develop defensive systems to defend civilian and military assets around the world. The systems engineering approach has been applied to improve an existing process that takes too long to perform. The functions to perform the vulnerability assessment are identified, and an automated process incorporating current off-the-shelf technologies is discussed. The operational requirements, maintenance concept, and functional analysis are presented for the automated vulnerability quantification process. A conceptual design is then outlined for the system elements. A cost comparison between the current system and the automated system is calculated in the conceptual design. ; Master of Science