1.

Record Nr.

UNINA9910132647003321

Autore

Boullier Dominique

Titolo

Opinion mining et sentiment analysis méthodes et outils : méthodes et outils / / Dominique Boullier et Audrey Lohard

Pubbl/distr/stampa

OpenEdition Press, 2012

France : , : OpenEdition Press, , 2012

ISBN

9782821812277

9782821812260

Descrizione fisica

1 online resource : digital, PDF file(s)

Collana

Sciences Po, médialab

Soggetti

Engineering & Applied Sciences

Computer Science

Lingua di pubblicazione

Francese

Formato

Materiale a stampa

Livello bibliografico

Monografia

Note generali

Bibliographic Level Mode of Issuance: Monograph

Nota di bibliografia

Includes bibliographical references.

Sommario/riassunto

Opinion mining is on the way to becoming a real industry, just as strategic as that of polls. The promises made are impressive: the computing power of computer tools would make it possible to follow all the evolutions of opinion on the web in real time, whatever the volume. Moreover, the linguistic processing capacities would make it possible to detect the tones of all the verbatims collected, thanks to the so-called “sentiment analysis” methods. The state of the art of commercial and technological offers presented in this book takes account of this effervescence but also underlines its excessiveness, by taking care to distinguish the real results from the sometimes misleading promotional slogans. he book written by researchers from the medialab of Sciences Po, a laboratory specializing in the processing of the masses of data available on the web for the social sciences, also makes it possible to situate the interest of these new technical means for research, in the context of of what are now called “digital humanities”. Finally, anxious to allow each reader to take in hand these tools, certainly powerful but with very real limits of validity, the authors describe step by step all the phases of a project mobilizing the methods of opinion mining, specifying the pitfalls. and the imperatives



of intervention of human expertise, always necessary. Largely illustrated, this book should encourage researchers, public opinion and marketing professionals as well as computer scientists and specialists in “web studies” to exchange views in order to advance these common tools.

2.

Record Nr.

UNINA9910787190103321

Autore

Egg Matthias

Titolo

Scientific realism in particle physics : a causal approach / / Matthias Egg

Pubbl/distr/stampa

Boston, [Massachusetts] ; ; Berlin, [Germany] : , : Walter de Gruyter Inc., , 2014

©2014

ISBN

3-11-038351-9

3-11-035440-3

Descrizione fisica

1 online resource (200 p.)

Collana

Epistemische Studien : Schriften zur Erkenntnisund Wissenschaftstheorie, , 2198-1884 ; ; Band 29

Classificazione

CC 3500

Disciplina

530.1202855133

Soggetti

Particles (Nuclear physics) - Data processing

Physics - Philosophy

Realism

Lingua di pubblicazione

Inglese

Formato

Materiale a stampa

Livello bibliografico

Monografia

Note generali

Description based upon print version of record.

Nota di bibliografia

Includes bibliographical references and index.

Nota di contenuto

Part I. The recent debate on scientific realism -- Part II. Causal realism -- Part III. The quantum challenge.

Sommario/riassunto

Particle physics studies highly complex processes which cannot be directly observed. Scientific realism claims that we are nevertheless warranted in believing that these processes really occur and that the objects involved in them really exist. This book defends a version of scientific realism, called causal realism, in the context of particle physics. The first part of the book introduces the central theses and arguments in the recent philosophical debate on scientific realism and discusses entity realism, which is the most important precursor of causal realism. It also argues against the view that the very debate on



scientific realism is not worth pursuing at all. In the second part, causal realism is developed and the key distinction between two kinds of warrant for scientific claims is clarified. This distinction proves its usefulness in a case study analyzing the discovery of the neutrino. It is also shown to be effective against an influential kind of pessimism, according to which even our best present theories are likely to be replaced some day by radically distinct alternatives. The final part discusses some specific challenges posed to realism by quantum physics, such as non-locality, delayed choice and the absence of particles in relativistic quantum theories.