Proceedings of the International Workshop on Ubiquitous and Decentralized User Modeling:in conjunction with 11th International Conference on User Modeling, UM 2007. Corfu, Greece, June 26th, 2007
Notice bibliographique
Résumé
Personalising or intelligently adapting an application to user's behaviour and context requires gathering and processing some information about the user.Data collected for user modelling may contain personal and sometimes confidential information.Hence, privacy issues raise with the use of personalisation and context-awareness technology.Is personal data collected and used appropriately, with full awareness and consent of the user?Is the user granted the adequate access rights to manage their personal data?Local privacy regulations have a strong impact on business operations and vary a lot from country to country.Another related issue is that of the security of collected user data; is it kept secured, uncorrupted, with appropriate tracking mechanisms and protection from unauthorized access?The ownership of that data can also be questioned: user profiles may describe how the user relates to their environment (social, digital or physical).As such, it contains both information about the user and their environment.Does such information fully belong to the user?Or to tuples of users involved in a given transaction or communication?Can it be owned by the content provider?By the network or service operator?Or by the device manufacturer?Last but not least, beyond legal and security matters, personalisation systems are only accepted when they are of good-enough quality and accuracy.In order to be accepted by users, their efficiency must exceed some quality threshold for users to accept them.By adopting personalisation systems, they not only put some of their privacy at risk but also their time and attention as they may receive irrelevant recommendations from poorly performing personalisation systems, as well as their money since the cost of the system is at the end somehow supported by end-users.When is a personalisation system good enough for being compliant with the users' expectations in terms of quality of service as well as in terms of cost?The economic viability of content-related business models may vastly rely on the accuracy allowed by the current state of the art in personalisation systems.For instance, how far can mobile advertisements targeting go in proposing business tradeoffs between the need of the advertisers to achieve high efficiency in advertising campaigns, the need for privacy of users and their hostility toward being distracted by irrelevant ads on their personal mobile device?Is there an appropriate price for letting users abandon their privacy and accept reading poorly targeted ads?Can advanced user modelling systems do anything better than that?All these issues will be presented and discussed through our talk planned during the UbiDeUM'2007 workshop (International Workshop on Ubiquitous and Decentralized User Modelling).
Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.
Comment cette classification a été obtenuedéplier
Prédiction machine sur la base complète
Imitation des enseignantsNi prévalence calibrée, ni vérité terrain. Validation humaine à venir. Le volet Gemma est une étiquette directe du modèle pour chaque travail de la base, lue sur la notice réduite au titre. Le volet Codex est un classifieur appris des 10 348 étiquettes directes de Codex et calibré sur les taux pondérés de l'échantillon; les champs sans appui suffisant ne portent aucun appel Codex. Le mode candidate est l'union des deux volets; le consensus est leur intersection. Ces sorties portent le statut machine_predicted_unvalidated et ne sont pas des étiquettes humaines.
Scores du classifieur distillé par catégorie (deux têtes)
| Catégorie | Codex | Gemma |
|---|---|---|
| Métarecherche | 0,006 | 0,008 |
| Méta-épidémiologie (sens strict) | 0,001 | 0,001 |
| Méta-épidémiologie (sens large) | 0,002 | 0,001 |
| Bibliométrie | 0,001 | 0,001 |
| Études des sciences et des technologies | 0,002 | 0,001 |
| Communication savante | 0,005 | 0,006 |
| Science ouverte | 0,002 | 0,004 |
| Intégrité de la recherche | 0,002 | 0,003 |
| Charge utile insuffisante (le modèle a refusé de juger) | 0,019 | 0,005 |
Scores machine (provisoires)
Les deux têtes enseignantes du modèle étudiant, lues sur ce travail. Un score ordonne la base pour la relecture; il n'affirme jamais une catégorie, et le statut de validation accompagne chaque rangée tel quel.
Scores de référence d'un modèle non mature (critères de maturité non atteints, 7 itérations). Un score ordonne; il n'affirme jamais une catégorie.
score_only:v0-immature-baseline · tel quel depuis la passe de notation : score_only signifie que le nombre peut ordonner les travaux, et qu'aucune étiquette de catégorie n'en découleClassification
machine, non validéePrédiction automatique; un appel candidat d’une seule source (Gemma direct ou Codex distillé), pas un consensus.
Le détail, modèle par modèle et score par score, se trouve en fin de page sous « Comment cette classification a été obtenue ».