Stratégies d'aide en ligne contextuelles : acquisition d'expertises, modélisation et évaluation expérimentale

Abstract : Current online help systems usually are unsuitable to the user's actual needs, and fail to help efficiently. Recent research focused on solving issues concerning the design of dynamic help systems. Although, few research has focused on the appraisal of the use of such help systems and their influence on the end user's information assimilation. Therefore, we performed an experimental study, based on a bottom-up approach, in order to study contextual dynamic help systems. First of all, we analysed a corpus of help dialogs involving human experts and novice users, in order to elicit the help strategies used by the experts. The implementation of the identified strategies in an experimental online help system allowed us to evaluate them during an experimentation based on the Wizard of Oz protocol. That study led us to propose a set of recommendations for the realization of contextual online help systems.
Document type :
Theses
Complete list of metadatas

https://hal.univ-lorraine.fr/tel-01748641
Contributor : Thèses Ul <>
Submitted on : Thursday, March 29, 2018 - 11:43:51 AM
Last modification on : Tuesday, April 24, 2018 - 1:31:00 PM
Long-term archiving on : Friday, September 14, 2018 - 2:39:25 AM

File

SCD_T_2002_0286_CAPOBIANCO.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : tel-01748641, version 1

Collections

Citation

Antonio Capobianco. Stratégies d'aide en ligne contextuelles : acquisition d'expertises, modélisation et évaluation expérimentale. Autre [cs.OH]. Université Henri Poincaré - Nancy 1, 2002. Français. ⟨NNT : 2002NAN10282⟩. ⟨tel-01748641⟩

Share

Metrics

Record views

23

Files downloads

33