Figure - uploaded by Olga Muñoz Ramos
Content may be subject to copyright.
-Mean Text Length and Mean Quality Ratings for each of the search conditions and subtasks. Standard deviations appear in parenthesis.
Source publication
With the use of the computers, the task of writing is intertwined with the task of searching for information that can be relevant for the document that is being written, however very little research has been done to understand how the two tasks intertwine. In this paper we present an initial attempt to develop a model of writing and information see...
Contexts in source publication
Context 1
... = 0.07. As can be seen in the first row of Table 2, during planning the text length tended to be longer in the Proactive condition. These differences were not significant when considering translation and reviewing with ...Similar publications
Motivation -- Finding appropriate information while writing a scientific paper is essential, but also difficult and time consuming. A Proactive Recommender System (PRS) retrieves information relevant to the text being written, and presents it automatically. However, current PRSs overlook that writing is a demanding task, affected by interruptions....
Citations
... The recommendation of documents can improve task performance by the reduction of the number of computer interactions required, and has been showed to improve the perceived usability of an information re-finding system (Wakeling et al., 2014). Especially for the task of writing, the time to complete the task can be shortened and the quality of the written document can be improved when relevant information is pro-actively recommended (Melguizo et al., 2010). This suggests that a recommender system for re-finding information can be useful for a knowledge worker. ...
... Thus, the user does not need to select his context, which is an advantage over the context-aware recommendation agents described in the previous section. The context query that is used can be formulated from the document a person is writing (Budzik and Hammond, 2000;Melguizo et al., 2010), the blogpost he is writing (Gao and Bridge, 2010), e-mail messages that are being read or composed (Dumais et al., 2004), the news that is being broadcasted (Henzinger et al., 2005) or the text that is visible on screen together with the location, person, date and subject information (Rhodes, 1997). ...
In this article we evaluate context-aware recommendation systems for information re-finding by knowledge workers. We identify 4 criteria that are relevant for evaluating the quality of knowledge worker support: context relevance, document relevance, prediction of user action, and diversity of the suggestions. We compare 3 different context-aware recommendation methods for information re-finding in a writing support task. The first method uses contextual prefiltering and content-based recommendation (CBR), the second uses the just-in-time information retrieval paradigm (JITIR), and the third is a novel network-based recommendation system where context is part of the recommendation model (CIA). We found that each method has its own strengths: CBR is strong at context relevance, JITIR captures document relevance well, and CIA achieves the best result at predicting user action. Weaknesses include that CBR depends on a manual source to determine the context and in JITIR the context query can fail when the textual content is not sufficient. We conclude that to truly support a knowledge worker, all 4 evaluation criteria are important. In light of that conclusion, we argue that the network-based approach the CIA offers has the highest robustness and flexibility for context-aware information recommendation.
... Interruption management has been frequently addressed in human-computer interaction research. There have been efforts showing the effects of interruptions on web-based systems and the necessity of managing them (Puerta et al.,2008;Bailey et al., 2008). ...
This work explores the benefits of supporting learners affectively in a context-aware learning situation. This features a new challenge in related literature in terms of providing affective educational recommendations that take advantage of ambient intelligence and are delivered through actuators available in the environment, thus going beyond previous approaches which provided computer-based recommendation that present some text or tell aloud the learner what to do. To address this open issue, we have applied TORMES elicitation methodology, which has been used to investigate the potential of ambient intelligence for making more interactive recommendations in an emotionally challenging scenario (i.e. preparing for the oral examination of a second language learning course). Arduino open source electronics prototyping platform is used both to sense changes in the learners’ affective state and to deliver the recommendation in a more interactive way through different complementary sensory communication channels (sight, hearing, touch) to cope with a universal design. An Ambient Intelligence Context-aware Affective Recommender Platform (AICARP) has been built to support the whole experience, which represents a progress in the state of the art. In particular, we have come up with what is most likely the first interactive context-aware affective educational recommendation. The value of this contribution lies in discussing methodological and practical issues involved.
