Short Papers - Abstracts
Do Students Trust Their Open Learner Models? (page 255)Norasnita Ahmad and Susan Bull
Open learner models (OLM) enable users to access their learner model to view information about their understanding. Opening the learner model to the learner may increase their perceptions of how a system evaluates their knowledge and updates the model. This raises questions of trust relating to whether the learner believes the evaluations are correct, and whether they trust the system as a whole. We investigate learner trust in various OLM features: the complexity of the model presentation; the level of learner control over the model contents; and the facility to release one's own model for peer viewing.
A Framework for the Development of Distributed, Context-Aware Adaptive Hypermedia Applications (page 259)Liliana Ardissono, Anna Goy, and Giovanna Petrone
The CAWE framework supports the development of contextaware, Service Oriented applications which integrate heterogeneous services and customize the cooperation among multiple users. We present the techniques adopted in the framework to manage a context-sensitive interaction with the users.
Collection Browsing through Automatic Hierarchical Tagging (page 263)Korinna Bade and Marcel Hermkes
In order to navigate huge document collections efficiently, tagged hierarchical structures can be used. For users, it is important to correctly interpret tag combinations. In this paper, we propose the usage of tag groups for addressing this issue and an algorithm that is able to extract these automatically for text documents. The approach is based on the diversity of content in a document collection. For evaluation, we use methods from ontology evaluation and showed the validity of our approach on a benchmark dataset.
Aspect-Based Personalized Text Summarization (page 267)Shlomo Berkovsky, Timothy Baldwin, and Ingrid Zukerman
This work investigates user attitudes towards personalized summaries generated from a coarse-grained user model based on document aspects. We explore user preferences for summaries at differing degrees of fit with their stated interests, the impact of length on user ratings, and the faithfulness of personalized and general summaries.
What Can I Watch on TV Tonight? (page 271)David Bueno, Ricardo Conejo, David Martin, Jorge Leon, and Javier G. Recuenco
This paper presents the methods used in a TV Recommender System that helps users in the difficult task of finding an interesting TV program from among the hundreds of channels that we can find nowadays on TV. Our aim is to cover not only user preferences but also user restrictions while watching TV. The recommendations use a hybrid method, combining content based and folksonomy (collaborative and social recommendations). We also present interesting initial results of some experiments that try to show the accuracy of the users recommendations.
Adaptive Navigation Support, Learner Control and Open Learner Models (page 275)Susan Bull, Norasnita Ahmad, Matthew Johnson, Rasyidi Johan, Andrew Mabbot, and Alice Kerly
We consider open learner models (OLM) with reference to adaptive navigation support and learner control. Our purpose is to assess the potential of a greater range of OLMs in adaptive educational hypermedia. We introduce five OLMs, discuss how these might be applied, and present learner reactions.
News@hand: A Semantic Web Approach to Recommending News (page 279)Ivan Cantador, Alejandro Bellogin, and Pablo Castells
We present News@hand, a news recommender system which applies se-mantic-based technologies to describe and relate news contents and user preferences in order to produce enhanced recommendations. The exploitation of conceptual in-formation describing contents and user profiles, along with the capability of infer-ring knowledge from the semantic relations defined in the ontologies, enabling different content-based collaborative recommendation models, are the key distinctive aspects of the system. The multi-domain portability, the multimedia source applicability, and addressing of some limitations of current recommender systems are the main benefits of our proposed approach.
A SOA-Based Framework to Support User Model Interoperability (page 284)Federica Cena and Roberto Furnari
This paper presents an approach to achieve User Model (UM) interoperability exploiting Web Service technologies for syntactic interoperability, and Semantic Web languages for semantic interoperability, together with negotiation techniques based on dialogue. We propose a SOA-based framework where a central UDDI registry, enhanced with UM specific capabilities, is used to support and promote the cooperation between UM-based applications.
Integrated Speaker Classification for Mobile Shopping Applications (page 288)Michael Feld and Gerrit Kahl
This paper presents an approach to how speaker classification can be used to enable new ways for recommender systems in a mobile shopping environment to bootstrap user models and avoid common problems such as the 'early rater'. In a concrete shopping scenario, we introduce the speech-controlled Mobile ShopAssist demonstrator that allows a new customer to more quickly find a product that fulfills his or her demographic group's specific requirements by exploiting features extracted from speech using the Agender speaker classification system.We propose a method for computing preference scores based on the user's profile and demonstrate how the application's GUI can be adapted to deliver the recommendations to the user.
The Authoring Tool of ADULT: Adaptive Understanding and Learning Text Environment (page 292)Alexandra Gasparinatou, Grammatiki Tsaganou, and Maria Grigoriadou
Previous research in the domain of text comprehension in Informatics has demonstrated that readers with little knowledge in this domain benefit from a coherent text, whereas high-knowledge readers benefit from a minimally coherent text. With respect to educational applications, these findings suggest constructing several versions of a text in order to adapt to varying levels of knowledge among readers. In this paper we present the design of the authoring tool of the learning environment ADULT (Adaptive Understanding and Learning from Texts), capable of supporting authors while constructing texts of different coherence in the domain of Informatics, accompanied by questions or tasks designed to access students' comprehension on line. This way students will be activated to use their background knowledge while reading and more students will have the opportunity to achieve better learning results in learning from Informatics texts than reading a single textbook in Informatics targeted at an average reader.
Interoperability between MOT and Learning Management Systems: Converting CAF to IMS QTI and IMS CP (page 296)Fawaz Ghali and Alexandra I. Cristea
The chain of applying adaptivity to Learning Management Systems (LMS) is still deficient; there is a gap between authoring adaptive materials and delivering them in LMS. In this paper, we extend My Online Teacher (MOT), an adaptive authoring system, by adding compatibility with IMS Question & Test Interoperability (QTI) and IMS Content Packaging (CP). Thus, the authors can utilize the authored materials for learning process adaptation on any standards-compatible LMS. From a technical perspective, we initialize the creation of adaptive LMS by converting Common Adaptation Format (CAF), XML representation of MOT database, into IMS QTI and IMS CP, to ensure a wider uptake and use of adaptive learning systems. Finally, this work represents a significant step towards the little explored avenue of adaptive collaborative systems based on extant learning standards and popular LMS.
Proactively Adapting Interfaces to Individual Users for Mobile Devices (page 300)Melanie Hartmann and Daniel Schreiber
The amount of functionality offered by nowadays applications is constantly growing, mostly leading to more and more complex user interfaces. This often decreases their usability, especially in mobile settings where we have to deal with limited input and output capabilities. We state that adapting the interface to the available devices as well as to the user's current needs is the key to improving usability. In this paper, we present the AUGUR system that can automatically generate user- and device-adapted interfaces. We thereby focus on the FxL* algorithm that determines which user interface elements are currently relevant for a user.We show that it clearly outperforms algorithms that do not take the user or her situation into account.
Reuse Patterns in Adaptation Languages: Creating a Meta-level for the LAG Adaptation Language (page 304)Maurice Hendrix and Alexandra Cristea
A growing body of research targets authoring of content and adaptation strategies for adaptive systems. The driving force behind it is semantics-based reuse: the same strategy can be used for various domains, and vice versa. Whilst using an adaptation language (LAG e.g.) to express reusable adaptation strategies, we noticed, however, that: a) the created strategies have common patterns that, themselves, could be reused; b) templates based on these patterns could reduce the designers' work; c) there is a strong preference towards XML-based processing and interfacing. This has leaded us to define a new meta-language for LAG, extracting common design patterns. This paper provides more insight into some of the limitations of Adaptation Languages like LAG, as well as describes our meta-language, and shows how introducing the meta-level can overcome some redundancy issues.
Implementing a Multimodal Interface to a DITA User Assistance Repository (page 308)Aidan Kehoe and Ian Pitt
User assistance systems can be extended to enable multimodal access to user assistance material. Implementing multimodal user assistance introduces new considerations with respect to authoring and storage of assistance material, transformation of assistance material for effective presentation on a range of devices, and user interaction issues. We describe an implementation of a multimodal interface to enable access to a DITA user assistance repository.
Analysing Hig h-Level Help-Seeking Behaviour in ITSs (page 312)Moffat Mathews, Tanja Mitrovic, and David Thomson
In this paper, we look at initial results of data mining students' help-seeking behaviour in two ITSs: SQL-Tutor and EER-Tutor. We categorised help given by these tutors into high-level (HLH) and lowlevel help (LLH), depending on the amount of help given. Each student was grouped into one of ten groups based on the frequency with which they used HLH. Learning curves were then plotted for each group. We asked the question, Does a student's help-seeking behaviour (especially the frequency with which they use HLH) affect learning? We noticed similarities between results for both tutors. Students who were very frequent users of HLH showed the lowest learning, both in learning rates and depth of knowledge. Students who were low to medium users of HLH showed the highest learning rates. Least frequent users of HLH had lower learning rates but showed higher depth of knowledge and a lower initial error rate, suggesting higher initial expertise. These initial results could suggest favouring pedagogical strategies that provide low to medium HLH to certain students.
Data-Driven Prediction of the Necessity of Help Requests in ILEs (page 316)Manolis Mavrikis
This paper discusses the data-driven development of a model which predicts whether a student could answer a question correctly without requesting help. This model contributes to a broader piece of research, the primary goal of which was to predict affective characteristics of students working in ILEs. The paper presents the bayesian network which provides adequate predictions, and discusses how its accuracy is taken into account when the model is integrated in an ILE. Future steps to improve the results are briefly discussed.
A Dynamic Content Generator for Adaptation in Hypermedia Systems (page 320)David Merida, Ramon Fabregat, Xavier Prat, David Huerva, and Jeimy Velez
The heterogeneity problem (in terms of different types of access devices, network bandwidth, preferences/characteristics of the user, etc.) has become a major problem for the Internet. Different alternatives have been developed to allow universal access to any type of content. Adaptive Hypermedia Systems (AHS) have emerged as a solution for this. In previous works we proposed the SHAAD model, which includes the concepts of adaptability, adaptivity and dynamism to adapt web contents. Based on this model we implemented MAS-SHAAD, a multiagent system implementation of SHAAD. In this work we present the design and development of a dynamic content generator that can be added to any JAVA AHS implementation, such as MAS-SHAAD. The structure of the generator is defined by an ontology; therefore, a standard behavior can be obtained for any object included in the web pages generated and stored in the content repository.
Automatic Generation of User Adapted Learning Designs: An AI-Planning Proposal (page 324)Lluvia Morales, Luis Castillo, Juan Fernandez-Olivares, and Arturo Gonzalez-Ferrer
A Learning Design(LD) definition under the IMS-LD standard is a complex task for the instructor because it requires a lot of time, effort and previous knowledge of the students group over which will be defined the knowledge objectives. That is why, taking advantage from diffusion of learning objects(LO) labeling using IMS-MD standard, we have proposed to realize a knowledge engineering process, represented as an algorithm, over LO labels and user profiles to automaticaly define a domain that will be used by an intelligent planner to build a LD. This LD will be finally implemented in the ILIAS Learning Management System (LMS).
Guaranteeing the Correctness of an Adaptive Tutoring System (page 329)Pilar Prieto-Linillos, Sergio Gutierrez, Abelardo Pardo, and Carlos Delgado Kloos
This paper presents an approach to create adaptive web-based educative systems that can be automatically audited by means of standard web testing tools. The auditing tool takes the role of a learner interacting with the system, checking that no errors are present. The tool can communicate with the exercises to know the correct answers to them; a configurable ratio of correct to incorrect answers allows the tool to behave as a range of different students. More complex checking techniques will be tested in the future using this architecture.
Designing a Personalized Semantic Web Browser (page 333)Melike Sah, Wendy Hall, and David C. De Roure
Web browsing is a complex activity and in general, users are not guided during browsing. Our hypothesis is that by using Semantic Web technologies and personalization methods, browsing can be supported better. However, existing personalization mechanisms on the Web are obstructive; users need to log in to multiple websites and enter their personal information and preferences, and the profiles are different for each site. There is a need for generic user profiles, which can also support the user's browsing. In this paper, we propose a novel Semantic Web browser using an ontology-driven user modeling architecture to enable semantic and adaptive links. We also introduce a new behavior-based user model. With our approach, users need to log in to their Web browser only and personalization is achieved on different websites.
Towards Inferring Sequential-Global Dimension of Learning Styles from Mouse Movement Patterns (page 337)Danilo Spada, Manuel Sanchez-Montanes, Pedro Paredes, and Rosa M. Carro
One of the main concerns of user modelling for adaptive hypermedia deals with automatic user profile acquisition. In this paper we present a new ap-proach to predict sequential/global dimension of Felder-Silverman's learning style model that only makes use of mouse movement patterns. The results obtained in a case study with 18 students are very promising. We found a strong correlation between maximum vertical speed and sequential/global dimension score. Moreover, it was possible to predict whether students' learning styles are global or sequential with high accuracy (94.4%). This suggests that mouse movement patterns can be a powerful source of information about certain user features.
VUMA: A Visual User Modelling Approach for the Personalisation of Adaptive Systems (page 341)Melanie B. Sp√§th and Owen Conlan
Current approaches to explicit user modelling are generally time consuming and tedious for the user. Oftentimes poor usability and overly long questionnaires deter the end user from reusing such modelling tools, thus only facilitating explicit personalisation once as they enter the system. This paper proposes a visual approach to user modelling resulting in the VUMA (Visual User Modelling Approach) tool that can be used in a playful and dynamic manner repeatedly during a user's engagement with a personalisation system. This work proposes and evaluates a visually empowering, usable, highly configurable and playful user modelling interface that is utilised to elicit user interests and preferences in a chosen knowledge domain.
Bookmark Category Web Page Classification Using Four Indexing and Clustering Approaches (page 345)Chris Staff
Web browser bookmark files store records of web pages that the user would like to revisit. We use four methods to index and automatically classify documents referred to in 80 bookmark files, based on document title-only and full-text indexing and two clustering approaches. We evaluate the approaches by selecting a bookmark entry to classify from a bookmark file, re-creating a snapshot of the bookmark file to contain only entries created before the selected bookmark entry. The baseline algorithm is 39% accurate at rank 1 when the target category contains 7 entries. By fusing the recommendations of the 4 approaches, we reach 78.7% accuracy on average, recommending at most 3 categories.
Personalization Using Ontologies and Rules (page 349)Thanh Tran, Haofen Wang, Steffen Lamparter, and Philipp Cimiano
Adaptive hypermedia systems can alleviate information overload on the Web by personalising the delivery of resources to the user. These systems are however afflicted with difficulties in the acquisition of user data as well as the general lack of user control on and transparency of the systems' adaptive behavior. In this paper, we argue that the use of rules on top of ontologies can enable adaptive functionality that is both transparent and controllable for users. To this end, we sketch ODAS, a domain ontology for adaptive hypermedia systems, and a model for the specification of adaptation rules.
RSS-Based Interoperability for User Adaptive Systems (page 353)Yiwen Wang, Federica Cena, Francesca Carmagnola, Omar Cortassa, Cristina Gena, Natalia Stash, and Lora Aroyo
This paper presents an approach to exploit widely used tag annotations to address two important issues in user-adaptive systems: the cold-start problem and the integration of distributed user models. The paper provides an example of re-use of user interaction data (tags) generated by one application into another one in similar domains for providing cross-system recommendations.
Assisting in Reuse of Adaptive Hypermedia Creator's Models (page 357)Nadjet Zemirline, Yolaine Bourda, Chantal Reynaud, and Fabrice Popineau
The design of Adaptive Hypermedia is a difficult task which can be made easier if generic systems and AH creators' models are reused. We address this design problem in the setting of the GLAM platform only made up of generic components. We present a rule-based approach helping an AH creator in reusing its user and domain models to create a specific adaptive hypermedia. This semi-automatic approach takes the creator's models as specialisations of GLAM generic models and requires the creator to express a minimum set of mappings between his models and the generic ones. The process results in a merged model consisting of the generic and the corresponding specific model. This merged model can be used by the adaptation model.