Dynamic generation of hypertext and its adaptation and personalization to particular users is a powerful and useful concept. It is particularly helpful for the reduction of the information overload such as is frequently experienced on the Internet. But it is equally helpful for guiding users towards “interesting” topics, products, artifacts or descriptions thereof in electronic shops, libraries or museums, or for filtering appropriate items from a general or domain-specific news feed.
Reference models and generic architectures unify a community and provide a leading generic model and/or architecture that spawns research activities in many directions. Examples of such generic models are AHAM for adaptive hypermedia and FOHM for open hypermedia. A nice example of a resulting generic implementation is the AHA! system that was last described in ACM Hypertext'06.
The research fields of hypertext and adaptive hypermedia (or adaptive web-based information systems) however, have been growing rapidly during the past ten years and this has resulted in a plethora of new terms, concepts, models and prototype systems. As a result the established models no longer include many of the recent new concepts and phenomena. In particular, open corpus adaptation, ontologies, group adaptation, social network analysis and data mining tools for adaptation are not or at least insufficiently supported.
The DAH'09 workshop1 organized in conjunction with the 20th ACM International Conference on Hypertext and Hypermedia and held on June 29, 2009, in Turin, Italy. The workshop provides a focused international forum for researchers to discuss new developments in generic methods and technologies for dynamic and adaptive hypertext. Topics discussed during the workshop include: adaptation and personalization including such issues as open-corpus adaptation, group adaptation, higher order adaptation, and sharing of user models; adaptive and dynamic hypertext authoring (e.g. authoring conceptual adaptation models); service-oriented approaches for adaptation; use of data mining for user and domain modeling, and automatic generation of adaptation rules.
These proceedings include six accepted contributions to the workshop. We would like to thank the authors for their interest in the workshop and for submitting their contributions. Special thanks to the PC Members for their help in reviewing the submitted papers.
The first paper by Levacher et al. “A Framework for Content Preparation to Support OpenCorpus Adaptive Hypermedia” proposes a novel framework for open-corpus content preparation that allows processing documents existing in open and closed corpora and producing coherent conceptual sections of text with associated descriptive metadata. This work is important for enabling the mainstream adoption of AHS in web applications, which is impossible without enabling the repurposing existing content. The authors adopt state-of-the-art information extraction and structural content analysis techniques for an on-demand provision of tailored, atomic information objects.
Next contribution by van der Slijs et al. “GAL: A Generic Adaptation Language for describing Adaptive Hypermedia” argues that despite a large variety of personalization and adaptation features that different emerging Web applications offer, it is possible to distinguish a central and unifying objective of adaptive engines that facilitates these various features, that is, to create an adaptive navigation structure. The authors present Generic Adaptation Language (GAL) that specifies the engine independent basic adaptive navigation structure that allows using any authoring environment in combination with any adaptive engine as long as there is a corresponding GAL compiler.
The following two papers present the service-based view on the development of adaptive hypermedia. Harrigan and Wade in “Towards a Conceptual and Service-Based Adaptation Model” consider the limitations of the current practice in adaptation modeling which assume that concepts and relationships between concepts are the fundamental building blocks of any adaptive content and introduce a representation of a conceptual and service-based adaptation model. With their approach 1 The scientific programme overview, and other workshop-related information as well as the link to the online proceedings can be found at http://www.win.tue.nl/~mpechen/conf/dah09/.
I having additional expressive power it would be possible to facilitate activity-based and processoriented adaptive applications, including the adaptation of the process itself.
Koidl et al. in “Non-Invasive Adaptation Service for Web-based Content Management Systems” consider the architectural and technical issues related to provision of third party adaptive services pluggable into existing web-based content management systems (WCMS) like Wiki. The authors introduce a third party Adaptive Service that also contributes to mainstreaming the adoption of AHS. This work introduces a principled way of embedding adaptive technologies within existing WCMS allowing to reduce or avoid the expense of re-engineering such systems.
The paper by Hargood et al. “Investigating a thematic approach to narrative generation” considers the potential, the challenges and open issues in integrating themes in narrative generation systems that attempt to generate content within a narrative or story framework.
Knutov et al. in “Versioning in Adaptive Hypermedia” present an approach that reuses the key concepts and ideas behind versioning and applies them to the adaptive hypermedia field. The authors illustrate with a couple of intuitive examples how such an approach helps to facilitate authoring, managing, storing, maintenance, logging and analysis of behaviour because of providing more flexibility and maintainability of the systems. Particular, versioning helps to create, maintain and reuse concurrent versions of an application or a model or a particular property and value, saving authoring effort and that is not less important facilitating provenance analysis.
June 2009
Organization Organizing Committee
Programme Committee
III Killian Levacher, Eamonn Hynes, Seamus Lawless, Alex O'Connor and Vincent Wade A Framework for Content Preparation to Support Open Corpus Adaptive Hypermedia .................... 1
"$ % "$ % & $ ’
+ $ # # ((( +’ ’ " "
$ ’ $ # ( $ ’ (
)’’ * ( $ , -.’ $ ’ # $ ( $ " # " % ! " # # !! "$
% ! # # ! # ! "& !!! $ # ! # " " ’( " " " ! )! # *!& " # #%
+ &* # ,-.,/. 0 & " " " $ & " !
"! #! $ , #
2 * 3 ()* $ 4
% 0 ( , , $ % ) * ! "# & ’ ’
"# 7 ’ 5 < = 5 7 5 : 7 ! $ !
" # ’%( & !
)*+ ’+, -./0 % 1 # & " / " " % &" ’ , 1 , *++ 2 - *++ * $ (" ( ,$ (" + &
) "##% & ( . + + ( , & ( + & ( + + ( International Workshop on Dynamic and Adaptive Hypertext: Generic Frameworks, Approaches and Techniques (DAH'09) 9 $% & ’ * + ,-./.*00123 +3 4 " $% & ’
) " !" # $ %&&’( ) *
+ $ , %&&-( . / 0 1 * + ) 1
, 2 111 %&&&( 3 4+ 5/ 2 $6+ ! 7 6 $+ 1+
6 %)88)( 9 4 " 7 1+ #
, $ : 2 5+ ; $ ; <+0 2 %;<2( 2 )) %)88&( %&&:( +,--%/ &,>1&>+,--0/ ,2>1&%2 a potentially powerful solution to the problem of generating engaging and relevant information dynamically. Narrative generation seeks to automatically create bespoke narratives for a variety of uses, either generating requested narratives from scratch or by representing existing information as a narrative.
Existing narrative generation systems while often successfully generating short narrative can find their results bland or unvaried depending on the limitations of their approach. We suggest a thematic approach to narrative generation where the addition of themes will enrich generated narratives making them closer to human authored narratives with a thematic objectivity beyond the base communication of the information present within the narrative.
A thematic model has been developed [
The focus of this paper is how such a thematic system could be integrated with existing narrative generation approaches in order to enrich the resulting narratives. We review existing approaches to narrative discourse generation and explore at what level a thematic system should be involved with narrative generation, what approaches it best compliments, and what benefits are likely to emerge from an integration. 2
Narratology Narratology is the study of narrative within literature and as such is primarily focused on the analysis of narrative. However it does provide a useful insight into how stories are constructed.
Structuralism is an area of narratology concerned with deconstructing
narratives to identify the components from which a story is built and the structures
that they build within a story. Because of the tangible nature of structuralism its
ideas are very useful for narrative generation as its clear definition of structures
and elements give an insight into what narrative generation systems should be
generating. Most structuralists assert that a narrative is composed of an
authored sequence of human experiences [
The story element is the sum of all experiences and elements that make up the narrative. The discourse represents what parts of the story are exposed in the narrative (the story selection) and how it is told (the story presentation).
Discourse is a complicated process concerning many different decisions including how the story is presented, what medium is used, the style, the genre, and the themes of the narrative. Thematics approaches themes with a structuralist method of deconstruction and identifies the narrative elements that communicate themes.
Tomashevsky identified the thematic elements of themes (broad ideas such
as ‘politics’ or ‘drama’) and motifs (more atomic elements directly related to the
narrative such as ‘the helpful beast’ or ‘the thespian’) [
As a process narrative generation can be broken down into three stages; story,
plot, and presentation generation. Depending on the project in question these
stages can be consolidated together or separated, (for example, in the virtual
storyteller, presentation generation is broken down in narration and
presentation [
According to Riedl and Young [
Character centric narrative generation systems often use agent technology to
suitably simulate the characters and their behaviors with a purpose built agent
taking the part of each character such as in work by Cavazza [
Automatic generation of story elements is rare in character centric
narrative generation. This is because elegantly written characters with sophisticated
behavior are key to narratives being successfully emergent from the generated
result and at present the only way to ensure this is to build the characters by
hand. Some story elements are generated by using character archetypes with
cliche behavior such as with the supporting characters in work by Cavazza [
Plot generation in character centric generation is therefore a direct result of
the characters behavior as dictated by the agents playing them or the intelligence
modeling all of the characters. The actions they take to achieve their goals builds
the relationships between story elements and the sequence of events that makes
a plot. Presentation generation is not specifically tied to the character centric
approach but the focus on entities and modeling their actions make character
centric approaches ideal for presentation in game engines (for example AConf
used the UT engine through the mimesis project [
The main weakness of character centric narrative generation is its reliance on an engaging narrative successfully emerging from the exposition of the characters actions. Often these systems generate bland stories that merely report on a series of uninteresting actions. These stories are thus often sensible and varied but lack narrative richness or interesting plot.
Author Centric Author Centric narrative generation seeks to model the authorial process itself rather then the content of the narrative. The systems seek to model the process by creating rule based systems or narrative grammars that use well defined structures that are typical of the desired genre of narrative in order to generate stories.
Author centric narrative generation also lends itself better to the
representation of existing knowledge as narrative as its story elements are not necessarily
the narrative devices such as characters and objects but the devices the author
needs to construct a story. Systems such as ArtEquAKT [
Some systems do model the contents of the narrative to be generated as part
of story generation but remain author centric. Universe [
Plot generation in these systems is a case of applying the rules of the
system for the desired genre, utilizing the grammar with the available resources,
or filling a story template with appropriate resources. Presentation for author
centric systems is often text based, either using templates such as Universe[
Author centric systems tend to be highly specialized for one particular type of narrative, making them inflexible and also often not with a view to generic narrative generation. The stories are seldom varied as they all follow a similar authoring process with the same rules and/or grammars and as such can generate engaging but not often varied narratives.
Compromise Approaches Many narrative generation systems often seek a
compromise between these two approaches in order to counteract the weakness
of using one approach or another. Some systems such as Facade[
The virtual storyteller [
AConf[
The presentation generation for these systems also vary. In the virtual
storyteller [
These systems experienced mixed success with both reporting the generation of successful narratives. However both suffered from similar problems to character centric approaches, while the addition of measures to ensure the narratives structure is engaging does have a positive effect the engaging narrative can at times still fail to emerge from the result and the systems can be reliant of stories that are heavily predefined at the request stage rather then being entirely generated. 3
The Thematic Model
Using a thematic model [
For example, we might view a digital photo as a natom, and the tags on that photo as the features that denote a particular motif. Thus a photo tagged with ‘daffodil’ could denote the motif of ‘flower’, which connotes the theme of ‘Spring’. Themes can themselves build up into new themes, for example the theme of ‘Christmas’ can be used to connote the theme of ‘Winter’. To facilitate an evaluation of the effectiveness of the model a prototype was built that could use an instance of the model to select images from a group of Flickr 1 images based on their ability to connote a desired theme. The prototype went under the working name of the Thematic Model Builder (TMB). 1 http://www.flickr.com Fig. 3. The Thematic Model
As an original instance of the model four themes were modeled (Winter,
Spring, celebration, family) along with all sub themes and motifs of these themes,
XML was used to build this instance. Defining an instance of the model for
particular themes is a complex and subjective process. We explored a systematic
method for building themes based on semiotics [
The prototype was written in java with a simple JSP front end and Flickr
was used as a source of natoms. As a folksonomy its items have rich semantic
annotations in metadata [
The system then followed an algorithm of measuring the thematic quality of
each natom in the fabula. It returns the natoms with the highest scores according
to two metrics:
– Component coverage: the proportion of high-level sub-themes or motifs that
a natom has features for - this is useful for measuring how strongly a natom
matches the desired theme. (for example, winter expands several high-level
sub-theme and motifs including christmas, snow and cold. A natom matching
just one of these has less coverage than one that matches many)
– Thematic coverage: the proportion of desired themes that a natom has
features for - this is useful for searches with multiple themes
The TMB prototype allows us to compare the effectiveness of selecting photos
according to their theme with the process of selecting photos based directly on
their tags. A pilot evaluation of the prototype has been completed in [
Referring back to the division of narrative generation illustrated in figure 2, we can explore the possibility of a thematic systems involvement at different levels of narrative generation. Themes are intangible concepts, a subtext rather then a core focus of the narrative, and for this reason it seems at first that narrative generation would be benefit from thematic involvement at the presentation level. Here themes could be connoted by emphasis given through the presentation to the features within the narrative that denote motifs which in turn connote the desired themes. At this level a thematic subtext would become present through elaboration on the presentation of the plot. However this is a process that could potentially fail if there were no relevant features present in the narrative that could be elaborated upon to help connote the desired theme, the system might find that at the presentation level a thematic system might only be able to offer from a subset of themes.
At the story level of narrative generation a thematic systems involvement would be in some ways the opposite of its involvement at a presentation level. Instead of offering elaboration on existing narrative features at the story level a thematic system would generate additional narrative elements based on a shopping list of required motifs for the desired themes. This way themes would become apparent through the presence of certain story elements that connoted the desired themes. This could potentially fail however if the systems plot generation did not make use of the thematic story elements or they were not properly exposed possibly leading to absence of key motifs. Also, such an approach could damage the generated narrative more then help it potentially flooding the system with elements irrelevant to the plot. For some systems as well story generation integration is not always an option, at least on a fully autonomous level, with many systems generating plot out of pre written and defined story elements. These semi automatic approaches to story generation require a very different approach perhaps with thematic guidance on the creation of these elements as supposed to influencing the automatic generation process in others.
At the plot generation level thematics could play a role in the story selection of narrative elements as well as the way relationships build. The first part of this would be similar to how the TMB prototype builds photo montages in that a list of desired motifs would be compiled and this would be used to thematicly score potential story elements and as such influence their selection and inclusion in the plot. Also, the relationships between story elements and actions of elements could in turn be factored in as features that denote motifs, as such potential actions at the plot generation of the story could be thematicly scored influencing what occurs. For example a story in which violence is a desired theme might see the protagonist kill the antagonist rather then banish or imprison them. However, like inclusion at the story level it is possible that heavy thematic involvement could damage the plot itself, making its involvement a dangerous balancing act, potentially forcing plot actions that damage the quality of the narrative. Furthermore like involvement at the presentation level, a lack of complete control over the story elements could potentially restrict available themes.
On the question of which approach to take character centric approaches are perhaps the most different from the current implementation using the thematic model. Rather then the natoms of narrative segments that our model describes, character centric approaches start by simulating the content of the narrative itself, modeling characters, locations, entities, and events. However, through the process of plot and presentation, character centric approaches still go on to generate natoms that contain features and by extension denote motifs. An integration would have to seek to ensure that certain features were planted in order for the themes to become apparent in the finished narrative. To do this involvement at the story level seems obvious as this is where the elements present within the narrative are generated. However, as a more semi automatic approach with predefined elements is more common then automatic generation at this level in character centric approaches it could be difficult to integrate a thematic approach with the prewritten characters. At the plot and presentation levels an integration seems more possible, potential character actions and story events can be thematicly scored to influence actions taken to be conducive with desired themes and then presented in a way that emphasises the relevant thematic content. Character centric generations frequent use of game engines means that integration at the presentation level may be easier where knowledge of the entities present in a particular scene is much more exact then in natural language. However as already discussed a reliance on integration at these levels potentially limits the available themes.
Author centric approaches are more similar in process to the current implementation of the thematic approach in that they’re heavily based on structures and largely concerned with the authoring process rather then modeling the content of the narrative. The story generation process for some is about composing a pool from large collections of potential natoms, often from the Web, based on their relevance to required parts of the narrative structure. This is very similar to the way the TMB currently puts together selections for montage, and it is easy to see that with author centric projects that work this way thematic integration would be a relatively simple process of scoring potential segments to generate. At the plot level, integration could be a similar process to the integration that would be used with character centric approaches, in that elements selected for exposure could be chosen on their thematic qualities rather then only narrative ones. However, for rule based systems of plot generation thematic rules would need to be written for the system. The feasibility of this would need to be added on a system by system basis. At a presentation level natural language generation poses difficulties for thematic integration as a full lexicon for desired features would need to be developed and integrated with the system. Forcing it to uses a small subset of words might make the language clumsy and its important to remember that the thematic model was created with the theory of structuring a narrative in mind where as the structure of individual pieces of language is very different. However, for those systems that use templating or selected pre authored text presentation, using thematics becomes more feasible where narrative techniques such as emphasis (spatially or visually) can be used to highlight relevant segments to help connote a theme.
The possibilities apparent from this investigation are summarised in the table in figure 4. Decisions and selections made in generation may be influenced thematically by making the objective of the decision thematic as well as for plot objectives. Further thematic integration can be achieved through emphasis at the presentation or plot levels and other presentation choices such as style may have an influence that could be worked in favor of desired themes. 5
In this paper we describe an investigation into the potential challenges of integrating a thematic system with narrative generation. It is possible that a thematic approach may improve the quality of narrative generation by enriching the resulting narratives which, could give adapted hypertexts tailored to the user and of a high quality. However, the question of integration between narrative systems and thematics is a complex one. At what point in the process of generation should there be thematic influence and what style of approach or compromise makes the prospect of a thematic narrative generation system most feasible?
Narrative generation is a rich and varied field with a wide variety of approaches. This variety is maintained in modern work and testament to the fact that no single perspective is a perfect solution to the problem, with this compromise approaches are becoming more popular as systems aim to find a successful middle ground. The complexity of the problem also means there are multiple stages of a system that it being integrated could exert an influence and effect the resulting narrative in different ways, all of which pose different constraints.
It seems that a combination of issues make integrating thematics with character centric narrative generation difficult. As the quality of such systems is reliant on rich and complex characters, they are often authored by hand which makes influencing the stories fundamental elements in an automatic way difficult, at the same time, abandoning thematic involvement at this level instigates constraints on the available themes to be used at later levels. These constraints mean that thematic involvement at the story level (the initial generation of elements) is important to assure a wide variety of available themes. However, in order to assure they are exposed correctly involvement at either the plot or presentation level would be necessary as well. Similarities between parts of plot generation for both approaches and existing implementations of the thematic model make plot generation the easier option. The constraints surrounding thematic involvement at the initial story level however do not necessarily rule it out as a possibility for integration as compromise approaches may allow for thematic elements to be introduced to a story using a director agent along side complex characters.
With an initial exploration of the issues complete, the future of this work is experimentation with integrating thematics and narrative generation. There are still also many remaining questions surrounding this process such as how the thematic scoring would be balanced in an effective way so as to include themes without spoiling a narrative and also how the effectiveness of a resulting system could be evaluated. Versioning in Adaptive Hypermedia
Evgeny Knutov, Paul De Bra, Mykola Pechenizkiy Eindhoven University of Technology, Department of Computer Science
PO Box 513, NL 5600 MB Eindhoven, The Netherlands
debra@win.tue.nl, {e.knutov, m.pechenizkiy}@tue.nl Abstract. This paper presents an approach that uses the terms, operations and methods of versioning applied to the Adaptive Hypermedia (AH) field. We show a number of examples of such a use which helps to facilitate authoring, managing, storing, maintenance, logging and analysis of AHS behaviour, providing extensive flexibility and maintainability of the systems.
Keywords: Adaptive Hypermedia, Versioning Systems, Revision Control, Authoring, User Model Updates, Context Changes. 1 Introduction
Versioning (also known as revision control, source control or code management) is the management of multiple revisions of the same unit of information. It is commonly used in engineering and software development to manage the development of digital documentation/code that are usually produced and maintained by a team of software developers. Changes to these pieces of information are usually identified by incrementing an associated version and binding it historically with the person, group or system making the change. On a large scale it is becoming the only way to keep track of all the changes, perform roll back operations if needed, merge modifications between different systems, individual users and groups and facilitate provenance analysis.
It is becoming obvious that tracking all the changes and operations over evolving structures makes reasonable and feasible to apply versioning methodologies in adaptive hypermedia systems. In this paper we will consider basic principles and operations of revision control applications in Adaptive Hypermedia Systems (AHS). We will not only start investigating the use of versioning for designing the structure and content of AHS applications but also take a look at the user model evolution in a versioning context. 2 Background
Engineering version control systems emerged from a formalized processes of tracking document versions. The main idea was to give a possibility to roll back to an early state of the document. Moreover, in software engineering, version control was introduced to track and provide access and control over changes to source code. Nowadays it is widely used to keep track of documentation, source files, configuration settings, etc.
As a system is designed, developed and deployed, it is quite common for multiple versions of the same system to be deployed with minor or major differences within different environments, furthermore system configuration/settings, state or/and context evolve which results in multiple co-existing system versions as well.
If we look at the revision control system as a system used to deliver a certain version of source code, documents, tables, or any file type or folder to a particular user at a given time or on request it may resemble delivering adapted content to a user in an Adaptive Hypermedia System. Of course version control systems are straightforward and don’t take into account any user model or context changes, don’t have any semantic or proof layers or any conceptual knowledge representation, but on the other hand provide very flexible techniques to deliver selected content in a certain context to a designated user, and facilitate methods to save and track system changes in a very efficient way.
At the same time AH systems store and process a lot of information, which is highly sensible to a context data and vice versa - context data itself, is highly dynamic and influences the adaptation process. Thus tracking AHS alterations and environment changes is a tightly connected task which requires storage and computational resources. 3 Revision Control in AHS
Hereafter we consider basic aspects of versioning in application to AHS, provide analysis of core terms and operations in parallel with AH. We take up major types of changes in a context of AH systems, models and process flow, thus coming up with a taxonomy of AH versioning techniques, types of changes and applications areas. As a result we come up with two examples typical for the AH field: one is an e-learning application providing an adaptive course and the other is a recommender system. We encapsulate these examples in a versioning context. 3.1 Types of Changes and Versioning Flow
In a layered structure of a generic AHS [
Fig. 1 Versioning process highlights
System deployment and analysis – changes done to an AHS to deploy the system in different environments, usually performed by the domain experts, whereas taking up the specifics of the domain to set up system logging, and analysis facilities.
We are not discussing technological aspects in depth, but can speculate about
versioning (and related) applications to be used, and can foresee a number of
wellknown methodologies such as conventional source control systems and historical data
bases for storing and querying for changes. OLAP technologies would suit to perform
analysis. In addition a variety of custom-build systems (such as ontology versioning
available in OntoView [
Fig. 2 AHS versioning taxonomy 3.2 Concepts and Operations of Revision Control
Here we would like to discuss the basic concepts of revision control and how they refer to the concept of evolution in AH systems. We will try to give a clear idea of how a particular concept or operation can be applied and will draw parallels with evolution and changes in AH through examples. 3.2.1 Basic versioning terms
Baseline – an approved version, usually the most stable, from which all the consequent versions are derived. In the AH field this is usually the core application, consequently used in different environment settings, within different user groups and thus being tuned up for those particular settings, which results in evolution and therefore new versions of the application. If we talk about UM – then the default version with which the user started can be considered as the baseline. In case of further analysis baseline is always the point of comparison of the application functionality and stability.
Branch – an instance is considered to be branched at a point in time, when from that time and on, two concurrent but independent versions of the same instance coexist. As we have mentioned above, any AHS may result in a new set of settings or being used by a completely different group of people, thus a tune-up of the system will be required to meet the new settings or requirements. In these terms every AH system or model emerges in a branch of the initial baseline, representing a new authored version of the application (model/sub-system). Hence these systems may coexist.
Branch can be also used as a prior concept of the User Model update, presentation generation or any other functionality which preview or try-out is desirable to analyze system behaviour. Even though the branch itself may not be used in the forthcoming version or update, a preview of a presentation or a try-out version of the application or engine will help to identify, observe and analyze the “what-if” case, and with the successful outcome commit branched changes, label as a new baseline or a head revision or merge them. This can be applied in terms of UM as well, considering the case when committing user properties to the latest application state is not desirable, however would be advantageous to see what can happen with the user in the new application environment. Though this UM case deviates from the original concept of the branch in a source control, it still uses the same principle of a concurrent instance, and in our case it is mapped on a context of UM in AH.
Head – the most recent commit, in other words the most latest version of the application/model existing on a single or multiple branches, with the latest functionality aspects prescribed to this branch(es). For UM the head revision is always the latest state of the model with all the corresponding values being updated to the latest state in the application.
Commit – applying changes to a central repository (branched or baselined) from
one or multiple working sources. Usually the authoring and set-up processes require a
number of experts, correspondingly a number of working instances which should
emerge in a baseline version which in turn will require to commit all the authors’
changes. In User Modeling we consider committing the changes to the user model
values. At the same time not all the latest user changes should be committed to the
user model in order to provide system stability (for example as it was done in AHA!
system [
Conflict – conflict occurs when the system can’t reconcile changes made to the same instance of the structure. E.g. when the same concept properties are changed by two different authors, trying to commit them in the same course may result in a conflict which will require reconsideration of a concept structure and manual interference. Though conflict resolution may not guarantee such a property of the AH system as confluence, we can anticipate that versioned and annotated structures of the Domain or Adaptation model will be helpful in confluence analysis.
Change (difference, delta) – represents a specific modification under the version control. Though the granularity of change may vary from system to system, change always represents the basic structure of versioning concept. Changes evolve from the baseline, are branched, become new baselines, are committed, then resolved and merged. Depending on AH system, different instances can be changed (concepts, relationships, properties, content, context, rules or even a complete model or subsystem).
Moreover considering changes – is the best way to compare system functionality and settings, which is very important for analysis of inherited and evolved (branched – in terms of version systems) functionality of AHS.
Merge – in conventional source control systems – merge usually refers to a reconciliation of multiple changes made to different copies of the same file. In AH merge can be considered as part of an authoring or set-up process, where all the development and changes from different authors or domain experts come together. In general we can perform the merge operation over concepts, relationships, properties, individual user properties and user groups, ontologies, rule sets, etc. As a result merge provides a clear picture how (from which sources) the concerned concept or a property has emerged, facilitating application playback and analysis of changes. In applications where merge regulates UM values this operation can be granted to the user as well.
Resolve – is a user intervention in conflict resolution of 2 or more conflicting versions. We refer this to the authoring stage and let the experts decide on the outcome, or facilitate end-user resolution when dealing with UM.
Tag, Label – refer to an important snapshot of the system or its part in time implementing a certain functionality which has to be marked/tagged to settle a new state of the application.. This may be a stable Domain Model structure or a new user group with common interests (for instance).
At the same time an explicitly labeled structure of the model may be used in the context of OLAP (On-Line Analytical Processing). We will elaborate on this case below.
As we conclude, all aforementioned concepts of versioning and source control as they are applied purely in software engineering and document control can be represented in terms of Adaptive Hypermedia Systems, facilitating system flexibility and extensibility 3.2.2 Versioning operations
Having considered the basic concepts of versioning in the context of AHS, we can
come up with the following classes of operations which will reflect typological and
structural types of changes. The following taxonomy of operations was proposed in
[
Transformation – is usually a set of actual changes taking place over the evolving structure. These may be: changing properties of concept, classes (in case of ontologies), adding or removing concept structures.
Conceptual changes (may include concept and concept relationship changes) will refer to conceptual and abstract changes of the representation, structure and relationships within the AH system. This type of changes includes changes of types, relations, conceptual representation of a knowledge. It may also include relations between constructs of two versions of the model. Conceptual changes should be always supervised and carried out manually.
Descriptive changes – usually deal with metadata describing the reason, intentions, author’s credentials, etc., regarding the new version of the model, ontology, relation or a concept. Descriptive changes don’t contribute to the actual change, however may help to reason over different versions of the same instance.
Context changes - will describe the environment in which the current update was made and in which it is going to be valid. At the same time all the environment settings for a particular change can be considered as a context change. E.g. changing a concept in the Domain Model we can consider the state of all the concept relationships as well as some other Domain specific information as a local context of this change. At the same the complete system environment is a context of embedding a new adaptive engine or re-organizing user groups. As we can see from these examples – context changes can be and usually are the most space and effort consuming, moreover domain experts usually analyze all the modifications in terms of contextual changes to capture the complete picture of a particular change. 4 Use-Cases
In order to demonstrate our conceptual view of versioning in AH, we present two examples that are typical for the AH field. 4.1 E-Learning application use-case
Let’s consider the following: the same discipline is taught in two different universities, where adaptive courses Course 1 and Course 2 on this subject are provided to the students (see fig. 3). After one year, a curriculum plan in the first university (University 1) changes and one of the teachers is asked to change (refine) the course navigation structure and some of the content which results in a new version of the course (Course 1 ver.2). At the same time one of the students moves from the second university (University 2) to the first one (University 1), having received the credits for a first part of the course (Course 2). The aforementioned teacher is creating a new version of the course, reusing the complete structure and introducing only designated changes (in navigation structure, updating adaptive engine with the associated changes in rules and changing content). Thus the initial version of the navigational structure is preserved to trace back, compare and analyze how the user behaviour changes (paths, clickstreams). At the same time the student’s User Model (which has been created in another university) (UM ver.1) emerges into a new version (UM ver.2), providing the student and his supervisor the possibility to compare basic concepts of Course 2, perform reconciliation and merge the corresponding UM values (of the results that the student has achieved taking the similar course in another university), and at the same time keep the history of his studies. On the other hand the automatic concordance is also possible if only the changes made in the course are corroborated with the additional descriptive information (is accompanied by the descriptive change (see section 3.2.2)) which contains extra explanations how to reason and update values (in this case UM values) due to the evolved course structure and/or content.
The complete picture of courses updates and user (student) development becomes transparent and the course instructor may trace back all the changes done to the course, re-use, update, merge any particular part without creating anything from the scratch. He can analyze differences in and between courses.
Fig. 3 Versioning adaptive courses 4.2 Recommender system use-case
In our second use-case we would like to outline the advantages of applying revision control in recommender systems. Typically, comparing a UM state to some reference characteristics, and predicting the user’s choice, the system evolves from the initial user profile. Each recommendation step (viewing an item, ranking it and getting recommendations) corresponds to a change in the User Model, which in a conventional system is committed to the latest state not taking into account user interactions and updates. Considering a versioning structure, we can store the difference in UM between two commits, thus logging user behaviour, and annotating it with the action (ranking) done by the user at each step. Later such a pattern of successful behaviour can serve as a recommendation to the other users providing the provenance of each recommendation step (how was this recommended and when, in what context, what was the user model state at that point in time and what has influenced the recommendation). The same happens with the recommended items, which evolve, can be modified and in general can have their own history of ratings.
If the user moves from one system to another, he will face a problem of the user profile alignment (like in the previous use-case sec. 4.1), where the history of ratings that he made may help to resolve conflicts. On the other hand labeling and tagging of user and system behaviour may become a basis for OLAP analysis. Building an OLAP cube where numeric facts (or measures), categorized by dimensions can have a set of labels or associated metadata (which is essentially corresponding to the labels used in the versioning system) will allow to organize everything in a form of multidimensional conceptual view and apply ‘slice and dice’ operation as the most
Fig. 4 Versioning in Recommender system
To conclude this use-case we would like to outline the ideas and principles used in a context of Recommender System. A new modification of product in the stock can be designated by a new version, it inherits corresponding properties and ratings. Versioning makes possible for products (product concurrent versions) to co-exist. It also helps to store product updates and a context of these changes efficiently. Thus we can create concurrent instances of products and ratings within different applications, inherit ratings, properties and merge these changes. The User Model versioning allows to analyze user step-by-step behavior (incl. user trend, etc.), compare changes with other users in order to provide similar recommendation patterns in a case of successful outcome. 4.3 Use-cases conclusions: For the aforementioned use-cases we would like to conclude with a summary of approaches we suggested in the introduction and outline the advantages of versioning: Authoring – versioning helps to create, maintain and re-use concurrent versions of an application, model or a particular property and value (e.g. course and a corresponding content), saving authoring effort.
Storing – versioning offers an efficient way to store changes, annotate them, label, present in a hierarchical structure. This saves space for a large scale system and keep track of the system history.
Maintenance – structured changes and a set of basic concepts and operations (merge, resolve, commit, tag/label, head and branch) are sufficient to maintain and reconcile application conflicts, create concurrent versions or comprise functionality implemented in different systems in one application (merging changes from different branches).
Logging – logging incremental changes (of the application, user model or a context) provides playback possibilities and serves as a ground of system analysis. Logging in terms of UM updates (or steps taken) will provide a basis for comparison of behaviour patterns and advance provided recommendations.
Analysis – versioning facilitates analysis of the step-by-step system and user behaviour to identify trends; tagging and labeling can facilitate more complex analytical approaches (such as OLAP) providing the required structure and description of the data; hierarchical incremental logging results in a clear and transparent structure of system changes and an overall evolution picture. 5 Conclusions and Further Work
In this article we tried to map a conventional versioning approach onto the field of Adaptive Hypermedia, providing clear parallels in terms of versioning concepts and operations. We tried to make first steps in this direction in order to show advantages of this approach and outline the perspective of applying these techniques and methodologies in order to facilitate the transparency of AHS evolution through versioning.
Considering further work directions we can think of describing layers of a generic adaptation framework in a versioning context (or essentially looking at the basics of adaptation through versioning), investigate technologies (e.g. source control, historical data bases, etc.) that meet the requirements of the framework and come up with the detailed structure and operations to devise these versioning facilities. Acknowledgements. This research is supported by NWO through the “GAF: Generic Adaptation Framework” project.