This doctoral thesis will exploit the advantages that can be obtained by leveraging ubiquitous learning and Linked Data benefits to support teachers in creating ubiquitous situations. In the literature there are several research proposals that explore this combination. However, to the best of our knowledge, existing proposals ofer ubiquitous learning experiences that are limited to specific knowledge and tied to a single (virtual or physical) space. For this reason, the research question “How to support teachers to exploit Linked Open Data to generate and reuse multi-domain and multi-space tasks for ubiquitous learning?” is been addressed. This proposal aims at answering this research question following the Systems Development Research Methodology using an iterative approach throughout the thesis lifespan. Two contributions are expected from this thesis: 1) to characterise the stages to be followed by teachers in order to generate ubiquitous educational experiences from diferent Linked Open Data sources; 2) propose and materialise a distributed software architecture that provides support to complete the above stages.
facilitate this management data. Authoring tools make it simple for teachers to generate content for their students The widespread use of mobile devices (laptops, tablets, using technologies with which they are familiar. Over mobiles) ofers an excellent opportunity to use them in the years, diferent authors have tried to take advantage education, blurring the limits that may exist between of the benefits of u-learning and the growing popularity learning that takes place in- and out-the-classroom [1, of personal mobile devices by developing systems that 2]. This type of learning is called ubiquitous learn- include authoring tools. An example is QuesTInSitu [8], ing (u-learning) [3]. Among the main advantages of u- a map-based application that supports the enactment of learning [4] is the increased autonomy of learners, as they learning tasks located in geospatial objects through mocan learn without the limitation of location or time [5] bile devices. This application allows teachers to create through diferent spaces. As indicated in [ 1], a space is learning tasks (multiple-choice questions (MCQs), mulunderstood as “the dimensional environment in which tiple response and true or false) and routes. Another objects and events occur, and in which they have rela- example of u-learning application is the aforementioned tive position and direction.” An application that could be SmartZoos [6]. It includes an authoring tool that allows considered u-learning is “Google Arts & Culture”.1 This to position learning tasks (information, MCQs, multiapplication enables students to explore cultural places ple responses, freeform answer, match pairs, photo and around the world. SmartZoos [6] is another u-learning embedded content) and to create activities (or routes) application that allows students to carry out ubiquitous composed of diferent tasks. Its graphical interface is also learning tasks. These learning tasks are created through based on maps. an authoring tool. In the above two examples of u-learning systems, the
Teachers may encounter dificulties in facilitating per- teachers must create the learning tasks positioned for vasive activities with their students if they are unable to their students. Repeating this process for a large numinclude learning tasks customised for individual needs. ber of tasks can become tedious and the workload can For this reason, existing solutions, as authoring tools [7], become excessive. In addition, it is possible that the information entered by teachers in such tasks may only be Proceedings of the Doctoral Consortium of the 18th European Confer- valid for that application (“isolated data silos” [9]) forcence on Technology Enhanced Learning, 4th September 2023, Aveiro, ing teachers to repeat the creation of the content if they P$orptuagbal olgz@gsic.uva.es (P. García-Zarza) wish to use it in another environment and avoiding that 0000-0001-5121-0786 (P. García-Zarza) teachers’ work can be reused by other teachers. Teachers
Copyright © 2023 for this paper by its authors. Use permitted under Creative Commons License who want to generate activities on u-learning systems 1 hCPWrEooUrctkReshtdoinpgpssIhStpN:/c1e:6u1r3-w/-0s.o7r3g/arACttstEraibUnutRdiocnWu4.0lotInurtekrrensa.hgtiooonpaol g(PClCreoB.cYcoe4.m0e).d/ings (CEUR-WS.org) may also be confronted with the cold–start problem [9].
system that still lacks content because only few previous users have made contributions up to that moment.
The principles of Linked Data [10] (LD) were intro- Teachers Students duced to enhance the data sharing and reusability (avoiding silos) and its use by both humans and machines by means of so-called Semantic Web technologies. Further- How to use LOD in more, when LD is shared with an open license, it is called How to manage different virtual and Linked Open Data [10] (LOD). There is a lot of LOD LOD without physical spaces? information available (as in the projects Wikidata2 and teecxhpneircasle? DBpedia3) and some of them are already being used in How to use LOD with the community of Technology Enhanced Learning (TEL) educa onal purpouses as LOD is suitable for education [11]. LOD datasets can in different domains? be from many domains.4 For u-learning, the geospatial domain plays a key role as it allows for more contex- Access, author and reuse Linked Open Data tualised learning. For example, Wikidata repository –a general purpose repository– stored5 more than 300,000 items that have geographic coordinates in Spain. Therefore, education-related LOD-based systems seek to take advantage of the huge amount of open data shared by different providers (in many cases, governmental agencies Mul domain Mul ple spaces following open data policies), which can help to avoid cold–start problem [9]. Figure 1: Stakeholders to be supported and issues to be ad
The GSIC–EMIC6 research group has created some ap- dressed in the doctoral thesis. plications that combine the use of u-learning and LOD to take advantage of the benefits mentioned in the previous paragraphs. An example of an application that uses u- plications are an excellent tool for teachers to generlearning in combination with LOD is “Casual Learn” [12]. ate ubiquitous activities. However, as Figure 1 shows, It is a distributed application that ofers students the op- we have identified the problems shown in the following portunity to carry out ubiquitous learning tasks related paragraphs that none of these systems fully address. A to Art History via a map-based interface. The learning summary of the limitations overcome by each of the extasks used in this system are located in Castile and León isting solutions in the literature can be found in Table 1. (Spain) and were generated semi-automatically using in- This dissertation aims to bring together the benefits of formation from multiple open repositories and a series u-learning and LOD to answer these problems. of templates validated by teachers [13]. This early at- With respect to the first limitation , for an educational tempt is limited to a single domain (Art History) and to solution based on LOD it is necessary for teachers to be a specific area of Spain due to the scope of some of the able to adapt and personalise information represented by sources of information used. "Casual Learn" does not means of Semantic Web technologies to their educational have an authoring tool for teachers to generate or adapt needs. By adapt and personalise information, we mean new information or learning tasks. Another application accessing, adding, and editing existing information so from the same research group is LOD4Culture [14], an that teachers can enrich it whenever possible. Teachers application that allows the visualisation and filtering of are not used to working with Semantic Web technoloLOD information related to Cultural Heritage and it has a gies, so, for example, they cannot be asked to use the global coverage (its LOD sources are Wikidata and DBpe- SPARQL language to query RDF data.7 Similarly, teachdia). This application does not allow to create or carry out ers cannot be asked to make SPARQL Update requests to learning tasks. As with “Casual Learn”, LOD4Culture is carry out modifications to information. It is necessary to an application limited to a single domain and it does not provide them a human–centered interface for accessing allow teachers to generate or adapt existing information. and managing LOD in order to support u-learning. This
Due to the advantages ofered by LOD, some of them may be of particular interest in domains where little or highlighted in the preceding paragraphs, LOD-based ap- incomplete information is available because, as we have previously indicated, they will be able to complete and enrich it. However, the annotation of places and learning tasks can be a very tedious job for teachers if they want
3https://www.dbpedia.org/ 4http://cas.lod-cloud.net/ 5https://w.wiki/6hzx 6https://www.gsic.uva.es/index.php?lang=en
complete the tasks proposed by their teacher (e.g., take pictures of the diferent types of leaves, record a video explaining the main characteristics of the tree they are looking at). In this example, learning is not being limited to a single space, it is continuous across several [17, 18] (physical and virtual space). Providing support to multiple spaces would foster one of the desired features of u-learning.
The rest of this document is structured as follows. Section 2 will present the research question and the goals to be achieved in the thesis. The methodology and the temporal organisation will be the focus of Section 3. Section 4 will discuss the current work that has been completed and the prototype’s status. Finally, there will be a brief discussion in Section 5 showing the main conclusions reached and the future work. to conduct ubiquitous activities involving many places if they have to do it manually. In an analogous way as in 2. Research question and goals non-LOD-based applications for u-learning, they would have to add the places, which, in the simplest case, should Given the research context and problem described beinclude their location and description. In addition, teach- fore, this dissertation aims to answer the research quesers would have to add to each place the learning tasks tion (RQ): how to support teachers to exploit Linked they would like their students to carry out, and possibly Open Data to generate and reuse multi-domain and some of these tasks would be similar (or the same task multi-space tasks for ubiquitous learning? could be reused for several places). For this, the auto- Figure 1 shows teachers and students as the stakeholdmatic (or semiautomatic) creation of learning tasks and ers of this thesis. Teachers are the main stakeholders places from LOD information can be helpful. Moreover, of this thesis since they will be supported for making even if one teacher had to enter this information man- use of LOD to create and enact u-learning situations. ually, other teachers could reuse it. Other authors have These experiences will be composed of learning tasks studied the problems of access and management of LOD and geospatial objects (known as“features” in the spatial outside the educational domain [15, 16]. domain [19]) positioned according to some coordinate
The second limitation identified is related to the fact reference system [20]. Students will also be, to a lesser that some solutions in the literature are often related to degree, stakeholders. Teachers will generate ubiquitous a single domain (e.g., [12, 14, 6]). This is ineficient as activities so that students can learn in multiple spaces. teachers developing ubiquitous educational activities will In addition, they could also suggest new learning tasks have to learn to work with multiple tools. In addition, to be carried out by their classmates. multiple tools are being developed to do similar activities To answer the RQ, and as can be seen in Figure 2, three in diferent domains (e.g., [ 12, 6]). Both problems could goals have been defined. The first of these goals is stated be addressed by proving an approximation that could be as “to enable teachers to access and author geospatial obgeneralised to as many domains as possible. jects based on Linked Open Data for educational purposes”.
Similarly, the existing solutions (e.g., [14, 6]) limit stu- Therefore, the first objective aims to work on supporting dents to work in a single physical or virtual space. The teachers to access and reuse information stored as LOD third limitation can be a disadvantage if teachers wish and to reduce the workload involved in generating new to conduct activities that involve multiple spaces. Let information or enriching existing data. us imagine a high school teacher that wants to explain The multidisciplinary nature of the information stored to her biology students what the distinctive features of as LOD is aimed to be used to answer the RQ as can be some species of local trees are. To do so, the teacher seen in the second objective, “To provide teachers with could base her explanation on diferent electronic re- support to generate learning tasks for diferent domains sources (e.g., virtual maps, pictures, videos) and assess based on information from Linked Open Data”. The meanher students through an electronic form while they are ing of automatically (or semiautomatically) generated in–class (virtual space). Learners who have done the pre- learning tasks for some learners or others will depend on liminary activity with their teacher, on their way home the quantity and quality of LOD available. If the domain from school, can use their mobile devices to physically data to be used by the teachers is very detailed (because (physical space) check the characteristics of the trees and the data was already that way originally or because the
- Learning across mul ple spaces using electronic devices - In-class and out-of-class learning - Students are aware of their environment Lack of support for teachers to manipulate seman c data Reuse LOD for u-learning Teachers’ contribu ons u-learning applica ons based on LOD
- Avoid cold start - Informa on is wri en once and shared - Mul -domain data Solu ons limited to specific domain
Lack of support for across mul ple space learning RQ: How to support teachers to exploit Linked Open Data to generate and reuse mul -domain and mul -space tasks for ubiquitous learning? G1_V&M To enable teachers to visualize and author geospa al objects based on LOD for educa onal purposes
G2_LT To provide teachers with support to generate learning tasks for different domains based on informa on from LOD
G3_EXP To support teachers in the crea on of LOD-based u-learning experiences
GOALS EVALUATION Evalua on the relevance of the approach for teachers in the genera on of ubiquitous learning situa ons.
Evalua on of the usability and usefulness of the applica on by carrying out pilot experiences with students and teachers. teachers have enriched it through the results of the first amount of LOD that exists in that domain or the level of objective) students of diferent academic levels will be detail of its information. This phase should also define able to use the learning tasks generated from it. However, the types of tasks that can be generated and that can be if the available information is not detailed enough for most useful for teachers and students. When defining the the academic level at which teachers want to work, they types of learning tasks, it should be considered that the may need to manually generate some of the learning solution will support multiple spaces. It makes sense to tasks. An example case is a tree where only its location, ask students to take a photograph in the physical space a brief description and that Pinus sylvestris is the tree but not in the virtual space. Although, for the virtual species. For elementary school teachers this information space, students will be able to search for a photograph. may be enough to generate a learning task to distinguish Another task to be undertaken in this phase will be to Pinus sylvestris from Picea abies (these species are often determine the process of creating places and learning present in the same areas). However, for university teach- tasks for teachers. This will also be solved by reviewing ers this information may be insuficient if they want that the existing literature. their students select the tree species and have automatic After this first step, SDRM defines the stage “ develop feedback. a system architecture” which will allow researchers to
Finally, the objective “to support teachers in the cre- build the system by specifying its functionalities and comation of Linked Open Data based ubiquitous learning ex- ponents, and defining the relationships between these periences” aims to enable teachers to provide meaning to components. At this stage, the actors that will interact educational tasks by grouping them. As indicated in the with the system will also be established. To this end, example of the third limitation in Section 1, there will the lessons learned from the state of the art will be used. be teachers who want to make the experiences ubiqui- In the final phase of this second step, the researchers tous in various places (in– and out–class) by connecting present an architecture for the system. For this thesis, the tasks that their students will carry out. To this end, the proposed architecture should take into account the teachers will need to be able to choose which tasks they diferent stakeholders (teachers, students and LOD reposiwant to be part of their experience and in which space tories), and the functions they will perform (visualisation, learners will complete them. management and reuse of content and performance of educational tasks), and the relationships that exist between them (e.g., teachers visualise and reuse LOD to 3. Research methodology generate educational tasks, students perform educational tasks designed by their teachers, LOD repositories take advantage of the new relationships generated by teachers’ annotations, etc.).
The third stage of SDRM, “analyze and design the system”, researchers will have to design how the data will be structured, the design of the interfaces to be used by users, how the communications between the distinct parts of the architecture will be designed, etc. [22]. In our case, the design of the data structure has taken into account that part of the data will be public (annotations generated by teachers) and part will be private (access credentials of users, metadata of students’ answers, etc.).
Moreover, popular design guidelines such as “Material Design”8 will be used for user interfaces.
The design created in the previous stage will be used in “build the system” for the implementation of the prototype.
This prototype should, at least partially, solve the problems highlighted in Section 1 (and shown in Figure 2).
We will review existing options to ease the development of the prototype.
In the last part of SDRM, “experiment, observe, and evaluate the system”, the researchers can check the system that their has implemented against the requirements that were established in the design stage. At this stage, The methodology chosen to answer the RQ and achieve the proposed objectives is Systems Development Research Methodology [21] (SDRM). This methodology is composed of iterative stages. Iterations make it possible to answer the RQ and refine it. The choice of this methodology for the development of the dissertation is due to the pragmatic worldview [23]. For me it is particularly important that the expected results of the thesis have a practical meaning (at least in the educational domain).
Thanks to the composition of the phases of SDRM and the possibility to refine the research question, the results obtained will solve a latent need of the main stakeholders (teachers). Figure 3 shows the tasks to be carried out in each of the SDRM stages. These tasks have been distributed over the four years in which the RQ is expected to be answered.
SDRM consists of five iterative stages. Its first stage “construct a conceptual framework” the researchers have to justify whether the RQ is meaningful. In this case, the framework will be related to u-learning and LOD and it will be built using current state of the art. At this stage, eforts will also be made to define some of the domains in which the expected solution can be used. There are domains in which LOD will be able to contribute more to the educational field than in others. This is due to the
together with the teachers, the pilots for the evaluation be used and relate their terms to a general ontology that of the system will be designed. These pilots are expected will be common to all domains. This ontology will be to involve both teachers and real students in which the based on the SLEek ontology [24] and has been previsystem will be monitored. After the educational experi- ously extended as indicated in [25]. ences have been carried out, the system will be evaluated In a second stage, templates [9] will be generated with to verify whether it is answering the research question which learning tasks will be created on the fly. Templates or whether it needs to be refined. As Figure 3 shows, are defined in [ 13] as a set of rules for generating learndissemination of the results has also been included in ing tasks from some specific types of data. Knowledge this stage. engineers and teachers will be involved in the template creation process. At this stage the knowledge engineers are aware of what kind of data are stored in the LOD 4. Current progress repositories to be used. These engineers should be able to pass this information on to the teachers, who are the As it has been seen throughout the previous sections, people to provide educational value to the LOD data. main stakeholders to be supported in this thesis are teach- They will be able to create a set of templates that can ers. To help them in the design and implementation be applied to the diferent entities for the creation of of ubiquitous educational experiences based on Linked the learning tasks. To do so, they will establish a seOpen Data, this doctoral thesis aims to have among its ries of restrictions (filters ) that will limit the scope of expected results two contributions (as can be seen in Fig- each of the templates (e.g., for this type of question only ure 2 and schematically in Figure 4). With the first contri- Gothic style cathedrals are valid). In the builder of the bution we want teachers, data engineers, and knowledge template the structure of the question will be indicated engineers together to be able to adapt information from (e.g., in the {{{cathedralName}}} take a picture of the LOD repositories to be used in education. To achieve this {{{elementName}}}). These templates will be saved as goal, in a first stage data engineers together with knowl- LODs in case other people want to reuse them. The adedge engineers must analyse the information sources to
Stage 1
Adapta on of LOD repository to the general
ontology Can we adapt these LOD to the ontology? What information is available?
Stage 2 Genera on of templates for learning tasks We have these data available
These learning tasks can
be generated
Stage 3
Crea ng educa onal experiences and carrying out
tasks
I can generate u-learning situations
I can complete u-learning situations Teacher tool
Student tool
Linked Open Data vantage of using templates to generate learning tasks on will be done through an intermediate tool that facilitates the fly instead of materialising them is that it avoids hav- the use of this type of data by people who are not experts ing to store a large amount of information. In addition, in the Semantic Web. Currently, within the thesis, this is the information used to generate a learning task on the the stage in which most work has been done. lfy will be carried out with the most updated data possible. The second contribution expected from the thesis is the However, it is possible that generating learning tasks in proposal of a distributed software architecture that makes this way may have several disadvantages associated with it possible for teachers and students to utilise the results it, such as the time required to recover the information of the first contribution. In the last year, a distributed softnecessary to generate the task. This technical problem will be studied in the thesis to decide the viability of this solution.
In the last stage shown in Figure 4, teachers will be - Manage LOD - Use LOD able to add geospatial objects that are not in the LOD -- MMuull pdloemsapiances -- MMuull dploemsapiances repositories used, generate specific learning tasks for Teachers Students characteristic elements of geospatial objects (tasks that would not make sense to generate from a template since they could only be applied to a single entity) and group Access, author and reuse LOD learning tasks and geospatial objects into itineraries in order to give a broader educational meaning to the experience (as was already done by [8, 6] systems). The teacher will be free to decide whether the students must complete the itineraries in a specific order or let the stu- Mul domain Mul ple spaces dents decide how to complete them. Students will be able to complete learning tasks individually or simply Figure 5: Problems that the current CHEST prototype aims visualise the description of the geospatial objects. Access, to address. and creation in the case of teachers, to the information ware architecture [25] and a prototype [25, 26] related propose a sequence of stages. These stages will indicate to Cultural Heritage has been designed and developed how, starting from a repository of LOD where a set of to support teachers in the creation of ubiquitous educa- geospatial objects can be found, teachers can generate tional situations. CHEST (Cultural Heritage Educational ubiquitous learning situations that their students can Semantic Tool) allows teachers to add new geospatial ob- follow and other teachers can reuse. The second contrijects, learning tasks and itineraries to a LOD repository. bution aims to propose an architecture that supports the The preliminary version of CHEST9 has already been ifrst contrition. used by real teachers and students in several pilot expe- Over the development of the thesis, we have imriences. In this preliminary version, CHEST is a Stage 3 plemented a prototype that allowed teachers to genersystem. As shown in Figure 5, CHEST has attempted to ate learning tasks, LOD-based geospatial objects, and solve the problems related to the access and authoring itineraries (ordered or unordered clustering of learning of LOD information and to the possibility of carrying tasks and geospatial objects) related to the domain of out ubiquitous activities in multiple spaces. However, Cultural Heritage. The initial data used by this system for now it is limited to the domain of Cultural Heritage. had been generated in previous work. For this reason and To see if the experience gathered with the Cultural Her- considering what is represented in Figure 3, CHEST is a itage domain can be transferred to other domains, work type of Stage 3 system. As future work, the steps taken to has started with the forest domain. In this exploration, materialise the learning tasks used as initials for CHEST the low volume of LOD data Wikidata, the LOD repos- should be extrapolated. In this way, we should try to itory that had been used as a reference until now, has extrapolate the steps followed and establish an approach been detected. For this reason, other data sources are that would allow the implementation of other domains being explored for domains diferent from Cultural Her- such as the forestry domain or the school preservation itage. Now we are working with the information stored education domain (domains where other researchers in in OpenStreetMap10 whose information is generated by the GSIC–EMIC research group have worked). a large community of users (over 10.6 million registered To check that the research question is being answered, users)11. As future work, as shown in Figure 3, the possi- an evaluation of the results achieved will be carried out. bility of the creation of learning tasks based on LOD and This evaluation will consist of testing the sequence of the templates explained in the first contribution remains stages in domains that have not been worked with. For to be explored with CHEST. this, the involvement of teachers (who teach at diferent levels of education) and domain experts (data and knowledge engineers) external to the research group where the 5. Discussion thesis is being developed will be desirable. On the other hand, a series of pilots will be developed to study how In the analysis of the state of the art that brings together the architecture (and the prototype implementing this ubiquitous learning and Linked Open Data principles, architecture) is able to support teachers in generating it has been found that none of the current systems, to ubiquitous educational experiences and how students can the best of our knowledge, jointly provide support for carry out learning tasks in diferent spaces and domains. teachers to manage Linked Open Data and support multi- The collaborative system that can be obtained as a domain and multi-space ubiquitous activities. For this contribution of this thesis is not without problems. For reason, this research thesis aims to answer the RQ: “how example, allowing teachers to create geospatial objects to support teachers to exploit Linked Open Data to gen- can lead to several teachers annotating the same object, erate and reuse multi-domain and multi-space tasks for which can hamper the overall usability of the system. ubiquitous learning?”. To answer this question, it has This could be because, for example, teachers want to inbeen further subdivided into three goals: 1) to enable clude diferent information in the geospatial object (e.g. teachers to access and author geospatial objects based on a description adapted for their students), they want to Linked Open Data for educational purposes; 2) to provide own it so that they can modify it, or they simply did teachers with support to generate learning tasks for dif- not realise that it was already present in the system. To ferent domains based on information from Linked Open avoid this problem, a number of checks will have to be Data; 3) to support teachers in the creation of Linked made (e.g., checking that the new feature has a diferOpen Data based ubiquitous learning experiences. It is ent name from the other near geospatial objects) and expected that by the end of the dissertation the objec- enabling a community validation system at the time of tive will have been achieved through the completion of creating these features. In addition, it could be asked two contributions. The first and most theoretical will whether this information on geospatial objects could be more useful if instead of being stored in a public reposi190hhtttptpss:/://c/whewstw.g.osipce.unvstar.eesetmap.org/ tory of the research group, it could be stored directly in 11https://planet.openstreetmap.org/statistics/data_stats.html a more general repository (e.g., Wikidata).
EMIC for their ideas and contributions. The author would especially like to thank J.I. Asensio-Pérez, M.L. Bote-Lorenzo and G. Vega-Gorgojo for their support. This work was supported in part by Grant PID2020-112584RB-C32 funded by MCIN/AEI/10.13039/501100011033, and in part by Grant LOD.For.Trees (TED2021-130667B-I00) funded by NextGenerationEU and MCIN/AEI. Pablo García Zarza has been funded by the call for predoctoral contract UVa 2022, co-funded by Banco Santander.
Throughout the Doctoral Consortium paper it has been indicated that teachers are the ones who will annotate learning tasks and geospatial objects, but why limit this annotation to teachers? Why not allow students to generate ubiquitous learning situations? It is probable that more verification systems would need to be implemented when making public the information generated by students. However, students could create activities for their classmates (and the rest of the community) with a diferent focus than their teachers might have.
If the expected contributions of the thesis are achieved and the contributions are successful (a large number of teachers and students from multiple domains use the system), besides the technical problems associated with all the information that could be created (and even more so if students can also annotated LOD), it could be very dificult to find specific information. For this reason, it will be necessary to adapt the information displayed to each user (for example, only show the user the tasks that correspond to their educational level) or let them filter it (e.g., students who are using the system in the domain of cultural heritage may at a certain moment want to see only the palaces of the area where they are located). of cultural heritage linked open data, Semantic Web [25] P. García-Zarza, M. Bote-Lorenzo, G. Vega-Gorgojo, Preprint (2023) 1–30. doi:10.3233/SW-233358, J. Asensio-Pérez, Towards a teacher application publisher: IOS Press. to support semantic annotations of learning tasks [15] A.-S. Dadzie, E. Pietriga, Visualisation of Linked in cultural heritage, in: Proceedings of the Ninth Data - Reprise, Semantic Web 8 (2017) 1–21. doi:10. ACM Conference on Learning @ Scale, L@S ’22, 3233/SW-160249, publisher: IOS Press. Association for Computing Machinery, New York, [16] T. Heath, J. Domingue, P. Shabajee, User Interac- NY, USA, 2022, p. 436–440. doi:10.1145/3491140. tion and Uptake Challenges to Successfully Deploy- 3528335. ing Semantic Web Technologies, in: 3rd Interna- [26] P. García-Zarza, M. Bote-Lorenzo, G. Vega-Gorgojo, tional Semantic Web User Interaction Workshop J. Asensio-Pérez, CHEST: A Linked Open (SWUI2006) co-located with the 5th International Data-based Application to Annotate and Carry Semantic Web Conference, 2006, p. 14. Out Learning Tasks About Cultural Heritage, [17] J. A. Muñoz-Cristóbal, V. Gallego-Lema, H. F. in: I. Hilliger, P. Muñoz-Merino, T. De Laet, Arribas-Cubero, J. I. Asensio-Pérez, A. Martínez- A. Ortega-Arranz, T. Farrell (Eds.), Educating for Monés, Game of blazons: Helping teachers conduct a New Future: Making Sense of Technologylearning situations that integrate web tools and Enhanced Learning Adoption, Lecture Notes in multiple types of augmented reality, IEEE Transac- Computer Science, Springer International Pubtions on Learning Technologies 11 (2018) 506–519. lishing, Cham, 2022, pp. 441–447. doi:10.1007/ doi:10.1109/TLT.2018.2808491. 978-3-031-16290-9_34. [18] S. Serrano-Iglesias, E. Gómez-Sánchez, M. L.
Bote-Lorenzo, J. I. Asensio-Pérez, A. Ruiz-Calleja, G. Vega-Gorgojo, Demonstration of scarlett: A smart learning environment to support learners across formal and informal contexts, in: T. De Laet, R. Klemke, C. Alario-Hoyos, I. Hilliger, A. OrtegaArranz (Eds.), Technology-Enhanced Learning for a Free, Safe, and Sustainable World, Springer International Publishing, Cham, 2021, pp. 404–408. [19] ISO 19101-1:2014, Geographic information — Reference model — Part 1: Fundamentals, Standard 19101, International Organization for Standardization, Geneva, Switzerland, 2014. [20] ISO 19111:2019, Geographic information — Referencing by coordinates, Standard 19111, International Organization for Standardization, Geneva,
Switzerland, 2019. [21] J. Nunamaker, M. Chen, T. D. Purdin, Systems development in information systems research, Journal of Management Information Systems 7 (1990) 89–106.
doi:10.1080/07421222.1990.11517898. [22] A. Ortega Arranz, Supporting Practitioners in the
Gamification of MOOCs through Reward-Based Strategies, Ph.D. thesis, Universidad de Valladolid,
Valladolid, 2021. doi:10.35376/10324/47535. [23] J. W. Creswell, Research Design: Qualitative, Quantitative, and Mixed Methods Approaches, 3rd ed.,
SAGE Publications, Thousand Oaks, Calif, 2008. [24] A. Ruiz-Calleja, M. Bote-Lorenzo, G. Vega-Gorgojo,
A. Martínez-Monés, J. Asensio-Pérez, E. GómezSánchez, S. Serrano-Iglesias, Y. Dimitriadis, Sleek: An ontology for smart learning in the web of data, in: IEEE 21st International Conference on Advanced Learning Technologies, ICALT 2021, Online, 2021, pp. 365–366. doi:10.1109/ICALT52272. 2021.00117.