ReGammend: A method for personalized recommendation of gamification designs Gabriel Vasconcelos 1, Wilk Oliveira 1,2, Ana Cláudia Guimarães Santos 1 and Juho Hamari 2 1 Institute of Mathematics and Computer Science, University of São Paulo, São Carlos, São Paulo, Brazil 2 Gamification Group, Faculty of Information Technology and Communication Sciences, Tampere University, Tampere, Finland Abstract Gamification personalization has been increasingly investigated as an avenue to improve the effects of gamification. While currently, empirical data exist to start making evidence-based gamification design, current guidelines and methods to bridge the gap of evidence and design is lacking. To start facing this challenge, we outline a point of departure proposing the recommendation system ReGammend (recommendation system for gamification designs). The system tailor gamification design based on users’ traits, contextual factors, goals, and other relevant moderating factors. The recommendation system uses information from the previous literature to recommend gamification designs with multiple game elements aiming to positively affect the positive outcomes stemming from gamification. The proposed system contributes to researchers and practitioners, providing a practical way to personalize gamification designs. Keywords1 Gamification design, recommendation system, user experience, user-centered design, user modeling 1. Introduction persuasive health strategies [12]. Despite the positive results that studies have reported over the years, a considerable number of negative or mixed In the last decade, gamification (i.e., the design results have highlighted that gamification could approach of products, activities, services and not affect all users in the same way [1], [10], [13]. systems to create similar motivational experiences With that, researchers started to search for ways as games usually create [1]), has increasingly to personalize gamification and create gamified become an important research topic in different environments that would be more suitable for the contexts [2]–[5]. The application of gamification different users’ profiles and preferences [3], [14]. in contexts such as education [6], health [7], and Nowadays player and user typologies are the government services [8], in general, seeks to most investigated users’ characteristic in affect the user behavior, engaging them during the personalized gamification, with indications that use of gamified environments [9]. the user preference over gamification designs Different studies indicated that applying depends on their user types [15]. Prior research gamification could have positive results in the has also indicated that the user types are dynamic users, as students having better learning outcomes [16], what would demand from designers and [10], the raise of users’ participation in fitness researchers a constant personalization of the courses [11], or the increase of the efficacy of 6th International GamiFIN Conference 2022 (GamiFIN 2022), April 26-29 2022, Finland EMAIL: gabriel.vasconcelos@usp.br (G. Vasconcelos); wilk.oliveira@usp.br (W. Oliveira); anaclaudiaguimaraes@usp.br (A. C. G. Santos); juho.hamari@tuni.fi (J. Hamari) ORCID: 0000-0002-9037-8378 (G. Vasconcelos); 0000-0003- 3928-6520 (W. Oliveira); 0000-0002-3498-0049 (A. C. G. Santos); 0000-0002-6573-588X (J. Hamari) ©️ 2022 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0). CEUR Workshop Proceedings (CEUR-WS.org) 85 gamified systems based on these user types. This gamification has shown through recent research process would be easier with the automation of the that people have different orientations and personalization, however, albeit the considerable preferences regarding gamification design [15], number of studies that sought to personalize and therefore are affected differently according to gamification over the years, the automation of this the type of gamification design they need to use process still remains a lack in the field [3]. [21, 22]. Based on these results, studies have One prominent possibility to make the process sought to identify the most suitable gamification of personalization easier for researchers and designs for each user, considering different users’ designers could be the use of recommendation aspects (e.g., user type, age, and demographic systems (RS). RS uses prior information to create data) [3]. Overall, studies on gamification a more suitable suggestion for the users, and have personalization are focused on i) identifying the been used especially in the e-commerce and relationships between different types of game entertainment industry [17]. The elements and the user profile [23], ii) evaluating recommendations provided by a RS can be done the effects of gamification personalization on the based on several aspects, such as user user experience [24, 25], or iii) proposing demographic information or purchases historic theoretical/conceptual models to personalize the [18] and would help the user to find what best gamification [26]. suits its preferences between all the items Albeit the different users’ characteristics that available [19]. have been investigated, the player and user Thus, in the field of personalized gamification, typologies have received major attention [3]. Over RS can be a useful tool to recommend the years, researchers have worked on how personalized designs, since they can indicate to different patterns could be grouped and therefore users the gamified activities that would better fit indicate different player/user types in games and to their preferences [17]. Thus, automation of gamification systems. These player/user types personalized gamification with RS also could help normally are grounded in motivational or designers implement gamification to users who psychological theories [27], player experiences have no previous experience in the usage of [28], or even neurobiological research [29]. The gamified systems, create a more efficient choice of the player/user typology that would be personalization, as well as avoid asking the user used in a personalized gamified system, can be about their preferences constantly. one major factor on the user motivation [30], and Albeit some studies have started to seek how therefore, should be an important aspect to be to implement RS in the gamification context [17], considered in the development of gamified [20], proposals of how to implement RS to define personalized systems. the gamification design remains a lack in the field. One way that has recently started to be To start to face this challenge, in this paper we discussed to improve the personalization of present a novel approach that is an evidence-based gamification is the use of RS [17], which in short RS that provide recommendations of which would are systems/algorithms capable of identifying be the most suitable gamification design for each aspects of individual users and provide dynamic user type, according to the users’ traits. The RS recommendations (e.g., design recommendations) proposed in this paper can be adapted and [31]. RS can be classified into different plugged-in different kind of gamified systems, categories: i) personalized, ii) collaborative, iii) thus, allowing designers and researchers to content-based, iv) knowledge-based, and v) provide automatic recommendations for evidence-based [32]. gamification design. At the same time, our work The personalized recommendation systems generates insights for future studies about use user profiles and some contextual parameters dynamic recommendation of gamification designs of users to provide personalized recommendations in terms of graphical user interface (GUI). [32]. The collaborative recommendation systems use user-profile, some contextual 2. Background and related works parameters, and data of the community to which the user belongs. It recommends a similar product to a user which other users of their community are In this section, we present the main topics buying [32]. The content-based addressed in this paper (i.e., personalized recommendation systems use the user-profile, gamification and RS in gamification), and the contextual parameters as well as features of the main related works. Personalization of product. Based on this, it recommends the product 86 to the user which has the same feature as the Santos et al. [15] investigated how Hexad user product he has already purchased before [32]. The types (i.e., Achiever, Disruptor, Free Spirit, knowledge-based recommendation systems use Philanthropist, Player, and Socialiser) are the user profile, contextual parameters, product associated with the preference and perceived features, and knowledge models which keeps sense of accomplishment from different track of certain event in users’ demographics and gamification designs (Performance, Ecological, accordingly do the recommendations (e.g., social, Personal, and Fictional). The study birthday recommends a certain product) [32]. conducted by Santos et al. [15] provides insights Evidence-based recommendation systems use into which gamification designs are suitable for the user profile and previews evidence collected each user type, however, does not provide (e.g., results of research) [32]. Evidence-based practical approach to implement this recommendation systems were of particular personalization in gamified systems. interest to us as it allowed us to use previous In summary, studies on the recommendation in acknowledgement from the literature to provide gamified systems focus on personalizing system recommendations. attributes (e.g., challenges and tasks), however, do Over the years, different approaches have been not focus on personalizing the gamification used to provide dynamic adaptation of GUI in design, and at the same time do not present how different areas [32]–[34]. In the field of to automate the personalization process. At the gamification, some studies involving best of our knowledge, this is the first evidence- recommendation also have been conducted. based RS for gamification design. Table 1 present Khoshkangini et al. [20] designed and a comparison between the related work. implemented a fully automated system for the dynamic generation and recommendation of Table 1 challenges, which are personalized and Related works comparison contextualized based on the preferences, history S Y UT UTR GD EB game status, and performances of each player. [20] 2021 No No No No They conducted a long-running open-field experiment (12 weeks) involving more than 400 active participants, however, they focused on [35] 2017 No No No No proposing recommendations for challenges in [36] 2017 No No No No gamified systems without proposing recommendations for gamification design itself. Herpich et al. [35] proposed a digital picture [17] 2018 Yes No No No frame that interleaves a picture display mode with a recommender mode to promote a healthy [15] 2021 Yes Yes Yes No lifestyle and to increase well-being of elderly people. Although they used gamification as a Key: S: study; Y: year of the study; UT: used a means to increase user appreciation of the system, user typology for gamification; UTR: provide the authors also did not provide recommendations the recommendations based on user’ traits; directly related to gamification designs. GD: provide recommendations for the Su et al. [36] proposed an adaptative path RS gamification designs; EB: provide an evidence- for the teaching of geometry. The authors also based RS. proposed and evaluated a gamified prototype within the system. The results indicated that personalized recommendations are important 3. Recommendation system [36], however, the authors also did not provide recommendations related to gamification design. In this section, we present the RS proposed in Tondello et al. [17] proposed a general this work. The system aims to provide framework for personalized gameful applications recommendations of gamification designs using RS (i.e., a framework to design RS for according to the user’s types. In summary, the RS gamified applications). The framework proposed receives as input the user type and provide as by Tondello et al. [17] does not provide a RS per output, a recommendation of gamification design se, but it helps the community to create RS for for the user. An example of implementation is gamified systems. presented. 87 3.1. Materials and method related Hexad user types with the gamification designs proposed by Toda et al. [40]. The work was organized in two general steps: To design the general architecture of the RS, i) RS design (general architecture) and ii) RS we used the framework proposed by Tondello et implementation (example of implementation). In al. [17]. This framework defines the general the first step, the general idea of the RS was inputs, processes, and outputs to implement planned according to the materials previously recommendation in gamified applications. described. In the second step, an example of Tondello’s framework was of our interest implementation was provided, so that it could be because, as far we know, is the only framework to used in different types of gamified systems. implement RS in gamified systems and can be adapted for different contexts. To define/identify the users’ types, in the 3.2. Recommendation system example of implementation presented in our design paper, we used the Hexad framework [27], which defines different orientations of users according to Initially, the general RS was modeled their preferences related to interaction with according to Tondello’s framework [17]. The gameful applications. The Hexad framework framework defines that a RS for gamification defines six different user types that a given user should have four Inputs (User profile, Items, can be (i.e., Achiever, Disruptor, Free Spirit, Transactions, and Context), a Recommendation Philanthropist, Player, and Socialiser). Hexad was model, and a Rating [17]. of special interest in our work because it is (as far The user profile should represent the user as we know) the only model for user types information that will be taken into account during identification for the gamification domain. At the the personalization process [17]. In our example, same time, it has already been validated in several we used the Hexad profile of users [37]. Items languages and is widely used in academia and must represent the system attributes that were industry [37]–[39]. However, in future uses of our used in the personalization process [17]. In our RS, Hexad can be replaced by another framework implementation example, we used the that better adapts to the application context. gamification design types proposed by Toda et al. To define the gamification designs to be [40]. recommended, in our example if implementation, The Transactions must represent how the we used the taxonomy proposed by Toda et al. personalization will be defined [17]. In our [40]. Toda’s taxonomy defines five gamification implementation example, transactions are the designs that are organized according to crossover between user types (Hexad) and motivation types and can be used to personalize gamification designs, defined according to the gamified environments. The taxonomy proposed model by Santos et al. [15]. The Context must by Toda et al. [40] was especially used in our represent the definitions made at the user level of work because it is, as far as we know, the only the system [17]. In our implementation example, taxonomy focused on proposing gamification we use information from the system designs, as well as because it is already widely administrator, who can make settings related to used in field studies. Also, in future uses of our the type of personalization they want RS, the taxonomy can be replaced by another that (accomplishment-based recommendation or better fits the application context. preference-based recommendation). Finally, to provide recommendations for The Recommendation must be the algorithm gamification designs, in our example of itself, where personalization is processed [17]. In implementation, we followed an evidence-based our implementation example, we used an recommendation model, using the study evidence-based algorithm to provide the conducted by Santos et al. [15], who identified gamification design recommendation according how the different gamification designs proposed to the results of the study by Santos et al. [15]. in the taxonomy of Toda et al. [40] affect the Finally, Ratings are the recommendations perceived sense of accomplishment and generated by the algorithm [17]. In our preference of users according to their Hexad user implementation example, ratings are the type. We used the study conducted by Santos et gamification design recommendations that should al. [15] as a basis for our recommendations for appear in the user interface. Despite the examples being, as far as we know, the only study that used/suggested in this study (e.g., Hexad [27], 88 Toda’s framework [40], and Santos’s by the User modeling unit. Therefore, it recommendations [15]), the proposed RS is contains the scores of each user type. independent of these examples and can be adapted • System database: The system database according to different needs. Figure 1 presents the contains data about previous study’ users so general structure for our RS. that the collaborative RS can compare them to the current user. Also contains the results obtained from the previous study and the variety of possible gamification designs. • Admin settings: The admin settings store information about which parameter the admin wants to use for recommending gamification designs. • Content managing unit: The content managing unit manages all the information coming from the User Model, System database, and Admin settings. It processes data to provide a rating for each possible gamification design. Returns the best design rating for the UO. Figure 1: Inputs and outputs of the • User object (UO): Contains the recommendation system (adapted from Tondello recommended gamification design for the et al. [17]). current user. Finally, the RS was defined in nine All of the RS components can be changed as per components: i) User, ii) Admin, iii) User profile, system needs. In other words, where we use iv) User modeling unit, v) User model, vi) System Hexad as a framework to identify user types, database, vii) Admin settings, viii) Content another framework that is more appropriate for managing unit, and ix) User object (UO): each context type can be used (e.g., BrainHex User: the user is the person who will use the [29], Bartle’s Archetypes [28]). Where we are gamification system. The system can identify using Toda’s taxonomy to define gamification the trait, for instance, the user can answer a designs [40], other more context-appropriate questionnaire (Hexad in our example) to taxonomies can be used. Where we are using the provide their user type to the system (as input) study by Santos et al. [15] to define transactions, and will receive the personalized system with other evidence-based information can be used. the most appropriate gamification designs for Figure 2 present the general RS architecture. their profile (as output). • Admin: The admin is responsible for managing the gamification system. It chooses which parameter to take into account when recommending a design. In the example of implementation provided in our work, the admin may choose between “user preference” or “user perceived sense of accomplishment” to the algorithm provide the recommendation. • User profile: The user profile consists of the answers to the questionnaire. • User modeling unit: The user modeling unit is the unit responsible for processing the user’s answers provided in the questionnaire. It returns the scores for each user type (e.g., Disruptor, Free Spirit, Achiever, Player, Socialiser, and Philanthropist) when using Figure 2: RS architecture Hexad. • User model: The user model is responsible to stores the information returned 89 3.3. Example of implementation Code 2 Content Managing Unit representation To provide a RS easily interpretable and const createRecommendation = async incremental, we implemented an example for the (req, res) => { RS in JavaScript (programming language highly const user = await compatible with different types of web systems). userModel.findById(req.params.user_id The system was register National Institute of ); // The User Model of a specific Industrial Property of Brazil. The current version user is taken of the system can be found in a GitHub repository with a commercial license2. Initially, following the architecture presented const accomplishment = [] in Figure 2, the User Modeling Unit was const preference = [] // Two implemented. Each user type is represented in an arrays are created to store the array. Another array is created to represent recommendations for both criteria possible choices as to the type of recommendation for (userType of (i.e., accomplishment-based or preference-based user.dominantTypes) { recommendation). Another array is created to let represent each dominant user’s type. Finally, the recommendation_based_accomplishment = last array is created to represent the recommendations possibilities (i.e., the available maxIndexBPTable designs to be recommended). The User Modeling (recommendationModel.BTable[userType] Unit is presented in the Code 1. [0], recommendationModel.PTable[userType][ Code 1 0]); User Modeling Unit representation let const userSchema = Schema({ recommendation_based_preference = maxIndexBPTable(recommendationModel.B userTypes: Array, Table[userType][1], recommendationModel.PTable[userType][ choiceRecommendationType: 1]); Array, accomplishment.push(recommendation_ba dominantTypes: Array, sed_accomplishment); recommendation: Array preference.push(recommendation_based_ }); preference); // Add recommendation to respective arrays To implement the recommendations (based on } Santos et al. [15]), two three-dimensional matrixes were created, representing the ß-value const recommendation = and the P-value, accordingly to the following [accomplishment, preference] indexation: return recommendation; recommendationTable[UserType][Criterion } ][Design]. In an example, considering the Code 2, to get the ß-value of the Philanthropist’s Finally, the function maxIndexBPTable get the preference for the social design, we have indexes which the value of ß-value is maximum, following processing, BTable[0][1][4]. get the most significant p-value and provide the recommendation. 2 Link to access the code: https://github.com/kibonusp/rs- gamification-design 90 4. Agenda for future studies suitable gamified system for the users [30], since they have different characteristics that could influence the use of the gamified system [3]. In this section, we present the work limitations, Other aspects such as demographic information, as well, an agenda for further studies. Our work gaming habits or personality traits can be contributes to the field of gamification design, addressed in future studies about RS, to providing a RS able to be adapted and plugged in investigate how the multiple user characteristics general gamified systems. However, we have (besides only user traits) can be combined into some limitations, which create the possibility to different recommendations for gamification propose future studies that would further the designs. knowledge in the field. Firstly, in our work we did Different studies have pointed out that the use not evaluate the RS. Therefore, future studies should evaluate the system in terms of of questionnaires to assess the user type present recommendation effects (considering users some limitations, as for example random preference and perceived sense of responses [41] or missing data [3]. Also, accomplishment), as well as plug the RS in measuring the user type only in the first system different systems and evaluate its efficacy in use, might not be the best option since their profile provide the personalized gamification. Future can change over time [16], [28]. Future studies studies also can compare the users’ experiences can adapt the RS to predict the user type based on when using a personalized (with the RS) and a user behavior data or their first answer to the non-personalized (without the RS) version of the questionnaire. In this way, the RS could provide gamified system. recommendations that would be adapted to the Recent studies have pointed out that using one user changes. single user characteristic to personalize gamification might not be sufficient to create a Table 2 Summary of the agenda for future studies Study proposal Motivation Type of study Contribution Studies to evaluate Validation of the RS Experimental Generation of evidences recommendation effects that automation of gamification could be done through RS Studies comparing the Validation of the RS Experimental Generation of evidences personalized and not that automation of personalized gamification could be done recommendation through RS Improvement of the RS Improvement of the Exploratory Further the literature on to create personalization recommendations and empirical how to create RS to based on multiple users' automation of gamification characteristics Studies adapting the RS Improvement of the Exploratory Further the literature on to predict the user type recommendations and empirical how to create RS to automation of gamification Studies about the impact Improvement of the Exploratory Further the literature on of the context in the recommendations and surveys how to create RS to recommendations automation of gamification Improvement of the RS Improvement of the Exploratory Further the literature on to create recommendations how to create RS to recommendations for automation of gamification game elements 91 In this version of the RS, we also did not 46505–46544, 2021, doi: consider the context of application, creating a RS 10.1109/ACCESS.2021.3063986. that could be used regardless domain. The context [3] A. C. T. Klock, I. Gasparini, M. S. can play an important role in the effectiveness of Pimenta, and J. Hamari, “Tailored gamification [30], and prior research have gamification: A review of literature,” indicated that studies about how the context International Journal of Human Computer impact the success of the implementation of Studies, vol. 144, Dec. 2020, doi: gamification strategies, are a gap in the field [1], 10.1016/j.ijhcs.2020.102495. [3], [16]. Future studies can use and adapt the RS [4] J. Majuri, J. Koivisto, and J. Hamari, to specific domains and evaluate if the “Gamification of education and learning: recommendations fit the user preferences, and A review of empirical literature,” in therefore, positively affecting their behavior. Proceedings of the 2nd International Finally, in this work we propose a RS based on GamiFIN Conference, 2018, pp. 11–19. gamification designs. To provide these Accessed: Feb. 15, 2022. [Online]. recommendations it was possible to find only one Available: http://ceur-ws.org/Vol- study in the literature that related the Hexad user 2186/paper2.pdf types with gamification designs [15]. Therefore, it [5] D. Johnson, S. Deterding, K. A. Kuhn, A. was not possible to provide individual game Staneva, S. Stoyanov, and L. Hides, elements recommendations for the users or create “Gamification for health and wellbeing: A recommendations based in different studies. Since systematic review of the literature,” there is a large number of studies that relates the Internet Interventions, vol. 6. Elsevier Hexad user types with the game elements B.V., pp. 89–106, Nov. 01, 2016. doi: individually (see [3] for a review), future studies 10.1016/j.invent.2016.10.002. can improve our RS using prior research to [6] L. Rodrigues et al., “Personalization provide recommendations for individual game Improves Gamification: Evidence from a elements for the users. These evaluations and Mixed-methods Study,” in Proceedings of comparisons studies would provide the field with the ACM on Human-Computer more evidence-based that the RS could be an Interaction, Sep. 2021, vol. 5, no. option to personalize gamification. Table 2 CHIPLAY. doi: 10.1145/3474714. summarize the agenda for future studies. [7] M. Altmeyer, P. Lessel, S. Jantwal, L. Muller, F. Daiber, and A. Krüger, 5. Final remarks “Potential and effects of personalizing gameful fitness applications using behavior change intentions and Hexad In this study, we propose a RS for gamification user types,” User Modeling and User- designs, capable of recommending gamification Adapted Interaction, vol. 31, no. 4, pp. designs according to the user type. Thus, we 675–712, Sep. 2021, doi: 10.1007/s11257- contribute to academia and to the industry. In 021-09288-6. future work, we aim to improve the RS, provide [8] S. K. Bista, S. Nepal, C. Paris, and N. recommendations based on other user Colineau, “Gamification for online characteristics. communities: A case study for delivering government services,” International 6. References Journal of Cooperative Information Systems, vol. 23, no. 2, 2014, doi: [1] J. Koivisto and J. Hamari, “The rise of 10.1142/S0218843014410020. motivational information systems: A [9] J. Hamari, “Gamification,” in The review of gamification research,” Blackwell Encyclopedia of Sociology, International Journal of Information Oxford, UK: John Wiley & Sons, Ltd, Management, vol. 45. Elsevier Ltd, pp. 2019, pp. 1–3. doi: 191–210, Apr. 01, 2019. doi: 10.1002/9781405165518.wbeos1321. 10.1016/j.ijinfomgt.2018.10.013. [10] S. Bai, K. F. Hew, and B. Huang, “Does [2] M. Trinidad, M. Ruiz, and A. Calderon, “A gamification improve student learning Bibliometric Analysis of Gamification outcome? Evidence from a meta-analysis Research,” IEEE Access, vol. 9, pp. and synthesis of qualitative data in educational contexts,” Educational 92 Research Review, vol. 30. Elsevier Ltd, [18] H. Tan, J. Guo, and Y. Li, “E-learning Jun. 01, 2020. doi: Recommendation System,” in 10.1016/j.edurev.2020.100322. International Conference on Computer [11] M. Altmeyer, M. Schubhan, A. Krüger, Science and Software Engineering, Dec. and P. Lessel, “A long-term investigation 2008, pp. 430–433. doi: on the effects of (personalized) 10.1109/csse.2008.305. gamification on course participation in a [19] J. Davidson et al., “The YouTube video gym,” in Proceedings of the 5th recommendation system,” in RecSys’10 - International GamiFIN Conference, 2021, Proceedings of the 4th ACM Conference pp. 60–69. Accessed: Feb. 15, 2022. on Recommender Systems, 2010, pp. 293– [Online]. Available: http://ceur- 296. doi: 10.1145/1864708.1864770. ws.org/Vol-2883/paper7.pdf [20] R. Khoshkangini, G. Valetto, A. Marconi, [12] R. Orji, G. F. Tondello, and L. E. Nacke, and M. Pistore, “Automatic generation and “Personalizing persuasive strategies in recommendation of personalized gameful systems to gamification user challenges for gamification,” User types,” in Conference on Human Factors Modeling and User-Adapted Interaction, in Computing Systems - Proceedings, Apr. vol. 31, no. 1, pp. 1–34, 2021, doi: 2018, vol. 2018-April. doi: 10.1007/s11257-019-09255-2. 10.1145/3173574.3174009. [21] L. Hassan, J. Rantalainen, N. Xi, H. [13] A. M. Toda, P. H. D. Valle, and S. Isotani, Pirkkalainen, and J. Hamari, “The “The dark side of gamification: An relationship between player types and overview of negative effects of gamification feature preferences,” in gamification in education,” in Proceedings of the 4th International Communications in Computer and GamiFIN Conference, 2020, vol. 2020, pp. Information Science, 2018, vol. 832, pp. 11–20. Accessed: Feb. 16, 2022. [Online]. 143–156. doi: 10.1007/978-3-319-97934- Available: http://ceur-ws.org/Vol- 2_9. 2637/paper2.pdf [14] L. Rodrigues, A. M. Toda, P. T. Palomino, [22] M. Altmeyer, M. Schubhan, P. Lessel, L. W. Oliveira, and S. Isotani, “Personalized Muller, and A. Krüger, “Using Hexad User gamification: A literature review of Types to Select Suitable Gamification outcomes, experiments, and approaches,” Elements to Encourage Healthy Eating,” in PervasiveHealth: Pervasive Computing in Extended Abstracts of the 2020 CHI Technologies for Healthcare, Oct. 2020, Conference on Human Factors in pp. 699–706. doi: Computing Systems, 2020, pp. 1–8. doi: 10.1145/3434780.3436665. 10.1145/3334480.3383011. [15] A. C. G. Santos et al., “The relationship [23] W. Oliveira and I. I. Bittencourt, between user types and gamification “Selecting the Most Suitable Gamification designs,” User Modeling and User- Elements for Each Situation,” in Tailored Adapted Interaction, vol. 31, pp. 907–940, Gamification to Educational 2021, doi: 10.1007/s11257-021-09300-z. Technologies, Springer, 2019, pp. 55–69. [16] A. C. G. Santos, W. Oliveira, J. Hamari, [24] S. Hallifax, E. Lavoué, and A. Serna, “To and S. Isotani, “Do people’s user types tailor or not to tailor gamification? An change over time? An exploratory study,” analysis of the impact of tailored game in Proceedings of the 5th International elements on learners’ behaviours and GamiFIN Conference, Jun. 2021, pp. 90– motivation,” International Conference on 99. Accessed: Feb. 15, 2022. [Online]. Artificial Intelligence in Education, no. Available: http://ceur-ws.org/Vol- May, 2020, doi: 10.1007/978-3-030- 2883/paper10.pdf 52237-7_18. [17] G. F. Tondello, R. Orji, and L. E. Nacke, [25] W. Oliveira et al., “Does Tailoring “Recommender systems for personalized Gamified Educational Systems Matter? gamification,” in UMAP 2017 - Adjunct The Impact on Students’ Flow Publication of the 25th Conference on Experience,” in Hawaii International User Modeling, Adaptation and Conference on System Sciences, 2020, pp. Personalization, Jul. 2017, pp. 425–430. 1226–1235. doi: doi: 10.1145/3099023.3099114. 10.24251/hicss.2020.152. 93 [26] G. F. Tondello, R. R. Wehbe, L. Diamond, [35] M. Herpich, T. Rist, A. Seiderer, and E. M. Busch, A. Marczewski, and L. E. André, “Towards a gamified Nacke, “The gamification user types recommender system for the elderly,” Hexad scale,” in CHI PLAY 2016 - ACM International Conference Proceedings of the 2016 Annual Proceeding Series, vol. Part F1286, pp. Symposium on Computer-Human 211–215, 2017, doi: Interaction in Play, Oct. 2016, pp. 229– 10.1145/3079452.3079500. 243. doi: 10.1145/2967934.2968082. [36] C. H. Su, “Designing and developing a [27] A. Marczewski, Even Ninja Monkeys Like novel hybrid adaptive learning path to Play: Gamification, Game Thinking and recommendation system (ALPRS) for Motivational Design. CreateSpace gamification mathematics geometry Independent Publishing Platform, 2015. course,” Eurasia Journal of Mathematics, [28] R. Bartle, “Hearts, clubs, diamonds, Science and Technology Education, vol. spades: Players who suit MUDs,” 1996. 13, no. 6, pp. 2275–2298, 2017, doi: [29] L. E. Nacke, C. Bateman, and R. L. 10.12973/EURASIA.2017.01225A. Mandryk, “BrainHex: A neurobiological [37] G. F. Tondello, A. Mora, A. Marczewski, gamer typology survey,” Entertainment and L. E. Nacke, “Empirical validation of Computing, vol. 5, no. 1, pp. 55–62, 2014, the Gamification User Types Hexad scale doi: 10.1016/j.entcom.2013.06.002. in English and Spanish,” International [30] S. Hallifax, A. Serna, J. C. Marty, G. Journal of Human Computer Studies, vol. Lavoué, and E. Lavoué, “Factors to 127, pp. 95–111, Jul. 2019, doi: consider for tailored gamification,” in CHI 10.1016/j.ijhcs.2018.10.002. PLAY 2019 - Proceedings of the Annual [38] A. Manzano-león et al., “Adaptation and Symposium on Computer-Human validation of the scale of types of users in Interaction in Play, Oct. 2019, pp. 559– gamification with the spanish adolescent 572. doi: 10.1145/3311350.3347167. population,” International Journal of [31] N. Idrissi and A. Zellou, “A systematic Environmental Research and Public literature review of sparsity issues in Health, vol. 17, no. 11, pp. 1–10, Jun. recommender systems,” Social Network 2020, doi: 10.3390/ijerph17114157. Analysis and Mining, vol. 10, no. 1, pp. 1– [39] N. Taşkın and E. Kılıç Çakmak, 23, 2020, doi: 10.1007/s13278-020-0626- “Adaptation of modified gamification user 2. types scale into Turkish,” Contemporary [32] G. Vaibhavi Subhash and P. Ramanuj, Educational Technology, vol. 12, no. 2, “Optimizing Recommender System: pp. 1–17, 2020, doi: Literature Review,” Turkish Journal of 10.30935/cedtech/7942. Computer and Mathematics Education, [40] A. M. Toda et al., “A taxonomy of game vol. 12, no. 10, pp. 3934–3939, 2021. elements for gamification in educational [33] G. Ghiani, M. Manca, and F. Paternò, contexts: Proposal and evaluation,” in “Dynamic user interface adaptation driven International Conference on Advanced by physiological parameters to support Learning Technologies, 2019, vol. 2161– learning,” EICS 2015 - Proceedings of the 377X, pp. 84–88. doi: 2015 ACM SIGCHI Symposium on 10.1109/ICALT.2019.00028. Engineering Interactive Computing [41] R. Kimpen, R. de Croon, V. vanden Systems, pp. 158–163, 2015, doi: Abeele, and K. Verbert, “Towards 10.1145/2774225.2775081. predicting hexad user types from mobile [34] T. Alves, J. Natálio, J. Henriques-Calado, banking data: An expert consensus study,” and S. Gama, “Incorporating personality in CHI PLAY 2021 - Extended Abstracts of in user interface design: A review,” the 2021 Annual Symposium on Personality and Individual Differences, Computer-Human Interaction in Play, vol. 155, p. 109709, 2020, doi: Oct. 2021, pp. 30–36. doi: 10.1016/j.paid.2019.109709. 10.1145/3450337.3483486. 94