Papers by Julita Vassileva
Combining social network information with collaborative filtering recommendation algorithms has s... more Combining social network information with collaborative filtering recommendation algorithms has successfully reduced some of the drawbacks of collaborative filtering and increased the accuracy of recommendations. However, all approaches in the domain of research paper recommendation have used explicit social relations that users have initiated which has the problem of low recommendation coverage. We argued that the available data in social bookmarking Web sites such as CiteULike or Mendeley could be exploited to connect similar users using implicit social connections based on their bookmarking behavior. In this paper, we propose three different implicit social networks-readership, co-readership, and tag-basedand we compare the recommendation accuracy of several recommendation algorithms using data from the proposed social networks as input to the recommendation algorithms. Then, we test which implicit social network provides the best recommendation accuracy. We found that, for the most part, the social recommender is the best algorithm and that the readership network with reciprocal social relations provides the best information source for recommendations but with low coverage. However, the co-readership network provide good recommendation accuracy and better user coverage of recommendation.
Studies in big data, Aug 23, 2017
Combining social network information with collaborative filtering recommendation algorithms has s... more Combining social network information with collaborative filtering recommendation algorithms has successfully reduced some of the drawbacks of collaborative filtering and increased the accuracy of recommendations. However, all approaches in the domain of research paper recommendation have used explicit social relations that are initiated by users. Moreover, the results of previous studies have shown that the recommendations produced cannot compete with traditional collaborative filtering. We argue that the available data in social bookmarking websites can be exploited to connect similar users using implicit social connections based on their bookmarking behavior. We explore the implicit social relations between users in social bookmarking websites such as CiteULike and Mendeley, and propose three different implicit social networks to recommend relevant papers/people to users. We showed that the proposed implicit social networks connect users with similar interests and the relations are propagated through the networks. In addition, we showed that implicit social networks connect more users than the two of well-known explicit social networks (co-authorship and friendship).
AIED Workshops, 2015
We claim that this marriage has never been closed and exclusive. It started because both AI and E... more We claim that this marriage has never been closed and exclusive. It started because both AI and Education share the goal of understanding the human process of knowing, and getting to know, i.e. learning. The difference is how the two areas exploit the understanding they aim to develop. AI is more focused on making machines that know and learn like people or better than them. AIED is more interested in supporting people to learn better.
Lecture Notes in Computer Science, 2012
In this short paper, we describe our RSS recommender system, KeepUP. Too often recommender system... more In this short paper, we describe our RSS recommender system, KeepUP. Too often recommender systems are seen as black box systems, resulting in general perplexity and dissatisfaction from users who are treated as passive, isolated consumers. Recent literature observes that recommendations rarely occur within such isolation and that there may be potential within more sociallyorientated approaches. With KeepUP, we outline the design of a recommendation process that is based on an implicit social network where the relevancy and meaning of information can be negotiated not only with the recommender system but also with other users. Our overall goal is to support the traditional notion of "word of mouth" rather than attempting to completely automate it.
Digital behaviour interventions aim to encourage and support people to change their behaviour, fo... more Digital behaviour interventions aim to encourage and support people to change their behaviour, for their own or communal benefits. Personalization plays an important role in this, as the most effective persuasive and motivational strategies are likely to depend on user characteristics. This tutorial covers the role of personalization in behaviour change technology, and methods and techniques to design personalized behaviour change technology.
International Journal of Artificial Intelligence in Education, Oct 31, 2017
Authoring tools have been broadly used to design Intelligent Tutoring Systems (ITS). However, ITS... more Authoring tools have been broadly used to design Intelligent Tutoring Systems (ITS). However, ITS community still lacks a current understanding of how authoring tools are used by non-programmer authors to design ITS. Hence, the objective of this work is to review how authoring tools have been supporting ITS design for non-programmer authors. In order to meet our goal, we conduct a Systematic Literature Review (SLR) to identify the primary studies on the use of ITS authoring tools, following a pre-defined review protocol. Among the 4622 papers retrieved from seven digital libraries published from 2009 to June 2016, 33 papers are finally included after applying our exclusion and inclusion criteria. We then identify the main ITS components authored, the ITS types designed, the features used to facilitate the authoring process, the technologies used to develop authoring tools and the time at which authoring occurs. We also look for evidence of the benefits of ITS authoring tools. In summary, the main findings of this work are: (1) there is empirical evidence of the benefits (i.e., mainly in terms of effectiveness, efficiency, quality of authored artifacts, and usability) of using ITS authoring tools for non-programmer authors, specially to aid authoring of learning content and to support authoring of modeltracing/cognitive and example-tracing tutors; 2) domain and pedagogical models have been much more targeted by authoring tools; (3) several ITS types have been authored, with an emphasis on model-tracing/cognitive and example-tracing tutors;
Springer eBooks, 2012
ABSTRACT This paper presents a novel peer-to-peer architecture for decentralized Online Social Ne... more ABSTRACT This paper presents a novel peer-to-peer architecture for decentralized Online Social Network and a mechanism that allows each node to filter out irrelevant social data, while ensuring a level of serendipity, by letting important information pass, even if it does not fall in the areas of interest of the user. The evaluation of the approach, using an Erlang simulation with 2318 nodes shows that it works as it was designed to: with the increasing number of social data passing through the network, the nodes learn to filter out irrelevant data, while serendipitous important data is able to pass through the network. Future work will implement the mechanism in a decentralized OSN and evaluate it in a real community of users, in an enterprise context.
Springer eBooks, 2012
In Online Social Networks (OSNs) users are overwhelmed with huge amount of social data, most of w... more In Online Social Networks (OSNs) users are overwhelmed with huge amount of social data, most of which are irrelevant to their interest. Due to the fact that most current OSNs are centralized, people are forced to share their data with the site, in order to be able to share it with their friends, and thus they lose control over it. Decentralized OSNs provide an alternative which allows users to maintain control over their data. This paper discusses an approach for propagation of social data in a decentralized OSN so as to reduce irrelevant data among users. The approach uses interaction between users to construct relationship model of interest. This relationship model acts as a filter later while propagating social data of the same interest group. This paper also presents a plan of a simulation to analyze our approach.
IEEE/WIC/ACM International Conference on Web Intelligence (WI'04), Mar 31, 2005
Decentralized P2P networks can benefit from forming interestbased communities which can provide p... more Decentralized P2P networks can benefit from forming interestbased communities which can provide peers with information about the resources shared in the community and collectively computed rating of their quality as well as about the agents in the community and their reputation. We propose a mechanism for forming communities in a P2P system for sharing academic papers. The mechanism requires each agent to compute its trust in the agents with whom it interacts. A simulation shows that such communities can benefit agents.
In this paper, we propose a novel community-based approach for web service selection where super-... more In this paper, we propose a novel community-based approach for web service selection where super-agents with more capabilities serve as community managers. They maintain communities and build community-based reputation for a service based on the opinions from all community members that have similar interests and judgement criteria. The community-based reputation is useful for consumer agents in selecting satisfactory services when they do not have much personal experience with the services. Experimental results show that our approach results in more effective service selection. A practical reward mechanism is also introduced to create incentives for super-agents to contribute their resources and provide truthful community-based reputation information, as strong support for our approach.
It is difficult for online users to keep track of their social friendships and friends' social ac... more It is difficult for online users to keep track of their social friendships and friends' social activities scattered across different social networking sites. We propose a usercentric approach for integrating social data from different social networking sites and allowing users to create personalized social and semantic contexts for their social data. Users can blend and group friends on different social networking sites. They can also rate friends and their activities as favourite, neutral or disliked. Our approach then provides personalized recommendations of friends' activities that may be interesting to each user. A prototype of a dashboard application called SocConnect (Social Connect) is also implemented to demonstrate the feasibility of our approach, along with evaluation on real users that confirms the appropriateness and effectiveness of our approach.
PGP Web of Trust where users can sign digital signatures on public key certificates of other user... more PGP Web of Trust where users can sign digital signatures on public key certificates of other users has been successfully applied in securing emails and files transmitted over the Internet. However, its rigorous restrictions on utilizable trust relationships and acceptable signatures limit its performance. In this paper, we first make some modification and extension to PGP Web of Trust by relaxing those constraints. In addition, we propose a novel method to further expand trusted neighborhood of users by merging the signatures of the trusted neighbors and finding the similar users based on the merged signature set. Confirmed by the experiments carried out in different simulated real-life scenarios, our method applied to both the modified and extended PGP methods can improve their performance. With the expansion of trusted neighborhood, the performance of the original PGP Web of Trust is also improved considerably.
Abstract This paper proposes to visualize the relationships among online community members as a f... more Abstract This paper proposes to visualize the relationships among online community members as a feedback that can motivate more active participation and reciprocation among community members. The approach is inspired by a combination of four theories of motivation from social psychology. The effect of the visualization was evaluated in an online community allowing Women in Science and Engineering to share personal stories. The results show that the visualization had a positive awareness effect, but the motivational ...
arXiv (Cornell University), Jan 14, 2021
Personalized gamification explores knowledge about the users to tailor gamification designs to im... more Personalized gamification explores knowledge about the users to tailor gamification designs to improve one-sizefits-all gamification. The tailoring process should simultaneously consider user and contextual characteristics (e.g., activity to be done and geographic location), which leads to several occasions to tailor. Consequently, tools for automating gamification personalization are needed. The problems that emerge are that which of those characteristics are relevant and how to do such tailoring are open questions, and that the required automating tools are lacking. We tackled these problems in two steps. First, we conducted an exploratory study, collecting participants' opinions on the game elements they consider the most useful for different learning activity types (LAT) via survey. Then, we modeled opinions through conditional decision trees to address the aforementioned tailoring process. Second, as a product from the first step, we implemented a recommender system that suggests personalized gamification designs (which game elements to use), addressing the problem of automating gamification personalization. Our findings i) present empirical evidence that LAT, geographic locations, and other user characteristics affect users' preferences, ii) enable defining gamification designs tailored to user and contextual features simultaneously, and iii) provide technological aid for those interested in designing personalized gamification. The main implications are that demographics, game-related characteristics, geographic location, and LAT to be done, as well as the interaction between different kinds of information (user and contextual characteristics), should be considered in defining gamification designs and that personalizing gamification designs can be improved with aid from our recommender system.
Combining social network information with collaborative filtering recommendation algorithms has h... more Combining social network information with collaborative filtering recommendation algorithms has helped to alleviate some drawbacks of collaborative filtering, for example, the cold start problem, and has increased the accuracy of recommendations. However, the user coverage of recommendation for social-based recommendation is low as there is often insufficient data about explicit social relationships among users. In this paper, we fuse recommendation that uses explicit social relations (friends and co-authors) with recommendations that use implicit social relations aiming to increase the user coverage with minimum recommendation accuracy loss. We found that fusing recommendations from friends with recommendations using implicit social networks increases both accuracy and user recommendation coverage while fusing recommendation from co-authors increase the coverage.
The 3 rd international workshop on Trust, Reputation and User Modelling (TRUM 22013) was held wit... more The 3 rd international workshop on Trust, Reputation and User Modelling (TRUM 22013) was held with the International Conference on User Modeling Adaptation and Personalization (UMAP 2013). The purpose of the workshop is : (a) to bring researchers together from the communities of trust, reputation and user modeling, and online communities where trust plays an important role, (b) to provide a forum for cutting-age research possibly not yet well evaluated, and (c) to initiate and facilitate discussions on the new trends in trust, reputation and user modeling, and to move the trends forward. In this preface, we briefly introduce the workshop, present the summary of the papers presented in the workshop and ackwledged people who have helped for the success of the workshop.
In this paper, we present a new authentication and privacy control mechanism for personalized mas... more In this paper, we present a new authentication and privacy control mechanism for personalized mashups of social networking sites. Current authentication and privacy control mechanisms lack flexibility and transparency. This mechanism can make the user model interoperation process for mashups more transparent to users. Users can have a clear understanding and control about which part of their data is being accessed by the mashup application. This mechanism is an important part of user model interoperability framework.
Springer eBooks, Jul 7, 2006
Providing mobile workers with mobile devices such as a Compaq iPaq with a CDPD card can support t... more Providing mobile workers with mobile devices such as a Compaq iPaq with a CDPD card can support them in retrieving information from centralized information systems. More specifically, mobile devices can enable mobile users to make notifications for schedule changes and add new data into the information system. In addition these devices can facilitate group communication anytime and anywhere. This paper presents different ways of providing non-critical information in a timely fashion for nomadic users of mobile devices using a wireless network. A distributed application prototype to support nomadic users is proposed, and a simulated environment is used to evaluate the prototype. Since solutions for seamless access are highly domain specific, the study involves homecare workers at Saskatoon District Health (SDH). By keeping track of the users' current context (time, location etc.) and a user task model, it is possible to predict the information needs of mobile users and to provide context dependent adaptation of both the content and the functionality. Moreover, to avoid interrupts in the user's interaction with the main information sources, methods for mobile transactions management using agent-based smart proxies that buffer, delay or pre-fetch information/data are introduced.
Uploads
Papers by Julita Vassileva
• The realization that wisdom of the crowd can be harvested only if people participate actively (e.g. rate products, provide feedback, answers, connect to other people, tag items, etc.);
• The realization that by correlating individual data of a community of users, it may be possible to generate personalized data-driven recommendations for health-related behaviour change.