Dynamic User Model Research Articles

Adapting Recommendations on Environmental Education Programs

Stakeholders in Environmental Education (EE) often face difficulties identifying and selecting programs that best suit their needs. This is due, in part, to the lack of expertise in evaluation knowledge and practice, as well as to the absence of a unified database of Environmental Education Programs (EEPs) with a defined structure. This article presents the design and development of a web application for evaluating and selecting EEPs. The certified users of the application can insert, view, and evaluate the registered EEPs. At the same time, the application creates and maintains for each user an individual and dynamic user model reflecting their personal preferences. Finally, using all the above information and applying a combination of Multi-Criteria Decision-Making Methods (MCDM), the application provides a comparative and adaptive evaluation in order to help each user to select the EEPs that best suit his/her needs. The personalized recommendations are based on the information about the user stored in the user model and the results of the EEPs evaluations by the users that have applied them. As a case study, we used the EEPs from the Greek Educational System.

Attention-based dynamic user modeling and Deep Collaborative filtering recommendation

Deep learning (DL) techniques have been widely used in recommender systems for user modeling and matching function learning based on historical interaction matrix. However, existing DL-based recommendation methods usually perform static user preference modeling by using historical interacted items of the user. In this article, we present a time-aware deep CF framework which contains two stages: dynamic user preference modeling based on attention mechanism and matching score prediction based on DL. In the first stage, short-term user preferences are modeled by the time-aware attention mechanism that fully considered the predicted item, the recent interacted items and their interaction time. The resulting short-term preferences are combined with long-term preferences for dynamic user preference modeling. In the second stage, high-order user-item feature interactions are learned by two types of DL models, Deep Matrix Factorization (DMF) and Multiple-Layer Perception (MLP), and the feature interaction vectors of the two models are fused in the last layer of the model to predict the matching score. Extensive experiments on five datasets indicate that our method is superior to the existing time-aware and DL-based recommendation methods in top-k recommendations significantly and consistently.

Adaptive Health Coaching Technology for Tailored Interventions.

Recent advances in sensor and communications technology have enabled scalable methods for providing continuity of care to the home for patients with chronic conditions and older adults wanting to age in place. In this article we describe our framework for a health coaching platform with a dynamic user model that enables tailored health coaching messages. We have shown that this can improve coach efficiency without a loss of message quality. We also discovered many lessons for coaching technology, most demonstrating the need for more coach input on sample message content, perhaps even requiring that individual coaches be able to modify the message database directly. Overall, coaches felt that the structure of the automated message generation was useful in remembering what to say, easy to edit if necessary and especially helpful for training new health coaches.

Dynamic user modeling for expert recommendation in community question answering

Community Question Answering (CQA) websites has become an important channel for people to acquire knowledge. In CQA, one key issue is to recommend users with high expertise and willingness to answer the given questions, i.e., expert recommendation. However, a lot of existing methods consider the expert recommendation problem in a static context, ignoring that the real-world CQA websites are dynamic, with users’ interest and expertise changing over time. Although some methods that utilize time information have been proposed, their performance improvement can be limited due to fact that they fail they fail to consider the dynamic change of both user interests and expertise. To solve these problems, we propose a deep learning based framework for expert recommendation to exploit user interest and expertise in a dynamic environment. For user interest, we leverage Long Short-Term Memory (LSTM) to model user’s short-term interest so as to capture the dynamic change of users’ interests. For user expertise, we design user expertise network, which leverages feedback on users’ historical behavior to estimate their expertise on new question. We propose two methods in user expertise network according to whether the dynamic property of expertise is considered. The experimental results on a large-scale dataset from a real-world CQA site demonstrate the superior performance of our method.

A multimodal conversational coach for active ageing based on sentient computing and m‐health

AbstractAs life expectancy increases, it has become more necessary to find ways to support healthy ageing. A number of active ageing initiatives are being developed nowadays to foster healthy habits in the population. This paper presents our contribution to these initiatives in the form of a multimodal conversational coach that acts as a coach for physical activities. The agent can be developed as an Android app running on smartphones and coupled with cheap widely available sport sensors in order to provide meaningful coaching. It can be employed to prepare exercise sessions, provide feedback during the sessions, and discuss the results after the exercise. It incorporates an affective component that informs dynamic user models to produce adaptive interaction strategies.

User Models for Personalized Physical Activity Interventions: Scoping Review.

BackgroundFitness devices have spurred the development of apps that aim to motivate users, through interventions, to increase their physical activity (PA). Personalization in the interventions is essential as the target users are diverse with respect to their activity levels, requirements, preferences, and behavior.ObjectiveThis review aimed to (1) identify different kinds of personalization in interventions for promoting PA among any type of user group, (2) identify user models used for providing personalization, and (3) identify gaps in the current literature and suggest future research directions.MethodsA scoping review was undertaken by searching the databases PsycINFO, PubMed, Scopus, and Web of Science. The main inclusion criteria were (1) studies that aimed to promote PA; (2) studies that had personalization, with the intention of promoting PA through technology-based interventions; and (3) studies that described user models for personalization.ResultsThe literature search resulted in 49 eligible studies. Of these, 67% (33/49) studies focused solely on increasing PA, whereas the remaining studies had other objectives, such as maintaining healthy lifestyle (8 studies), weight loss management (6 studies), and rehabilitation (2 studies). The reviewed studies provide personalization in 6 categories: goal recommendation, activity recommendation, fitness partner recommendation, educational content, motivational content, and intervention timing. With respect to the mode of generation, interventions were found to be semiautomated or automatic. Of these, the automatic interventions were either knowledge-based or data-driven or both. User models in the studies were constructed with parameters from 5 categories: PA profile, demographics, medical data, behavior change technique (BCT) parameters, and contextual information. Only 27 of the eligible studies evaluated the interventions for improvement in PA, and 16 of these concluded that the interventions to increase PA are more effective when they are personalized.ConclusionsThis review investigates personalization in the form of recommendations or feedback for increasing PA. On the basis of the review and gaps identified, research directions for improving the efficacy of personalized interventions are proposed. First, data-driven prediction techniques can facilitate effective personalization. Second, use of BCTs in automated interventions, and in combination with PA guidelines, are yet to be explored, and preliminary studies in this direction are promising. Third, systems with automated interventions also need to be suitably adapted to serve specific needs of patients with clinical conditions. Fourth, previous user models focus on single metric evaluations of PA instead of a potentially more effective, holistic, and multidimensional view. Fifth, with the widespread adoption of activity monitoring devices and mobile phones, personalized and dynamic user models can be created using available user data, including users’ social profile. Finally, the long-term effects of such interventions as well as the technology medium used for the interventions need to be evaluated rigorously.

Understanding Dynamic Cross-OSN Associations for Cold-Start Recommendation

Online social networks (OSNs) have become an essential part of people's daily life, and an increasing number of users are now using multiple OSNs for different social media services simultaneously. As a result, user's interests and preferences usually distribute in different OSNs. While most of the existing work mainly aggregates the distributed user behaviors or features directly, recently very few efforts have been focused on understanding the cross-OSN association from collective user behaviors. In this paper, we go one step further to consider the dynamic characteristic of user behaviors and propose a dynamic cross-OSN association mining framework. In this framework, dynamic user modeling is first conducted to capture the drift of user interest in each OSN. A session-based factorization method is then proposed to establish the cross-OSN association in a dynamic manner, by incrementally updating the derived association each time a new session of data arrives. Based on the derived dynamic association, we finally design a cold-start YouTube video recommendation application, by only utilizing users' behaviors in Twitter. Experiments are conducted using real-world user data from Twitter and YouTube. The results demonstrate the effectiveness of this proposed framework in capturing the underlying association between different OSNs and achieving superior cold-start recommendation performance.

Improvement of Semantic Search Results with Providing an Updatable Dynamic User Model

Improvement of Semantic Search Results with Providing an Updatable Dynamic User Model

Usability and acceptability assessment of an empathic virtual agent to prevent major depression

AbstractIn Human–Computer Interaction, the adaptation of the content and the way of how this content is communicated to the users in interactive sessions is a critical issue to promote the acceptability and usability of any computational system. We present a user‐adapted interactive platform to identify and provide an early intervention for symptoms of depression and suicide. In particular, we describe the work performed to assess users' system acceptability and usability. An empathic Virtual Agent is the main interface with the user, and it has been designed to generate the appropriate dialogues and emotions during the interactions according to the detected user's specific needs. This personalization is based on a dynamic user model nurtured with clinical, demographical and behavioural information. The evaluation was performed with 60 participants from the university community. The obtained results were promising, allowing the execution of a further clinical trial. The system's usability score was 75.7%, and the score of the user‐adapted content and the emotional responses of the Virtual Agent was 70.9%.

Topic-tracking-based dynamic user modeling with TV recommendation applications

One of the challenging issues in TV recommendation applications based on implicit rating data is how to make robust recommendation for the users who irregularly watch TV programs and for the users who have their time-varying preferences on watching TV programs. To achieve the robust recommendation for such users, it is important to capture dynamic behaviors of user preference on watched TV programs over time. In this paper, we propose a topic tracking based dynamic user model (TDUM) that extends the previous multi-scale dynamic topic model (MDTM) by incorporating topic-tracking into dynamic user modeling. In the proposed TDUM, the prior of the current user preference is estimated as a weighted combination of the previously learned preferences of a TV user in multi-time spans where the optimal weight set is found in the sense of the evidence maximization of the Bayesian probability. So, the proposed TDUM supports the dynamics of public users' preferences on TV programs for collaborative filtering based TV program recommendation and the highly ranked TV programs by similar watching taste user group (topic) can be traced with the same topic labels epoch by epoch. We also propose a rank model for TV program recommendation. In order to verify the effectiveness of the proposed TDUM and rank model, we use a real data set of the TV programs watched by 1,999 TV users for 7 months. The experiment results demonstrate that the proposed TDUM outperforms the Latent Dirichlet Allocation (LDA) model and the MDTM in log-likelihood for the topic modeling performance, and also shows its superiority compared to LDA, MDTM and Bayesian Personalized Rank Matrix Factorization (BPRMF) for TV program recommendation performance in terms of top-N precision-recall.

A Personal Data Framework for Distributed User Modelling in New Service Ecosystems

User models generated from personal data are becoming a powerful tool used by service providers to personalize their supply, or introduce new features by means of recommendations or prediction of user behavior, among other applications. However, traditional user modelling requires each service provider to acquire and manage their users' personal information and carry out complex modelling processes. In this paper, we present a comprehensive personal data framework that distributes the user modelling process so that the personal data scattered in heterogeneous sources is used to generate on-demand, dynamic user models that can be retrieved by service providers to meet their business needs. Our proposal has been validated in a financial services scenario.

Dynamic User Modeling in Social Media Systems

Social media provides valuable resources to analyze user behaviors and capture user preferences. This article focuses on analyzing user behaviors in social media systems and designing a latent class statistical mixture model, named temporal context-aware mixture model (TCAM), to account for the intentions and preferences behind user behaviors. Based on the observation that the behaviors of a user in social media systems are generally influenced by intrinsic interest as well as the temporal context (e.g., the public's attention at that time), TCAM simultaneously models the topics related to users' intrinsic interests and the topics related to temporal context and then combines the influences from the two factors to model user behaviors in a unified way. Considering that users' interests are not always stable and may change over time, we extend TCAM to a dynamic temporal context-aware mixture model (DTCAM) to capture users' changing interests. To alleviate the problem of data sparsity, we exploit the social and temporal correlation information by integrating a social-temporal regularization framework into the DTCAM model. To further improve the performance of our proposed models (TCAM and DTCAM), an item-weighting scheme is proposed to enable them to favor items that better represent topics related to user interests and topics related to temporal context, respectively. Based on our proposed models, we design a temporal context-aware recommender system (TCARS). To speed up the process of producing the top- k recommendations from large-scale social media data, we develop an efficient query-processing technique to support TCARS. Extensive experiments have been conducted to evaluate the performance of our models on four real-world datasets crawled from different social media sites. The experimental results demonstrate the superiority of our models, compared with the state-of-the-art competitor methods, by modeling user behaviors more precisely and making more effective and efficient recommendations.

Twitter is Faster

Traditional personalized video recommendation methods focus on utilizing user profile or user history behaviors to model user interests, which follows a static strategy and fails to capture the swift shift of the short-term interests of users. According to our cross-platform data analysis, the information emergence and propagation is faster in social textual stream-based platforms than that in multimedia sharing platforms at micro user level. Inspired by this, we propose a dynamic user modeling strategy to tackle personalized video recommendation issues in the multimedia sharing platform YouTube, by transferring knowledge from the social textual stream-based platform Twitter. In particular, the cross-platform video recommendation strategy is divided into two steps. (1) Real-time hot topic detection: the hot topics that users are currently following are extracted from users' tweets, which are utilized to obtain the related videos in YouTube. (2) Time-aware video recommendation: for the target user in YouTube, the obtained videos are ranked by considering the user profile in YouTube, time factor, and quality factor to generate the final recommendation list. In this way, the short-term (hot topics) and long-term (user profile) interests of users are jointly considered. Carefully designed experiments have demonstrated the advantages of the proposed method.

Panel: Perspectives on adaptivity in information retrieval interaction (PAIRI)

Adaptivity in IR interactions requires the IR systems adapting to users’ situations and the users adapting to the systems. System adaption entails dynamic user modeling, effective information architecture and enhanced search features such as search integration and relevance feedback; user adaptation through interactions entails mental model building and modification towards a coherent state of knowledge and learning. The panel is structured as follows. Initially we provide an overview of the panel contents, consisting of four central dimensions of adaptivity in IR interaction. These are adaption 1) through integration of information objects; 2) of information system to searcher; 3) of searcher to information system; and 4) to context and practice. The sequence follows the order of the panellists, i.e., each panellist is the prime mover of a particular dimension.

Cultivating Personalized Museum Tours Online and On-Site

Web 2.0 — the perceived second generation of the World Wide Web that aims to improve collaboration, sharing of information and interoperability — enables increasing access to digital collections of museums. The expectation is that more and more people will spend time preparing their visit before actually visiting the museum and look for related information reflecting on what they have seen or missed after visiting the museum. It can also be expected that museum curators want to enhance visitors' museum experiences in the more personalized, intensive and engaging way promised by an improved Web. In other words, to keep their visitors, they should adopt an immersive museum environment that combines the museum Web site (online) with the physical museum space (on-site). In this context, the CHIP (Cultural Heritage Information Presentation) project offers tools to the users to be their own curators, e.g. browsing the online collections, planning a personalized museum tour suiting their art interests, getting some recommendations about interesting artworks to see, and quickly finding their ways in the museum. In this paper, we present the new additions to the CHIP tools, which target such functionality: a Web-based museum Tour Wizard based on the user's interests and the Mobile Guide that converts the tours to a mobile device (PDA) used in the physical museum space. To connect the user's various interactions with these tools online and on-site, we built a dynamic user model. Online, the user model stores the user's personal background, ratings of artworks and art concepts, recommended or created museum tours. On-site, it is a conversion of the online user model stored in RDF into XML format which the mobile guide can parse. When the user rates artworks inside the physical museum, the on-site user model is updated and when the tour is finished, it is synchronized with the online user model. In such a way, we support a 'virtuous circle' of the museum visit, which links the personalized museum experiences both online and on-site.

Online Kranten, Een Model Van De (Tweede-)Taalleerder En Het Genereren Van Adaptief Lesmateriaal

This article explores the question whether or not a user model with an automated adaptive selection of online generated and enriched reading texts from newspapers ensures that readers of different vocabulary levels receive linguistic input that is best suited to their abilities and preferences. In a simulation study, 30 'readers' with a vocabulary size varying from 500 to 15,000 lemmas received three texts aimed at 88% text coverage (relative proportion of lemmas known to the reader) for ten days, to make sure that they understood the texts and could learn new words from them. All texts and words encountered were logged (usage data) to build up a dynamic user model. The study revealed that except for the two lower vocabulary levels (knowledge of 500 and 1000 lemmas) all texts had sufficient text coverage. After 20-25 texts, the number of words encountered for the first time in each text was about 4 to 6% (dependent on low or high vocabulary level), and not yet stable. The number of words learned after 30 texts - two to five words - was quite low. More research with far more texts is needed to find a stable pattern in the number of words learned while reading online newspapers.

A Long-Term Learning-Based Dynamic User Model in Image Retrieval

This paper proposes a probabilistic model incorporating long-term learning to estimate a dynamic user model. By using RF sequence as the user pattern, the approach can gradually update the prediction of current user based on matching the current user pattern with the user patterns in log. Compared with the invariant user model in PicHunter, the model is capable of dynamically adjusting when more user actions are observed, thus provides more accurate prediction for probability distribution. Experimental results on 11 000 images show that this approach can improve the retrieval accuracy apparently.

Understanding patrons’ micro-level information seeking (MLIS) in information retrieval situations

Elicitation is a communicative act to request information during a dialogue, which reflects the questioner’s problems at hand, perplexity, interests, or perhaps internal concerns, either voluntarily or unintentionally. A patron with an information need or a problem at hand approaches the retrieval system to ask their first question. However, many more communicative acts requesting information occur during the retrieval interaction. Elicitation during retrieval interaction, as distinct from the patron’s first search question, is termed micro-level information seeking (MLIS). What features does MLIS possess? And is MLIS predictable? These are salient research issues because if a patron’s MLIS is explainable and predictable, it is advantageous for the intermediary, be it human or an intelligent agent, to form a dynamic user model by taking the patron’s MLIS into account to provide better information retrieval support. This study explores the following research questions: (1) What are the purposes of patron elicitations? (2) When does the patron’s elicitation tend to occur? (3) Do patron elicitations differ from intermediary elicitations in terms of frequency of occurrence and time frame? (4) Does patron elicitation behavior relate to contextual variables, such as gender, age, status, knowledge, prior online search experience, individual intermediaries interacted with, or length of interaction? Qualitative and quantitative approaches including discourse analysis, content analysis and statistical analysis are applied. The major research findings include: (1) patrons’ and intermediaries’ elicitation behaviors differ in terms of frequency and time frame, supporting the prior assumption that intermediary elicitation is pre-planned and patron elicitation is situational; (2) patron’s perplexity is situational, being in most cases “search-assignment related”; and (3) patrons’ elicitation behavior is significantly related to their contextual variables. The study suggests dynamic user modeling to take account of the patron’s MLIS.

What exactly is wrong and why? Tutorial Dialogue for Intelligent CALL Systems

We introduce a dialogue-based explanation facility for Intelligent CALL (ICALL) Systems. Our prototype system, DiBEx, uses meta reasoning to build up an explanation (error) tree, given a faulty user input. It relies on correct grammatical subtheories, instead of explicit error taxonomies. DiBEx, thus, realizes anticipation free error diagnosis. The system enters in a tutorial dialogue with the student, where each explanation (dialogue) step is based on the principles of a single tutorial strategy and a dynamic user model.

User Modelling for Intent Prediction in Information Analysis

User modelling is a key element in successfully assisting intelligence analysts who must gather information and make decisions without being overloaded by the massive amounts of data available on a daily basis most of which are irrelevant. Furthermore, with user modelling, we can predict the goals and intentions of the analyst in order to better serve their information seeking tasks by providing better re-organization and presentation of data as well as pro-actively retrieve novel and relevant information as it arises. Our goal is to provide a dynamic user model of an analyst and work with him as he goes about his daily tasks.