Bayesian networks for student model engineering

Bayesian networks are graphical modeling tools that have been proven very powerful in a variety of application contexts. The purpose of this paper is to provide education practitioners with the background and examples needed to understand Bayesian networks and use them to design and implement student models. The student model is the key component of any adaptive tutoring system, as it stores all the information about the student (for example, knowledge, interest, learning styles, etc.) so the tutoring system can use this information to provide personalized instruction. Basic and advanced concepts and techniques are introduced and applied in the context of typical student modeling problems. A repertoire of models of varying complexity is discussed. To illustrate the proposed methodology a Bayesian Student Model for the Simplex algorithm is developed.     

The disenchantment of affect

In computing design, experience is often broken down, compartmentalized, and engineered: a process that often disenchants the original experience. In this paper, we demonstrate the possibility to design for experience, not by formalizing and rationalizing it, but instead by supporting open-ended engagement and appropriation. We illustrate this approach through Affector, a case study in affective computing, in which we focus on user interpretation and construction of emotional experience over its computational modeling. We derive design and evaluation strategies for enchantment that focus on supporting the ongoing construction and interpretation of experience by human participants over the course of interaction. We suggest that enchanting experiences may be designed only by approaching enchantment obliquely: not by engineering it in, but by providing opportunities where it may emerge.     

Predicting user personality by mining social interactions in Facebook

Adaptive applications may benefit from having models of users? personality to adapt their behavior accordingly. There is a wide variety of domains in which this can be useful, i.e., assistive technologies, e-learning, e-commerce, health care or recommender systems, among others. The most commonly used procedure to obtain the user personality consists of asking the user to fill in questionnaires. However, on one hand, it would be desirable to obtain the user personality as unobtrusively as possible, yet without compromising the reliability of the model built. On the other hand, our hypothesis is that users with similar personality are expected to show common behavioral patterns when interacting through virtual social networks, and that these patterns can be mined in order to predict the tendency of a user personality. With the goal of inferring personality from the analysis of user interactions within social networks, we have developed TP2010, a Facebook application. It has been used to collect information about the personality traits of more than 20,000 users, along with their interactions within Facebook. Based on all the collected data, automatic classifiers were trained by using different machine-learning techniques, with the purpose of looking for interaction patterns that provide information about the users? personality traits. These classifiers are able to predict user personality starting from parameters related to user interactions, such as the number of friends or the number of wall posts. The results show that the classifiers have a high level of accuracy, making the proposed approach a reliable method for predicting the user personality     

Designing for user benefit: a methodological cautionary tale

This paper is concerned with the problem of ‘losing sight’ of the user during the design of human-computer interfaces. A methodological practice is proposed which enables the designer to track the implications of design proposals with respect to both the benefits they confer on the user and their goodness of fit with user characteristics. The methodological practice, which is seen as a technique for early or formative evaluation, is then applied in retrospect to two adaptive prototypes which are currently being incorporated into a Multi-Author Document Preparation System. The results of this process are intended as a methodological cautionary tale for other designers, illustrating the potential consequences of failure to evaluate the implicit assumptions that are being made about users by the designer.     

Building a user model implicitly from a cooperative advisory dialog

This paper reviews existing methods for building user models to support adaptive, interactive systems, identifies sigificant problems with these approaches, and describes a new method for implicitly acquiring user models from an ongoing user-system dialog. Existing explicit user model acquisition methods, such as user edited models or model building dialogs put additional burden on the user and introduce artificial model acquisition dialogs. Hand coding stereotypes, another explicit acquisition method, is a tedious and error-prone process. On the other hand, implicit acquisition techniques such as computing presuppositions or entailments either draw too few inferences to be generally useful, or too many to be trusted.In contrast, this paper describes GUMAC, a General User Model Acquisition Component that uses heuristic rules to make default inferences about users' beliefs from their interaction with an advisory expert system. These rules are based on features of human action and conversation that constrain people's behavior and establish expectations about their knowledge. The application of these rules is illustrated with two examples of extended dialogs between users and an investment advisory system. During the course of these conversations, GUMAC is able to acquire an extensive model of the users' beliefs about the aspects of the domain considered in the dialog. These models, in turn, provide the sort of information needed by an explanation generator to tailor explanations the advisory system gives to its users.     

情感建模研究进展

情感建模在和谐人机交互及情感机器人方向有着广泛应用,是情感计算及人工心理理论研究的主要内容．本文对情感建模研究现状及其相关应用作了较为详细的分析与综述,指出了情感建模目前所面临的问题及难点．     

Modeling user behavior data in systems of engagement

The proliferation of mobile devices has changed the way digital information is consumed and its efficacy measured. These personal devices know a lot about user behavior from embedded sensors along with monitoring the daily activities users perform through various applications on these devices. This data can be used to get a deep understanding of the context of the users and provide personalized services to them. However, there are a lot of challenges in capturing, modeling, storing, and processing such data from these systems of engagement, both in terms of achieving the right balance of redundancy in the captured and stored data, along with ensuring the usefulness of the data for analysis. There are additional challenges in balancing how much of the captured data should be processed through client or server applications. In this article, we present the modeling of user behavior in the context of personalized education which has generated a lot of recent interest. More specifically, we present an architecture and the issues of modeling student behavior data, captured from different activities the student performs during the process of learning. The user behavior data is modeled and sent to the cloud-enabled backend where detailed analytics are performed to understand different aspects of a student, such as engagement, difficulties, and preferences and to also analyze the quality of the data.     

Exploring MARS: developing indoor and outdoor user interfaces to a mobile augmented reality system

We describe an experimental mobile augmented reality system (MARS) testbed that employs different user interfaces to allow outdoor and indoor users to access and manage information that is spatially registered with the real world. Outdoor users can experience spatialized multimedia presentations that are presented on a head-tracked, see-through, head-worn display used in conjunction with a hand-held pen-based computer. Indoor users can get an overview of the outdoor scene and communicate with outdoor users through a desktop user interface or a head- and hand-tracked immersive augmented reality user interface.     

情感计算数学模型的研究初探

针对计算机如何进行情感计算,提出一种情感空间的概率模型并对其进行了计算机仿真。通过构造状态的概率转移矩阵,得到每个情感状态的概率分布,从而计算出情感的熵值。情感的熵值,表达出所构造的情感的细腻程度。通过计算机仿真,验证了情感模型是符合人类情感规律的——不确定性同时又具有统计规律。最后通过不同参数的仿真,说明此模型可以产生不同的情感活动,为情感机器人数学模型的建立与情感决策支持系统提供了一种新的方法。     

A probabilistic model for interactive decision-making

A probabilistic reasoning model is defined where the decision maker (d.m.) is engaged in a sequential information-gathering process facing the trade-off between the reliability of the achieved solution and the associated observation cost. The d.m. is directly involved in the proposed flexible control strategy, which is based on information-theoretic principles. The devised strategy works on a Bayesian belief network that allows the efficient representation and manipulation of the knowledge base relevant to the problem domain. It is shown that this strategy guarantees a constant factor approximate solution with respect to the optimum of the decision problem. Some application examples are also discussed.     

Affective evaluation of user impressions using virtual product prototyping

This study aims to develop an effective and efficient method of analyzing user impressions of a product using virtual prototyping. A method to analyze the relationship between user impressions and design elements of a product using virtual prototyping was proposed, and then the method was applied to the case study of automobile interior design. Thirty‐two participants in the experiment evaluated 32 different virtual prototypes of automobile interiors generated from the combination of 17 design elements. The results of the case study showed that user impressions can be well explained by design elements. Physically large design elements, such as the shapes of frontal area, center fascia, door trim, and steering wheel, were critical areas of automobile interior design. The proposed method is expected to be an effective and efficient alternative for industrial practitioners to analyze user impressions of design alternatives at the early stage of the design process. © 2010 Wiley Periodicals, Inc.     

Numerical uncertainty management in user and student modeling: An overview of systems and issues

A rapidly growing number of user and student modeling systems have employed numerical techniques for uncertainty management. The three major paradigms are those of Bayesian networks, the Dempster-Shafer theory of evidence, and fuzzy logic. In this overview, each of the first three main sections focuses on one of these paradigms. It first introduces the basic concepts by showing how they can be applied to a relatively simple user modeling problem. It then surveys systems that have applied techniques from the paradigm to user or student modeling, characterizing each system within a common framework. The final main section discusses several aspects of the usability of these techniques for user or student modeling, such as their knowledge engineering requirements, their need for computational resources, and the communicability of their results.     

Issues in software ergonomics: user interface design and Chinese computing input methodology

This paper provides an overview of two ongoing research projects at the Institute of Systems Science (ISS). In the intelligent user interface project, the use of artificial intelligence in enhancing human-computer interaction is discussed. In the Chinese computing input methodology project, an input methodology derived from studies in cognitive psychology is discussed.     

A framework for evaluating the usability of mobile phones based on multi-level, hierarchical model of usability factors

As a mobile phone has various advanced functionalities or features, usability issues are increasingly challenging. Due to the particular characteristics of a mobile phone, typical usability evaluation methods and heuristics, most of which are relevant to a software system, might not effectively be applied to a mobile phone. Another point to consider is that usability evaluation activities should help designers find usability problems easily and produce better design solutions. To support usability practitioners of the mobile phone industry, we propose a framework for evaluating the usability of a mobile phone, based on a multi-level, hierarchical model of usability factors, in an analytic way. The model was developed on the basis of a set of collected usability problems and our previous study on a conceptual framework for identifying usability impact factors. It has multi-abstraction levels, each of which considers the usability of a mobile phone from a particular perspective. As there are goal-means relationships between adjacent levels, a range of usability issues can be interpreted in a holistic as well as diagnostic way. Another advantage is that it supports two different types of evaluation approaches: task-based and interface-based. To support both evaluation approaches, we developed four sets of checklists, each of which is concerned, respectively, with task-based evaluation and three different interface types: Logical User Interface (LUI), Physical User Interface (PUI) and Graphical User Interface (GUI). The proposed framework specifies an approach to quantifying usability so that several usability aspects are collectively measured to give a single score with the use of the checklists. A small case study was conducted in order to examine the applicability of the framework and to identify the aspects of the framework to be improved. It showed that it could be a useful tool for evaluating the usability of a mobile phone. Based on the case study, we improved the framework in order that usability practitioners can use it more easily and consistently.     

Evaluation of user authentication methods in the gadget-free world

In an ideal gadget-free environment the user is interacting with the environment and the services through only “natural” means. This imposes restrictions on many aspects of the interaction. One key element in this is user authentication, because it assures the environment and related services of the legitimacy of user’s actions and empowers the user to carry out his tasks. We present five high-level categories of features of user authentication in the gadget-free world including security, privacy and usability aspects. These are adapted and extended from earlier research on web authentication methods. We survey existing authentication methods together with some emerging technologies and evaluate these according to the features in our categories. Our results show, that no single authentication method can realise all these requirements for authentication. In conclusion, we give future research directions and open problems that stem from our observations. Especially, finding combinations of authentication factors and methods that achieve all requirements is an interesting problem in the gadget-free scenario.     

Considerate home notification systems: A user study of acceptability of notifications in a living-room laboratory

Recent field experiments on acceptability of notifications in the home showed that people generally want to be informed of urgent messages as soon as possible, whereas non-urgent messages should not be presented at all. A possible way to improve the acceptability of a notification might be to adjust the presentation mode and the timing of notifications to the message content and to the state of the user. For example, acceptability might be improved by considering user activities when selecting the best time to present the message. The relation between acceptability, presentation mode and timing has not been formally studied in a controlled home setting before. This paper presents the results of a user study, in which 10 participant couples were asked to engage in everyday home activities, and to subjectively rate factors that were expected to influence acceptability. The study was situated in a living-room laboratory in which the user activities and the timing of notifications were controlled. Questionnaire data was evaluated using cluster analysis in order to construct a semantic model that describes the relationship between user, system and environment. The key findings in the present study are: (1) acceptability could be improved by adjusting the level of intrusiveness of the presentation to message urgency: urgent messages should be presented intrusively, medium-urgent messages unobtrusively, and (2) non-urgent messages should be postponed until the message urgency has increased, or skipped if the message urgency never exceeds the predefined presentation threshold. Surprisingly, the user activities at the time of notification were not found to influence acceptability. These findings have resulted in a model of acceptability of notifications for the design of future home notification systems.     

Coalescing individual and collaborative learning to model user linguistic competences

A linguistic, pedagogic and technological framework for an ICALL system called COPPER is presented here, where individual and collaborative learning are combined within a constructivist approach to facilitate second language learning. Based upon the Common European Framework of Reference for Languages, the ability to use language is viewed as one of several cognitive competences that are mobilised and modified when individuals communicate. To combine the different types of learning underlying the European Framework, a student model has been developed for COPPER that represents linguistic competences in a detailed way, combining high granularity expert-centric Bayesian networks with multidimensional stereotypes, and is updated following student activities semi-automatically. Instances of this model are used by an adaptive group formation algorithm that dynamically generates communicative groups based upon the linguistic capabilities of available students, and a collection of collaborative activity templates. As well as the student model, which is a representation of individual linguistic knowledge, preferences, etc., there is a group model, which is a representation of how a set of students works together. The results of a student’s activity within a group are evaluated by a student monitor, with more advanced linguistic competences, thereby sidestepping the difficulties present when using NLP techniques to automatically analyse non-restricted linguistic production. The monitor role empowers students and further consolidates what has been previously learnt. Students therefore initially work individually in this framework on certain linguistic concepts, and subsequently participate in authentic collaborative communicative activities, where their linguistic competences can develop approximately as they would in ‘real foreign language immersion experiences’.     

Affectively intelligent and adaptive car interfaces

In this article, we describe a new approach to enhance driving safety via multi-media technologies by recognizing and adapting to drivers’ emotions with multi-modal intelligent car interfaces. The primary objective of this research was to build an affectively intelligent and adaptive car interface that could facilitate a natural communication with its user (i.e., the driver). This objective was achieved by recognizing drivers’ affective states (i.e., emotions experienced by the drivers) and by responding to those emotions by adapting to the current situation via an affective user model created for each individual driver. A controlled experiment was designed and conducted in a virtual reality environment to collect physiological data signals (galvanic skin response, heart rate, and temperature) from participants who experienced driving-related emotions and states (neutrality, panic/fear, frustration/anger, and boredom/sleepiness). k-Nearest Neighbor (KNN), Marquardt-Backpropagation (MBP), and Resilient Backpropagation (RBP) Algorithms were implemented to analyze the collected data signals and to find unique physiological patterns of emotions. RBP was the best classifier of these three emotions with 82.6% accuracy, followed by MBP with 73.26% and by KNN with 65.33%. Adaptation of the interface was designed to provide multi-modal feedback to the users about their current affective state and to respond to users’ negative emotional states in order to decrease the possible negative impacts of those emotions. Bayesian Belief Networks formalization was employed to develop the user model to enable the intelligent system to appropriately adapt to the current context and situation by considering user-dependent factors, such as personality traits and preferences.     

Evaluating design guidelines for reducing user disorientation in a desktop virtual environment

Navigation in virtual environments can be difficult. One contributing factor is user disorientation. Two major causes of this are the lack of navigation cues in the environment and problems with navigating too close to or through virtual world objects. Previous work has developed guidelines, informed by cinematography conventions, for the construction of virtual environments to aid user comprehension of virtual space to reduce user disorientation. To validate these guidelines, two user studies have been performed where users of a desktop virtual environment are to complete a navigation task in a virtual maze. In an initial study [12], collision detection with the maze walls was not enabled and the results indicated that the guidelines were effective for reducing disorientation but not for developing the users awareness of the environment space. A second study has been performed where collision detection was enabled. Results suggest that the use of the guidelines can help reduce the incidences of user disorientation and aid navigation tasks. However, the guidelines have little impact on users ability to construct cognitive maps of the desktop virtual environment.

Using dynamic user models in the recognition of the plans of the user

This paper is concerned with information-seeking dialogues in a restricted domain (we consider a consultation system for a Computer Science Department, delivering information about the various tasks that the users may want to perform: for example, how to access the library, get information about the courses of the Department, etc.) and presents a framework where a plan recognition and a user modeling component are integrated to cooperate in the task of identifying the user's plans and goals. The focus of the paper is centered on the techniques used for building the user model and exploiting it in the determination of the user's intentions. For this task, we use stereotypes and we propose some inference rules for expanding the user model by inferring the user's beliefs from both the sentences s/he utters and the information stored in the plan library of the system, that describes the actions in the domain. Moreover, we introduce some disambiguation rules that are applied to the information in the user model for restricting the set of ambiguous hypotheses on the user's plans and goals to the most plausible ones. This also simplifies a further clarification dialogue if it is necessary for a precise identification of the user's intentions.