Showing papers on "User modeling published in 2009"

Empirically building and evaluating a probabilistic model of user affect

Abstract: We present a probabilistic model of user affect designed to allow an intelligent agent to recognise multiple user emotions during the interaction with an educational computer game. Our model is based on a probabilistic framework that deals with the high level of uncertainty involved in recognizing a variety of user emotions by combining in a Dynamic Bayesian Network information on both the causes and effects of emotional reactions. The part of the framework that reasons from causes to emotions (diagnostic model) implements a theoretical model of affect, the OCC model, which accounts for how emotions are caused by one's appraisal of the current context in terms of one's goals and preferences. The advantage of using the OCC model is that it provides an affective agent with explicit information not only on which emotions a user feels but also why, thus increasing the agent's capability to effectively respond to the users' emotions. The challenge is that building the model requires having mechanisms to assess user goals and how the environment fits them, a form of plan recognition. In this paper, we illustrate how we built the predictive part of the affective model by combining general theories with empirical studies to adapt the theories to our target application domain. We then present results on the model's accuracy, showing that the model achieves good accuracy on several of the target emotions. We also discuss the model's limitations, to open the ground for the next stage of the work, i.e., complementing the model with diagnostic information.

Automatic Recommendations for E-Learning Personalization Based on Web Usage Mining Techniques and Information Retrieval

Abstract: Introduction Up to the very recent years, most e-learning systems have not been personalized Several works have addressed the need for personalization in the e-learning domain However, even today, personalization systems are still mostly confined to research labs, and most of the current e-learning platforms are still delivering the same educational resources in the same way to learners with different profiles In general, to enable personalization, existing systems used one or more types of knowledge (learners' knowledge, learning material knowledge, learning process knowledge, etc) Generally, personalization in e-learning systems concerns: adaptive interaction, adaptive course delivery, content discovery and assembly, and adaptive collaboration support The category of adaptive course delivery represents the most common and widely used collection of adaptation techniques applied in e-learning systems today Typical examples include dynamic course re-structuring and adaptive selection of learning objects, as well as adaptive navigation support, which have all benefited from the rise of using recommendation strategies to generate new and relevant links and items In fact, one of the new forms of personalization in e-learning environment is to give recommendations to learners in order to support and help them through the e-learning process A number of personalized systems have relied on explicit information given by a learner (demographic, questionnaire, etc) and have applied known methods and techniques of adapting the presentation and navigation (Chorfi et al, 2004) As explained in (Brusilovsky, 1996), two different classes of adaptation can be considered: adaptive presentation and adaptive navigation support Later, in (Brusilovsky, 2001), the taxonomy of adaptive hypermedia technologies was updated to add some extensions in relation with new technologies Then, the distinction between two modes of adaptive navigation support became a necessity, especially with the growth of recommender systems Automatic recommendation implies that the user profiles are created and eventually maintained dynamically by the system without explicit user information Examples include amazoncom's personalized recommendations and music recommenders like Mystrandcom in commercial systems (Mobasher 2006), smart recommenders in e-learning (Zaiane, 2002), etc In general, such systems differ in the input data, in user modeling strategies, and in prediction techniques Several approaches for automatic personalization have been reported in the literature, such as content-based or item-based filtering, collaborative filtering, rule-based filtering, and techniques relying on Web usage mining, etc (Nasraoui, 2005) Web recommender systems can be categorized depending on these approaches Content-based filtering (or item-based filtering) systems recommend items to a given user based on the correlation between the content of these items and the preferences of the user (Meteren et al, 2000) This means that the recommended items are considered to be similar to those seen and liked by the same user in the past Thus, there is no notion of a community of users, rather only one user profile is considered while making recommendations Classical examples of systems applying content based filtering approach include among other Personal webwatcher (Mladenic , 1996), syskill and webert (Pazzani et al, 1997), etc Collaborative filtering system recommends items that are liked by other users with similar interests Thus, the exploration of new items is assured by the fact that other similar user profiles are also considered Examples of such systems include GroupLens (Konstan et al, 1997) and (Sarwar et al, 1998) Hybrid recommender systems combine several recommendation strategies to provide better performance than either strategy alone Most hybrids work by combining several input data sources or several recommendation strategies …

Enhanced telephony computer user interface allowing user interaction and control of a telephone using a personal computer

Abstract: An enhanced telephony (ET) computer user interface that seamlessly integrates features of a personal computer (PC) and a telephone into a coherent user interface. The user is provided with a rich variety of functionality that leverages the fact that the PC has considerably more processing power and greater access to variety of data than the ordinary telephone. This processing power and data access is used to the user's advantage as the telephone's capabilities and functionality are greatly expanded. In general, the ET user interface includes a plurality of environments for the user to choose. These environments include a My Contacts environment, a communication preferences environment, and a Call History environment. Each of these environments contains certain available processes and features for controlling and managing telephones.

Towards Time-Dependant Recommendation based on Implicit Feedback

Abstract: Context-aware recommender systems (CARS) aim at improving users’ satisfaction by tailoring recommendations to each particular context. In this work we propose a contextual pre-ltering technique based on implicit user feedback. We introduce a new context-aware recommendation approach called user micro-proling . We split each single user prole into several possibly overlapping sub-proles, each representing users in particular contexts. The predictions are done using these micro-proles instead of a single user model. The users’ taste can depend on the exact partition of the contextual variable. The identication of a meaningful partition of the users’ prole and its evaluation is a non-trivial task, especially when using implicit feedback and a continuous contextual domain. We propose an o-line evaluation procedure for CARS in these conditions and evaluate our approach on a time-aware music recommendation sytem.

Mediation of User Models: for Enhanced Personalization in Recommender Systems

Abstract: Provision of personalized recommendations to users requires accurate modeling of their interests and needs. This work proposes a general framework and specific methodologies for enhancing the accuracy of user modeling in recommender systems by importing and integrating data collected by other recommender systems. Such a process is defined as user models mediation. The work discusses the details of such a generic user modeling mediation framework. It provides a generic user modeling data representation model, demonstrates its compatibility with existing recommendation techniques, and discusses the general steps of the mediation. Specifically, four major types of mediation are presented: cross-user, cross-item, cross-context, and cross-representation. Finally, the work reports the application of the mediation framework and illustrates it with practical mediation scenarios. Evaluations of these scenarios demonstrate the potential benefits of user modeling data mediation, as in certain conditions it allows improving the quality of the recommendations provided to the users.

User Interfaces for All: Concepts, Methods, and Tools

Abstract: Contents: G. Salvendy, Foreword. C. Stephanidis, Preface. Part I:Introduction. C. Stephanidis, User Interfaces for All: New Perspectives Into Human-Computer Interaction. Part II:Dimensions. D. Benyon, A. Crerar, S. Wilkinson, Individual Differences and Inclusive Design. A. Marcus, International and Intercultural User Interfaces. M. Maybury, Intelligent User Interfaces for All. L. Bass, Interaction Technologies Beyond the Desktop. P.L. Emiliani, Special Needs and Enabling Technologies: An Evolving Approach to Accessibility. G.C. Vanderheiden, S.L. Henry, Everyone Interfaces. Part III:Design. M. Wilson, Theory and Practice From Cognitive Science. T. Winograd, From Programming Environments to Environments for Designing. L.J. Bannon, V. Kaptelinin, From Human-Computer Interaction to Computer-Mediated Activity. M. Pieper, Sociological Issues in HCI Design. M. Antona, D. Akoumianakis, C. Stephanidis, "Generating" Design Spaces: An NLP Approach to HCI Design. Part IV:Software Technologies and Architectural Models. H. Ueda, The FRIEND21 Framework for Human Interface Architectures. J. Kay, User Modeling for Adaptation. A. Waern, K. Hook, Interface Agents: A New Interaction Metaphor and Its Application to Universal Accessibility. P. Korn, W. Walker, Accessibility in the Java(TM) Platform. Part V:Evaluation. D. Akoumianakis, D. Grammenos, C. Stephanidis, User Interface Adaptation: Evaluation Perspectives. N. Bevan, Quality in Use for All. Part VI:Unified User Interfaces. C. Stephanidis, The Concept of Unified User Interfaces. A. Savidis, C. Stephanidis, The Unified User Interface Software Architecture. A. Savidis, D. Akoumianakis, C. Stephanidis, The Unified User Interface Design Method. A. Savidis, C. Stephanidis, Development Requirements for Implementing Unified User Interfaces. D. Akoumianakis, C. Stephanidis, USE-IT: A Tool for Lexical Design Assistance. A. Savidis, C. Stephanidis, The I-GET UIMS for Unified User Interface Implementation. C. Stephanidis, A. Paramythis, M. Sfyrakis, A. Savidis, A Case Study in Unified User Interface Development: The AVANTI Web Browser. Part VII:Support Measures. D. Dardailler, J. Brewer, I. Jacobs, Making the Web Accessible. C. Stephanidis, D. Akoumianakis, N. Vernardakis, P.L. Emiliani, G. Vanderheiden, J. Ekberg, J. Ziegler, K-P. Faehnrich, A. Galetsas, S. Haataja, I. Iakovidis, E. Kemppainen, P. Jenkins, P. Korn, M. Maybury, H.J. Murphy, H. Ueda, Industrial Policy Issues. N. Vernardakis, D. Akoumianakis, C. Stephanidis, Economics and Management of Innovation. Part VIII:Looking to the Future. C. Stephanidis, G. Salvendy, D. Akoumianakis, N. Bevan, J. Brewer, P.L. Emiliani, A. Galetsas, S. Haataja, I. Iakovidis, J.A. Jacko, P. Jenkins, A.I. Karshmer, P. Korn, A. Marcus, H.J. Murphy, C. Stary, G. Vanderheiden, G. Weber, J. Ziegler, Toward an Information Society for All: An International Research and Development Agenda. C. Stephanidis, G. Salvendy, D. Akoumianakis, A. Arnold, N. Bevan, D. Dardailler, P.L. Emiliani, I. Iakovidis, P. Jenkins, A.I. Karshmer, P. Korn, A. Marcus, H.J. Murphy, C. Oppermann, C. Stary, H. Tamura, M. Tscheligi, H. Ueda, G. Weber, J. Ziegler, Toward an Information Society for All: HCI Challenges and R&D Recommendations.

I Like It... I Like It Not: Evaluating User Ratings Noise in Recommender Systems

Abstract: Recent growing interest in predicting and influencing consumer behavior has generated a parallel increase in research efforts on Recommender Systems. Many of the state-of-the-art Recommender Systems algorithms rely on obtaining user ratings in order to later predict unknown ratings. An underlying assumption in this approach is that the user ratings can be treated as ground truth of the user's taste. However, users are inconsistent in giving their feedback, thus introducing an unknown amount of noise that challenges the validity of this assumption. In this paper, we tackle the problem of analyzing and characterizing the noise in user feedback through ratings of movies. We present a user study aimed at quantifying the noise in user ratings that is due to inconsistencies. We measure RMSE values that range from 0.557 to 0.8156. We also analyze how factors such as item sorting and time of rating affect this noise.

Ontology based personalized route planning system using a multi-criteria decision making approach

Abstract: This study presents a generic ontology-based architecture using a multi-criteria decision making technique to design a personalized route planning system. The real world has become too complex to implement entirely within an information system such as geographic information system (GIS). A route planning technique is an essential geo-related decision support tool, especially in intelligent transportation systems (ITS). In ubiquitous GIS environments, personalization can be accomplished through a user's preferences stored on mobile appliances. In this manner, personalized and user-centric route planning services using semantic technologies and ontologies perceive user and context models to satisfy user demands and predict their requirements. In the past few years, several studies have been carried out regarding personalized services. However, the existing route finding algorithms suffer from a number of major difficulties, mainly owing to insufficient criteria modeling for a personalized system. Thus, the present study investigates how a user-centric route planning system can be implemented. In order to address this research issue, an ontology-based knowledge modelling technique using an analytic hierarchical process (AHP) is proposed. This technique can facilitate determination of the choice of criteria used for applying an impedance function in the route finding algorithm. From another perspective, AHP explicitly deals with a hierarchy structure and is essentially a theory of measurement and decision making methodology used for combining or synthesizing quantitative as well as qualitative criteria. User-centric results on real data illustrate the strengths of the present approach. It is anticipated that this new technique can be applied to develop new graph algorithms based on semantic web technology and can be used with new semantic graph structures.

Real-Time Game Adaptation for Optimizing Player Satisfaction

Abstract: A methodology for optimizing player satisfaction in games on the "playware" physical interactive platform is demonstrated in this paper. Previously constructed artificial neural network user models, reported in the literature, map individual playing characteristics to reported entertainment preferences for augmented-reality game players. An adaptive mechanism then adjusts controllable game parameters in real time in order to improve the entertainment value of the game for the player. The basic approach presented here applies gradient ascent to the user model to suggest the direction of parameter adjustment that leads toward games of higher entertainment value. A simple rule set exploits the derivative information to adjust specific game parameters to augment the entertainment value. Those adjustments take place frequently during the game with interadjustment intervals that maintain the user model's accuracy. Performance of the adaptation mechanism is evaluated using a game survey experiment. Results indicate the efficacy and robustness of the mechanism in adapting the game according to a user's individual playing features and enhancing the gameplay experience. The limitations and the use of the methodology as an effective adaptive mechanism for entertainment capture and augmentation are discussed.

System and method for streamlining user interaction with electronic content

Abstract: Various aspects and embodiments are directed to a graphical user interface that organizes interface elements into views of computer content for presentation to a user. Different views of are used to provide an interface that is responsive to configurations of the device and responsive to activity being performed by the user. Aspects include permitting the user to transition the device from one configuration to another during its use, for example from easel to laptop modes. Further the elements that comprise the graphical user interface are configured to present a summarized view of available actions and content, in order to simplify user interaction. The different views present different organizations of the interface elements and in some example display only certain ones of the modes of content in order to reduce the number of options a user must navigate to accomplish an objective. According to another aspect, methods and systems for streamlining user interaction with computer content are provided. In some embodiments, streamlining includes pre-configuring a user device based on received information. Other embodiments include presenting consistent visual representations used to navigated to views that present computer content.

SemFeel: a user interface with semantic tactile feedback for mobile touch-screen devices

Abstract: One of the challenges with using mobile touch-screen devices is that they do not provide tactile feedback to the user. Thus, the user is required to look at the screen to interact with these devices. In this paper, we present SemFeel, a tactile feedback system which informs the user about the presence of an object where she touches on the screen and can offer additional semantic information about that item. Through multiple vibration motors that we attached to the backside of a mobile touch-screen device, SemFeel can generate different patterns of vibration, such as ones that flow from right to left or from top to bottom, to help the user interact with a mobile device. Through two user studies, we show that users can distinguish ten different patterns, including linear patterns and a circular pattern, at approximately 90% accuracy, and that SemFeel supports accurate eyes-free interactions.

Method and apparatus for customizing a user interface menu

Abstract: Methods and apparatus enable a mobile device to suggest available applications or features in which a user may be interested to the user based upon the user's past and current mobile device usage patterns. The mobile device may monitor the specific application/features used and their frequency of use. The mobile device may determine other available applications/features that the user may be interested in using based upon the frequency of use of applications or features and information which indicates a likelihood of user interest in one application or feature based upon usage of another application or feature. Applications or features determined to be potentially of interest to the user may be presented to the user in the form of suggestions to be added to the user interface menu so that the user can elect to accept or rejection the suggestion to modify the menu.

Estimation of behavioral user state based on eye gaze and head pose--application in an e-learning environment

Abstract: Most e-learning environments which utilize user feedback or profiles, collect such information based on questionnaires, resulting very often in incomplete answers, and sometimes deliberate misleading input. In this work, we present a mechanism which compiles feedback related to the behavioral state of the user (e.g. level of interest) in the context of reading an electronic document; this is achieved using a non-intrusive scheme, which uses a simple web camera to detect and track the head, eye and hand movements and provides an estimation of the level of interest and engagement with the use of a neuro-fuzzy network initialized from evidence from the idea of Theory of Mind and trained from expert-annotated data. The user does not need to interact with the proposed system, and can act as if she was not monitored at all. The proposed scheme is tested in an e-learning environment, in order to adapt the presentation of the content to the user profile and current behavioral state. Experiments show that the proposed system detects reading- and attention-related user states very effectively, in a testbed where children's reading performance is tracked.

Design space for driver-based automotive user interfaces

Abstract: Over the last 100 years it has become much easier to operate a car. However in recent years the number of functions a user can control while driving has greatly increased. Infotainment, entertainment and comfort systems as well as driver assistance contribute to this trend. Interaction with these systems plays an important role, as on one hand this can improve the user experience while driving but on the other hand it may distract from the primary task of driving. User interfaces in cars differ regarding the number of input and output devices and their placement in the car to a great extent. In this paper, we introduce a first design space for driver-based automotive user interfaces that allows a comprehensive description of input and output devices in a car with regard to placement and modality. This design space is intended to provide a basis for analyzing and discussing different user interface arrangements in cars, to compare alternative user interface setups, and to identify new opportunities for interaction and placement of controls. We present a graphical representation of the design space and discuss its usage in detail based on several examples. To assess the completeness of the proposed design space we used it to classify and compare user interfaces from more than 100 cars shown at IAA2007, cars from the BMW museum, and from the A2Mac1 image database.

Latent Dirichlet Allocation with Topic-in-Set Knowledge

Abstract: Latent Dirichlet Allocation is an unsupervised graphical model which can discover latent topics in unlabeled data. We propose a mechanism for adding partial supervision, called topic-in-set knowledge, to latent topic modeling. This type of supervision can be used to encourage the recovery of topics which are more relevant to user modeling goals than the topics which would be recovered otherwise. Preliminary experiments on text datasets are presented to demonstrate the potential effectiveness of this method.

Adaptive Human Computer Interface (AAHCI)

Abstract: An Adaptive Human-Computer Interface (AAHCI) allows an electronic system to automatically monitor and learn from normal in-use behavior exhibited by a human user via responses generated by the supported input devices and to adjust output to the supported output devices accordingly. This Auto-Learning process is different than computer-directed training sessions and takes place as the user begins to use the device for the first time and with repeated use over time. The purpose of AHCI is to provide a user experience that is tailored to the skills, preferences, deficiencies and other personal attributes of the user automatically via machine-learned processes. This in turn provides an improved user experience that is more productive and cost efficient and that can automatically optimize itself over time with repeated use.

Structured annotations for 2D-to-3D modeling

Abstract: We present a system for 3D modeling of free-form surfaces from 2D sketches Our system frees users to create 2D sketches from arbitrary angles using their preferred tool, which may include pencil and paper A 3D model is created by placing primitives and annotations on the 2D image Our primitives are based on commonly used sketching conventions and allow users to maintain a single view of the model This eliminates the frequent view changes inherent to existing 3D modeling tools, both traditional and sketch-based, and enables users to match input to the 2D guide image Our annotations---same-lengths and angles, alignment, mirror symmetry, and connection curves---allow the user to communicate higher-level semantic information; through them our system builds a consistent model even in cases where the original image is inconsistent We present the results of a user study comparing our approach to a conventional "sketch-rotate-sketch" workflow

Combining Unsupervised and Supervised Classification to Build User Models for Exploratory Learning Environments

Abstract: In this paper, we present a data-based user modeling framework that uses both unsupervised and supervised classification to build student models for exploratory learning environments. We apply the framework to build student models for two different learning environments and using two different data sources (logged interface and eye-tracking data). Despite limitations due to the size of our datasets, we provide initial evidence that the framework can automatically identify meaningful student interaction behaviors and can be used to build user models for the online classification of new student behaviors online. We also show framework transferability across applications and data types.

The Hidden Agenda User Simulation Model

Abstract: A key advantage of taking a statistical approach to spoken dialogue systems is the ability to formalise dialogue policy design as a stochastic optimization problem. However, since dialogue policies are learnt by interactively exploring alternative dialogue paths, conventional static dialogue corpora cannot be used directly for training and instead, a user simulator is commonly used. This paper describes a novel statistical user model based on a compact stack-like state representation called a user agenda which allows state transitions to be modeled as sequences of push- and pop-operations and elegantly encodes the dialogue history from a user's point of view. An expectation-maximisation based algorithm is presented which models the observable user output in terms of a sequence of hidden states and thereby allows the model to be trained on a corpus of minimally annotated data. Experimental results with a real-world dialogue system demonstrate that the trained user model can be successfully used to optimise a dialogue policy which outperforms a hand-crafted baseline in terms of task completion rates and user satisfaction scores.

Identifying the Goal, User model and Conditions of Recommender Systems for Formal and Informal Learning

Abstract: This article addresses open questions of the discussions in the first SIRTEL workshop at the EC-TEL conference 2007. It argues why personal recommender systems have to be adjusted to the specific characteristics of learning in Learning Networks. Personal recommender systems strongly depend on the context or domain they operate in, and it is often not possible to take one recommender system with a specific purpose from one context and transfer it to another context or domain. The article describes a number of distinctive differences for personalised recommendation to learners when compared to recommendations for consumers. Similarities and differences for informal and formal learning are discussed and used to define the recommendation goal that recommender systems in informal learning networks have to address. The article further suggests an evaluation approach for recommender systems in Learning Networks.

User Interactive Reinsurance Risk Analysis Application

Abstract: A User Interactive Insurance Risk Analysis Application that is web-based which combines user imported insurance portfolios and geospatial technology with output from user defined actions and simulations providing an analysis tool for dynamic management of catastrophic exposures. The user may configure every aspect of the risk analysis application and use a variety of tools and features to create actions and receive risk information of all layers used for analysis. Precise risk information is given to the user due to the many services and modules which make up the risk analysis application, including a loss calculation service and dynamic, multi-dimensional configurable risk data display. The user receives estimated loss values and may configure which pieces of data analysis the user would like to see and how that data is presented without having to drilldown through all the data in the insurance portfolio.

Adaptive Experience Engine for Serious Games

Abstract: Designing games that support knowledge and skill acquisition has become a promising frontier of education techniques, since games are able to capture the user concentration for long periods and can present users with realistic and compelling challenges. In this scenario, there is a need for scientific and engineering methods to build games not only as more realistic simulations of the physical world but as means to provide effective learning experiences. Abstracting state of the art serious games' (SGs) features, we propose a new design methodology for the sand box serious games (SBSGs) class, decoupling content from the delivery strategy during the gameplay. This methodology aims at making design more efficient and standardized in order to meet the growing demand for interactive learning. The methodology consists in modeling an SBSG as a hierarchy of tasks (e.g., missions) and specifies the requirements for a runtime scheduling policy that maximizes learning objectives in a full entertainment context. The policy is learned by an experience engine (EE) based on computational intelligence. In this approach, the domain-expert author focuses on the creation and semantic annotation of tasks. Tasks are put in a repository and can then be exploited by game designers who define the expected learning curve and other requirements about education and entertainment for the game. The task sequencing that aims at matching such specifications with the real user profile is then presented to the EE. The EE can operate also in absence of the specification of the learning curve, continuously adapting the game flow without aiming at the achievement of target knowledge levels predefined by the author. We have implemented an EE module based on genetic computation and reinforcement learning (RL) atop of a state-of-the-art game engine. Test results show that the EE is able to define in real-time missions that meet the requirements expressed by the author.

Web-Based Tool for Collaborative, Social Learning

Abstract: A computerized-social network provides a community of users with features and tools facilitating an immersive, collaborative environment where users can learn a language or help others learn a language. One user (user A) can view another user's (user B) Web page or document and make suggestions or comments for selected content on that Web page. These suggestions are linked specifically to the selected content. User B can review the suggestions, and accept or reject the suggestions by user A and others.

A Theoretical Survey of User Interface Description Languages: Preliminary Results

Abstract: A user interface description language (UIDL) consists of a specification language that describes various aspects of a user interface under development. A comparative review of some selected user interface description languages is produced in order to analyze how they support the various stages of user interface development life cycle and development goals, such as support for multi-platform, device-independence, modality independence, and content delivery. There has been a long history and tradition to attempt capturing the essence of user interfaces at various levels of abstraction for different purposes, including those of development. The recent return of this effort today gains more attraction, along with the dissemination of XML markup languages, and gives birth to many proposals for various user interface description languages. Consequently, an in-depth analysis of the salient features that make these languages different from each other is desired in order to identify when and where they are appropriate for a specific purpose. The review is conducted based on a systematic analysis grid and some user interfaces implemented with these languages.

Sponsored application launcher suggestions

Abstract: A content delivery system that is based on user interest. The system includes client and server components. At each client, information about user interactions is collected and compared to information defining a set of user interest categories. In this way, the client can determine one or more interest categories in which the user fits. The client can then offer to the user content that is classified in accordance with the same set of user categories. The server can supply the client with the information defining user interest categories and a manifest of content options with associated categories. Additionally, the server can receive indications of when each content offering is presented to a user and when the user selects to acquire a content option. The server may also participate in offering the content. The content may be software applications.

Problem reporting system based on user interface interactions

Abstract: A system for reporting and analyzing problems encountered by computer users. The system includes a recording tool executing on a user computer to capture a sequence of user interactions in the context of a graphical user interface. When a problem or other stop event is encountered, the tool generates a report indicating user interactions leading to the stop event, including information such as the specific sequence of controls for specific programs accessed by the user. The report can be analyzed to identify a sequence of user interactions characteristic of a problem type, which in turn may be used to find a solution for a particular user's problem. The system may also include a server that receives and analyzes reports from multiple computer users to identify patterns of user interactions that characterize problem types. This information may be used for associating specific problems with future reports or to provide information to software developers for improving their products.