Showing papers on "User modeling published in 2010"

A Tutorial on Bayesian Optimization of Expensive Cost Functions, with Application to Active User Modeling and Hierarchical Reinforcement Learning

Abstract: We present a tutorial on Bayesian optimization, a method of finding the maximum of expensive cost functions. Bayesian optimization employs the Bayesian technique of setting a prior over the objective function and combining it with evidence to get a posterior function. This permits a utility-based selection of the next observation to make on the objective function, which must take into account both exploration (sampling from areas of high uncertainty) and exploitation (sampling areas likely to offer improvement over the current best observation). We also present two detailed extensions of Bayesian optimization, with experiments—active user modelling with preferences, and hierarchical reinforcement learning—and a discussion of the pros and cons of Bayesian optimization based on our experiences.

Personalized news recommendation based on click behavior

Abstract: Online news reading has become very popular as the web provides access to news articles from millions of sources around the world. A key challenge of news websites is to help users find the articles that are interesting to read. In this paper, we present our research on developing personalized news recommendation system in Google News. For users who are logged in and have explicitly enabled web history, the recommendation system builds profiles of users' news interests based on their past click behavior. To understand how users' news interests change over time, we first conducted a large-scale analysis of anonymized Google News users click logs. Based on the log analysis, we developed a Bayesian framework for predicting users' current news interests from the activities of that particular user and the news trends demonstrated in the activity of all users. We combine the content-based recommendation mechanism which uses learned user profiles with an existing collaborative filtering mechanism to generate personalized news recommendations. The hybrid recommender system was deployed in Google News. Experiments on the live traffic of Google News website demonstrated that the hybrid method improves the quality of news recommendation and increases traffic to the site.

Beyond DCG: user behavior as a predictor of a successful search

Abstract: Web search engines are traditionally evaluated in terms of the relevance of web pages to individual queries. However, relevance of web pages does not tell the complete picture, since an individual query may represent only a piece of the user's information need and users may have different information needs underlying the same queries. In this work, we address the problem of predicting user search goal success by modeling user behavior. We show empirically that user behavior alone can give an accurate picture of the success of the user's web search goals, without considering the relevance of the documents displayed. In fact, our experiments show that models using user behavior are more predictive of goal success than those using document relevance. We build novel sequence models incorporating time distributions for this task and our experiments show that the sequence and time distribution models are more accurate than static models based on user behavior, or predictions based on document relevance.

Scholarly paper recommendation via user's recent research interests

Abstract: We examine the effect of modeling a researcher's past works in recommending scholarly papers to the researcher. Our hypothesis is that an author's published works constitute a clean signal of the latent interests of a researcher. A key part of our model is to enhance the profile derived directly from past works with information coming from the past works' referenced papers as well as papers that cite the work. In our experiments, we differentiate between junior researchers that have only published one paper and senior researchers that have multiple publications. We show that filtering these sources of information is advantageous -- when we additionally prune noisy citations, referenced papers and publication history, we achieve statistically significant higher levels of recommendation accuracy.

Living Lab Research Landscape: From User Centred Design and User Experience towards User Cocreation

Abstract: New paradigms, such as Open Innovation (Chesbrough, 2003) and Web 2.0 (O'Reilly, 2004) as well as Living Labs operating as a User Centred Open Innovation Ecosystem (Pallot, 2009), promote a more proactive role of users in the RD Sanders, 2008) and later introduced in the domain of Living Lab research (Mulder & Stappers, 2009). It also discusses the links with existing theories such as Social Capital Theory (Nahapiet and Ghoshal, 1998) and Social Cognitive Theory (Bandura, 1986) as well as Socio-Emotional Intelligence Theory (Goleman, 1998). It also explores the creation of User Group Experience concept for bringing the socio-emotional perspective (Norman, 1995; 1998; 2004; 207; Goleman, 1998) into User Experience (Fleming, 1998) that appears too much focusing on individual users and usability.

Predicting searcher frustration

Abstract: When search engine users have trouble finding information, they may become frustrated, possibly resulting in a bad experience (even if they are ultimately successful). In a user study in which participants were given difficult information seeking tasks, half of all queries submitted resulted in some degree of self-reported frustration. A third of all successful tasks involved at least one instance of frustration. By modeling searcher frustration, search engines can predict the current state of user frustration and decide when to intervene with alternative search strategies to prevent the user from becoming more frustrated, giving up, or switching to another search engine. We present several models to predict frustration using features extracted from query logs and physical sensors. We are able to predict frustration with a mean average precision of 65% from the physical sensors, and 87% from the query log features.

Evaluating User Experience in Games: Concepts and Methods

Abstract: User Experience has become a major research area in human-computer interaction. The area of game design and development has been focusing on user experience evaluation for the last 20 years, although a clear definition of user experience is still to be established. The contributors to this volume explore concepts that enhance the overall user experience in games such as fun, playability, flow, immersion and many others. Presenting an overview of current practice from academia and industry in game development, the book shows a variety of methods that can be used to evaluate user experience in games, not only during game-play but also before and after the game play. Evaluating User Experiences in Games: Presents a broad range of user experience evaluation methods and concepts; Provides insights on when to apply the various user experience evaluation methods in the development cycle and shows how methods can be also applied to a more general HCI context; Includes new research on evaluating user experience during game play and after; and social play; Describes new evaluation methods; Details methods that are also applicable for exertion games or tabletop games. This comprehensive book will be welcomed by researchers and practitioners in the field.

Quality of Experience from user and network perspectives

Abstract: The impact of network performance on user experience is important to know, as it determines the success or failure of a service. Unfortunately, it is very difficult to assess it in real-time on an operational network. Monitoring of network-level performance criteria is easier and more usual. But the problem is then to correlate these network-level Quality of Service (QoS) to the Quality of Experience (QoE) perceived by the users. Efforts have been done in the previous years to map user behaviour to traffic characteristics on the network to QoS. However, being able to successfully relate these traffic characteristics to user satisfaction is not a simple task and still requires further investigations. In this work, we try to associate on one side the correlations between various traffic characteristics measured on an operational network and on the other side the user experience tested on an experimental platform. Our aim is to observe some pronounced trends regarding relationships between both types of results. More precisely, we want to validate how and to what extent the volumes of user sessions represent the level of user satisfaction. Along this way, we need to revise classical relationships between some of the network performance indicators such as loss, download time and throughput in order to strengthen the understanding of this impact on each other and on user satisfaction. This preliminary study is based on the application web.

Augmented reality for testing and training of human performance

Abstract: A system for continuously monitoring a user's motion and for continuously providing realtime visual physical performance information to the user while the user is moving to enable the user to detect physical performance constructs that expose the user to increased risk of injury or that reduce the user's physical performance. The system includes multiple passive controllers 100 A-F for measuring the user's motion, a computing device 102 for communicating with wearable display glasses 120 and the passive controllers 100 A-F to provide realtime physical performance feedback to the user. The computing device 102 also transmits physical performance constructs to the wearable display glasses 120 to enable the user to determine if his or her movement can cause injury or reduce physical performance.

The demographics of web search

Abstract: How does the web search behavior of "rich" and "poor" people differ? Do men and women tend to click on difffferent results for the same query? What are some queries almost exclusively issued by African Americans? These are some of the questions we address in this study. Our research combines three data sources: the query log of a major US-based web search engine, profile information provided by 28 million of its users (birth year, gender and ZIP code), and US-census information including detailed demographic information aggregated at the level of ZIP code. Through this combination we can annotate each query with, e.g. the average per-capita income in the ZIP code it originated from. Though conceptually simple, this combination immediately creates a powerful user modeling tool. The main contributions of this work are the following. First, we provide a demographic description of a large sample of search engine users in the US and show that it agrees well with the distribution of the US population. Second, we describe how different segments of the population differ in their search behavior, e.g. with respect to the queries they formulate or the URLs they click. Third, we explore applications of our methodology to improve web search relevance and to provide better query suggestions. These results enable a wide range of applications including improving web search and advertising where, for instance, targeted advertisements for "family vacations" could be adapted to the (expected) income.

Motion-based interactive shopping environment

Abstract: An on-screen shopping application which reacts to a human target user's motions to provide a shopping experience to the user is provided. A tracking system captures user motions and executes a shopping application allowing a user to manipulate an on-screen representation the user. The on-screen representation has a likeness of the user or another individual and movements of the user in the on-screen interface allows the user to interact with virtual articles that represent real-world articles. User movements which are recognized as article manipulation or transaction control gestures are translated into commands for the shopping application.

Web user behavioral profiling for user identification

Abstract: In this paper, we propose a simple, yet powerful approach to profile users' web browsing behavior for the purpose of user identification. The importance of being able to identify users can be significant given a wide variety of applications in electronic commerce, such as product recommendation, personalized advertising, etc. We create user profiles capturing the strength of users' behavioral patterns, which can be used to identify users. Our experiments indicate that these profiles can be more accurate at identifying users than decision trees when sufficient web activities are observed, and can achieve higher efficiency than Support Vector Machines. The comparisons demonstrate that profile-based methods for user identification provide a viable and simple alternative to this problem.

Suggesting connections to a user based on an expected value of the suggestion to the social networking system

Abstract: To suggest new connections to a user of a social networking system, the system generates a set of candidate users to whom the user has not already formed a connection. The system determines the likelihood that the user will connect to each candidate user if suggested to do so, and it also computes the value to the social networking system if the user does connect to the candidate user. Then, the system computes an expected value score for each candidate user based on the corresponding likelihood and the value. The candidate users are ranked and the suggestions are provided to the user based on the candidate users' expected value scores. The social networking system can suggest other actions to a user in addition to forming a new connection with other users.

Bayesian networks for student model engineering

Abstract: 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.

I tag, you tag: translating tags for advanced user models

Abstract: Collaborative tagging services (folksonomies) have been among the stars of the Web 2.0 era. They allow their users to label diverse resources with freely chosen keywords (tags). Our studies of two real-world folksonomies unveil that individual users develop highly personalized vocabularies of tags. While these meet individual needs and preferences, the considerable differences between personal tag vocabularies (personomies) impede services such as social search or customized tag recommendation. In this paper, we introduce a novel user-centric tag model that allows us to derive mappings between personal tag vocabularies and the corresponding folksonomies. Using these mappings, we can infer the meaning of user-assigned tags and can predict choices of tags a user may want to assign to new items. Furthermore, our translational approach helps in reducing common problems related to tag ambiguity, synonymous tags, or multilingualism. We evaluate the applicability of our method in tag recommendation and tag-based social search. Extensive experiments show that our translational model improves the prediction accuracy in both scenarios.

Editorial: Data mining for understanding user needs

Abstract: Data mining and data analysis have a long history in human-computer interaction, starting with early interests in tracking the users then trying to infer models of users for adaptive systems [Benyon and Murray 1993; Fischer 1993], to more recent interests in attentional user interfaces, notifier systems, and recommenders. Recommender systems have emerged as a research area meriting a conference series since 2007, while attentional UIs have been the subject of several special issues [Horvitz et al. 2003; McCrickard et al. 2003b]. The convergence of analytic techniques for establishing patterns and orders in large datasets—data mining—and using such analysis to improve the responsiveness, user fit, and functionality of interactive systems has not been explicitly synthesized even though it has been a persistent interest in HCI. This special issue is therefore timely in bringing the fields of data mining and HCI together, As technology has developed over the past few decades, vast amounts of data have been generated as a result of users’ interactions with a range of applications from e-commerce to social networking sites. Analyzing this data can help in understanding the users’ needs and evaluating the effectiveness of user interaction. In turn, this can be used to improve the interface and interaction design, determine more suitable content, and develop useful services targeted at individual users. Data mining, also known as knowledge discovery [Fayyad and Uthurusamy 1996], is the process of extracting valuable information from large amounts

Ontological technologies for user modelling

Abstract: This paper brings together research from two different fields – user modelling and web ontologies – in attempt to demonstrate how recent semantic trends in web development can be combined with the modern technologies of user modelling. Over the last several years, a number of user-adaptive systems have been exploiting ontologies for the purposes of semantics representation, automatic knowledge acquisition, domain and user model visualisation and creation of interoperable and reusable architectural solutions. Before discussing these projects, we first overview the underlying user modelling and ontological technologies. As an example of the project employing ontology-based user modelling, we present an experiment design for translation of overlay student models for relative domains by means of ontology mapping.

Expected browsing utility for web search evaluation

Abstract: Most information retrieval evaluation metrics are designed to measure the satisfaction of the user given the results returned by a search engine. In order to evaluate user satisfaction, most of these metrics have underlying user models, which aim at modeling how users interact with search engine results. Hence, the quality of an evaluation metric is a direct function of the quality of its underlying user model. This paper proposes EBU, a new evaluation metric that uses a sophisticated user model tuned by observations over many thousands of real search sessions. We compare EBU with a number of state of the art evaluation metrics and show that it is more correlated with real user behavior captured by clicks.

Pareto-improving congestion pricing and revenue refunding with multiple user classes

Abstract: This study investigates Pareto-improving congestion pricing and revenue refunding schemes in general transportation networks, which make every road user better off as compared with the situation without congestion pricing. We consider user heterogeneity in value of time (VOT) by adopting a multiclass user model with fixed origin–destination (OD) demands. We first prove that an OD and class-based Pareto-improving refunding scheme exists if and only if the total system monetary travel disutility is reduced. In view of the practical difficulty in identifying individual user’s VOT, we further investigate class-anonymous refunding schemes that give the same amount of refund to all user classes traveling between the same OD pair regardless of their VOTs. We establish a sufficient condition for the existence of such OD-specific but class-anonymous Pareto-improving refunding schemes, which needs information only on the average toll paid and average travel time for trips between each OD pair.

A study on user perception of personality-based recommender systems

Abstract: Our previous research indicates that using personality quizzes is a viable and promising way to build user profiles to recommend entertainment products Based on these findings, our current research further investigates the feasibility of using personality quizzes to build user profiles not only for an active user but also his or her friends We first propose a general method that infers users' music preferences in terms of their personalities Our in-depth user studies show that while active users perceive the recommended items to be more accurate for their friends, they enjoy more using personality quiz based recommenders for finding items for themselves Additionally, we explore if domain knowledge has an influence on users' perception of the system We found that novice users, who are less knowledgeable about music, generally appreciated more personality-based recommenders Finally, we propose some design issues for recommender systems using personality quizzes.

Providing information to users based on context

Abstract: Selecting and providing context information relevant to a user at a particular time and location. Input parameters such as a location and time are selected. Context information is obtained for the selected location and time based on the input parameters. Exemplary input parameters include a user activity history, user content such as calendar appointments, social networking data, and a state of a computing device of the user (e.g., as collected by sensors of the computing device). The computing device of the user presents the obtained context information to the user at the selected location and time.

Adjusting search results based on user social profiles

Abstract: Many contemporary computing scenarios involve the submission by a user of a search query to be applied to a data set (such as a set of web pages indexed by a web search engine.) Additionally, many users participate in social networking and have generated a social profile, including demographic information, interests, and associations with other users who also have social profiles. It may be advantageous to improve the presentation of search results from search queries by adjusting the search queries according to a prediction of user interest of the user in the topics associated with respective search results, based on the information in the social profile of the user. For example, search results relating to topics in which the user or the user's friends have expressed an interest within the social network may be presented before other search results, thereby improving the relevance of the search results to the user.

Improving the effectiveness of collaborative recommendation with ontology-based user profiles

Abstract: Collaborative recommendation is effective at representing a user's overall interests and tastes, and finding peer users that can provide good recommendations. However, it remains a challenge to make collaborative recommendation sensitive to a user's specific context and to the changing shape of user interests over time. Our approach to building context-sensitive collaborative recommendation is a hybrid one that incorporates semantic knowledge in the form of a domain ontology. User profiles are defined relative to the ontology, giving rise to an ontological user profile. In this paper, we describe how ontological user profiles are learned, incrementally updated, and used for collaborative recommendation. Using book rating data, we demonstrate that this recommendation algorithm offers improved coverage, diversity, personalization, and cold-start performance while at the same time enhancing recommendation accuracy.

System and method of selecting a relevant user for introduction to a user in an online environment

Abstract: Systems and methods of selecting a relevant user for introduction to a user in an online environment are disclosed. In one aspect, embodiments of the present disclosure include a method, which may be implemented on a system, of receiving a request to identify the relevant user to be introduced to the participating user in the social network, identifying a set of social history records that occurred among the multiple users in the social network, selecting, from the multiple users, the relevant user to be introduced to the participating user using the set of social history records. The relevant user that is selected is not a social connection of the participating user in the social network. One embodiment includes, determining whether the participating user is interested in meeting the relevant user in the social network based on an indication made by the participating user via interaction with the user device.

Enabling seamless access to a domain of an enterprise

Abstract: A user is allowed to access any of a number of domains associated with an enterprise using a credential for any one of the domains. An exemplary method includes steps of receiving, from a user and at a first domain of the enterprise, a user identification and a password; determining, at the first domain, whether the user identification is associated with the first domain; and upon determination that the user identification is not associated with the first domain, determining, at the first domain, whether the user identification is associated with a second domain of the enterprise. The user identification and the password are authenticated at the first domain, upon determination that the user identification is associated with the second domain. Upon successful authentication, the user is enabled to access the second domain of the enterprise. The user identification does not need to include a character directly reflecting a domain name.

The challenges in large-scale smartphone user studies

Abstract: We present preliminary results of a large-scale smartphone user study that examines how users interact with and consume energy on their personal mobile devices. Our dataset consists of over one millennium of user interaction traces from over 17300 BlackBerry users. Despite the scale and detail of the dataset, there are many research questions that it cannot answer; further user studies are therefore needed. We detail our insight into the major challenges in conducting a large-scale user study on BlackBerry devices.