User modeling

About: User modeling is a research topic. Over the lifetime, 10701 publications have been published within this topic receiving 278012 citations.

Web search personalization with ontological user profiles

Abstract: Every user has a distinct background and a specific goal when searching for information on the Web. The goal of Web search personalization is to tailor search results to a particular user based on that user's interests and preferences. Effective personalization of information access involves two important challenges: accurately identifying the user context and organizing the information in such a way that matches the particular context. We present an approach to personalized search that involves building models of user context as ontological profiles by assigning implicitly derived interest scores to existing concepts in a domain ontology. A spreading activation algorithm is used to maintain the interest scores based on the user's ongoing behavior. Our experiments show that re-ranking the search results based on the interest scores and the semantic evidence in an ontological user profile is effective in presenting the most relevant results to the user.

A survey of statistical user simulation techniques for reinforcement-learning of dialogue management strategies

Abstract: Within the broad field of spoken dialogue systems, the application of machine-learning approaches to dialogue management strategy design is a rapidly growing research area. The main motivation is the hope of building systems that learn through trial-and-error interaction what constitutes a good dialogue strategy. Training of such systems could in theory be done using human users or using corpora of human–computer dialogue, but in practice the typically vast space of possible dialogue states and strategies cannot be explored without the use of automatic user simulation tools.This requirement for training statistical dialogue models has created an interesting new application area for predictive statistical user modelling and a variety of different techniques for simulating user behaviour have been presented in the literature ranging from simple Markov models to Bayesian networks. The development of reliable user simulation tools is critical to further progress on automatic dialogue management design but it holds many challenges, some of which have been encountered in other areas of current research on statistical user modelling, such as the problem of ‘concept drift’, the problem of combining content-based and collaboration-based modelling techniques, and user model evaluation. The latter topic is of particular interest, because simulation-based learning is currently one of the few applications of statistical user modelling that employs both direct ‘accuracy-based’ and indirect ‘utility-based’ evaluation techniques.In this paper, we briefly summarize the role of the dialogue manager in a spoken dialogue system, give a short introduction to reinforcement-learning of dialogue management strategies and review the literature on user modelling for simulation-based strategy learning. We further describe recent work on user model evaluation and discuss some of the current research issues in simulation-based learning from a user modelling perspective.

Temporal sequence learning and data reduction for anomaly detection

Abstract: The anomaly-detection problem can be formulated as one of learning to characterize the behaviors of an individual, system, or network in terms of temporal sequences of discrete data. We present an approach on the basis of instance-based learning (IBL) techniques. To cast the anomaly-detection task in an IBL framework, we employ an approach that transforms temporal sequences of discrete, unordered observations into a metric space via a similarity measure that encodes intra-attribute dependencies. Classification boundaries are selected from an a posteriori characterization of valid user behaviors, coupled with a domain heuristic. An empirical evaluation of the approach on user command data demonstrates that we can accurately differentiate the profiled user from alternative users when the available features encode sufficient information. Furthermore, we demonstrate that the system detects anomalous conditions quickly — an important quality for reducing potential damage by a malicious user. We present several techniques for reducing data storage requirements of the user profile, including instance-selection methods and clustering. An empirical evaluation shows that a new greedy clustering algorithm reduces the size of the user model by 70%, with only a small loss in accuracy.

Methods and systems for processing playlists

Abstract: Various systems and methods described above permit a user's content experience (e.g. music playing experience) to be monitored and for metadata describing this experience to be collected. This metadata can be dynamically updated as a user experiences media content and then used to impart to the user a personalized experience that is tailored to that specific user. A user's metadata can, in some instances, provided across a wide variety of computing devices on which the user may experience the media content, thus standardizing the user's personalized media experience. In addition, intelligent or “smart” playlists can be provided which, in some instances, can be dynamically and automatically updated to reflect current user experiences, thus providing a highly personalized and enjoyable content experience.

Semantic user interface

Abstract: A system and method that allows a user to use their everyday language or user defined words to operate a computer in a highly efficient way. In short, every word, letter, control character and symbol is potentially actionable. A computer user's productivity is dramatically increased by making available those functions that enable a user to produce most of his work through simple, language-based commands. The present invention provides an intuitive interface, referred to as a semantic user interface (SUI), that enhances the operation of the current standard window-based interface in a manner that is simple, richer and natural. By leveraging all of the richness and power inherent in a user's language, the present invention provides an important tool that allows the personal computer to operate in a manner that is much closer to our natural way of interacting. A user is allowed to enter “commands” in his everyday natural language in order to control the operations of the computer. All commands are language-based and user-defined. These commands can be entered from any context of the user's computer (e.g., any application or operating system workspace). The commands allows a user to launch applications and navigate within applications by using language rather than clicks from a pointing device such as a mouse. It also allows the replacement of keystrokes with stored words or keystrokes. The system also keeps a complete archive record of all the text content the user provides as input, regardless of which application program or operating system window the user is operating in at the time. The combined set of all user defined commands and the memory of all the input text that is stored in the archive constitutes the personality profile and is transportable from one computer to another.