Marc Torrens

Recommender systems have been part of the Internet for almost two decades. Dozens of vendors have built recommendation technologies and taken them to market in two waves, roughly aligning with the web 1.0 and 2.0 revolutions. Today recommender systems are found in a multitude of online services. They have been developed using a variety of techniques and user interfaces. They have been nurtured with millions of users’ explicit and implicit preferences (most often with their permission)… Mostrar más

Recommender systems have been part of the Internet for almost two decades. Dozens of vendors have built recommendation technologies and taken them to market in two waves, roughly aligning with the web 1.0 and 2.0 revolutions. Today recommender systems are found in a multitude of online services. They have been developed using a variety of techniques and user interfaces. They have been nurtured with millions of users’ explicit and implicit preferences (most often with their permission). Frequently they provide relevant recommendations that increase the revenue or user engagement of the online services that operate them. However, when we evaluate the current generation of recommender systems from the point of view of the “recommendee,” we find that most recommender systems serve the goals of the business instead of their users’ interests. Thus we believe that the big promise of recommender systems has yet to be fulfilled. We foresee a third wave of recommender systems that act directly on behalf of their users across a range of domains instead of acting as a sales assistant. We also predict that such new recommender systems will better deal with information overload, take advantage of contextual clues from mobile devices, and utilize the vast information and computation stores available through cloud-computing services to maximize users’ long-term goals. Mostrar menos

When searching for multi-attribute services or products, understanding and representing user’s preferences is a crucial task. However, many computer tools do not afford users to adequately focus on fundamental decision objectives, reveal hidden preferences, revise conflicting preferences, or explicitly reason about tradeoffs with competing decision goals. As a result, users often fail to find the best solution.
From building decision support systems for various application domains, we have… Mostrar más

When searching for multi-attribute services or products, understanding and representing user’s preferences is a crucial task. However, many computer tools do not afford users to adequately focus on fundamental decision objectives, reveal hidden preferences, revise conflicting preferences, or explicitly reason about tradeoffs with competing decision goals. As a result, users often fail to find the best solution.
From building decision support systems for various application domains, we have observed some common areas of design pitfalls, which could lead to undesirable user behaviors and ineffective use of decision systems. By incorporating findings from behavior decision theory, we have identified and accumulated a set of principles for avoiding these design pitfalls: 1) provide a flexible order and choice in preference elicitation so that users can focus on fundamental objectives, 2) include appropriate information in a decision context to guide users in revealing hidden preferences, and 3) make tradeoff attributes explicit to facilitate preference revision and flexible decision making. We describe these principles and the corresponding interface affordances, and discuss concrete scenarios where they have been applied and tested. Mostrar menos

Nowadays, music fans are beginning to massively use mobile digital music players and dedicated software to organize and play large collections of music. In this context, users deal with huge music libraries containing thousands of tracks. Such a huge volume of music easily overwhelms users when selecting the music to listen or when organizing their collections. Music player software with visualizations based on textual lists and organizing features such as smart playlists are not really enough… Mostrar más

Nowadays, music fans are beginning to massively use mobile digital music players and dedicated software to organize and play large collections of music. In this context, users deal with huge music libraries containing thousands of tracks. Such a huge volume of music easily overwhelms users when selecting the music to listen or when organizing their collections. Music player software with visualizations based on textual lists and organizing features such as smart playlists are not really enough for helping users to efficiently manage their libraries. Thus, we propose new graphical visualizations and their associated features to allow users to better organize their personal music libraries and therefore also to ease selection later on. Mostrar menos

To find products in online environments, people increasingly rely on computerized search tools. The performance of such tools depends crucially on an accurate model of their users' preferences. Obtaining such models requires an adequate interaction model and system guidance.

We consider automated decision aids that help users select the best solution from a large set of options. For such tools to successfully accomplish their task, eliciting and representing users' decision preferences is a crucial task. It is usually too complex to get a complete and accurate model of their preferences, especially regarding the trade-offs between different criteria.We consider decision aid tools where users specify their preferences qualitatively: they are only able to state the… Mostrar más

We consider automated decision aids that help users select the best solution from a large set of options. For such tools to successfully accomplish their task, eliciting and representing users' decision preferences is a crucial task. It is usually too complex to get a complete and accurate model of their preferences, especially regarding the trade-offs between different criteria.We consider decision aid tools where users specify their preferences qualitatively: they are only able to state the criteria they consider, but not the precise numerical utility functions. For each criterion, the tool provides a standardized numerical function that is fixed and identical for all users and used to compare solutions. To compensate for the imprecision of this qualitative model, we let the user choose among a displayed set of possibilities rather than a single optimal solution. We consider the probability of finding the most preferred solution as a function of the number of displayed possibilities and the number of preferences. We present a probabilistic analysis, empirical validation on randomly generated configuration problems and a commercial application. We provide mathematical principles for the design of the selection mechanism, guaranteeing that users are able to find the target solution. Mostrar menos

Travel has many situations where context-aware computing can bring important benefits. In this paper, we describe an approach for integrating context-aware computing to a mobile travel assistant. Travel plans, generated using reality [2], are enriched within compact and powerful structures, called User Task Models. These structures are transferred to a mobile device enabling the support for the traveler during his trip.

Travel has many situations where context-aware computing can bring important benefits: pointing out notorious delays or bad weather during the planning phase, allowing the user to replan for handling unexpected situations, or suggesting flight alternatives to avoid strikes.This paper briefly describes an approach for integrating context-aware computing to a mobile travel assistant. We show how context-aware techniques can ease user interaction within mobile device applications. The presented… Mostrar más

Travel has many situations where context-aware computing can bring important benefits: pointing out notorious delays or bad weather during the planning phase, allowing the user to replan for handling unexpected situations, or suggesting flight alternatives to avoid strikes.This paper briefly describes an approach for integrating context-aware computing to a mobile travel assistant. We show how context-aware techniques can ease user interaction within mobile device applications. The presented ideas are illustrated with a scenario using a first version of a working prototype called Pocket reality. Mostrar menos

In many practical scenarios, users are faced with the problem of choosing the most preferred outcome from a large set of possibilities. As people are unable to sift through them manually, decisions support systems are often used to automatically find the optimal solution. A crucial requirement for such a system is to have an accurate model of the user's preferences.Studies have shown that people are usually unable to accurately state their preferences up front, but are greatly helped by seeing… Mostrar más

In many practical scenarios, users are faced with the problem of choosing the most preferred outcome from a large set of possibilities. As people are unable to sift through them manually, decisions support systems are often used to automatically find the optimal solution. A crucial requirement for such a system is to have an accurate model of the user's preferences.Studies have shown that people are usually unable to accurately state their preferences up front, but are greatly helped by seeing examples of actual solutions. Thus, several researchers have proposed preference elicitation strategies based on example critiquing. The essential design question in example critiquing is what examples to show users in order to best help them locate their most preferred solution.In this paper, we analyze this question based on two requirements. The first is that it must stimulate the user to express further preferences by showing the range of alternatives available. The second is that the examples that are shown must contain the solution that the user would consider optimal if the currently expressed preference model was complete so that he select it as a final solution. Mostrar menos

Many information systems are used in a problem solving context. Examples are travel planning systems, catalogs in electronic com- merce, or agenda planning systems. They can be made more use- ful by integrating problem-solving capabilities into the information systems. This poses the challenge of scalability: when hundreds of users access a server at the same time, it is important to avoid excessive computational load.
In this paper, we present an approach, called reality, that allows to… Mostrar más

Many information systems are used in a problem solving context. Examples are travel planning systems, catalogs in electronic com- merce, or agenda planning systems. They can be made more use- ful by integrating problem-solving capabilities into the information systems. This poses the challenge of scalability: when hundreds of users access a server at the same time, it is important to avoid excessive computational load.
In this paper, we present an approach, called reality, that allows to significantly extend the reach of electronic commerce in travel. Our application addresses in particular the challenge of modeling customers’ personal preferences and providing solutions that are tailored to just those preferences. In contrast to existing technology, which allow to optimize only a small and predefined set of prefer- ences, our tool allows a wide variety that can accurately model the preferences of different customers. Mostrar menos

The world today is full of information systems which make huge quantities of information available. This incredible amount of information is clearly overwhelming Internet end- users. As a consequence, intelligent tools to identify worthwhile information are needed, in order to fully assist people in finding the right information. Moreover, most systems are ultimately used, not just to provide information, but also to solve problems.
Encouraged by the growing popular success of Internet and… Mostrar más

The world today is full of information systems which make huge quantities of information available. This incredible amount of information is clearly overwhelming Internet end- users. As a consequence, intelligent tools to identify worthwhile information are needed, in order to fully assist people in finding the right information. Moreover, most systems are ultimately used, not just to provide information, but also to solve problems.
Encouraged by the growing popular success of Internet and the enormous business potential of electronic commerce, e-catalogs have been consolidated as one of the most relevant types of information systems. Nearly all currently available electronic catalogs are offering tools for extracting product information based on key-attribute filtering methods. The most advanced electronic catalogs are implemented as recommender systems using collaborative filtering techniques.
This dissertation focuses on strategies for coping with the difficulty of building intelli- gent catalogs which fully support the user in his purchase decision-making process, while maintaining the scalability of the whole system. The contributions of this thesis lie on a mixed-initiative system which is inspired by observations on traditional commerce ac- tivities. Such a conversational model consists basically of a dialog between the customer and the system, where the user criticizes proposed products and the catalog suggests new products accordingly.
Constraint satisfaction techniques are analyzed in order to provide a uniform frame- work for modeling electronic catalogs for configurable products. Within the same frame- work, user preferences and optimization constraints are also easily modeled. Searching strategies for proposing the adequate products according to criteria are described in de- tail. Mostrar menos

We consider constraint satisfaction problems where so- lutions must be optimized according to multiple crite- ria. When the relative importance of different criteria cannot be quantified, there is no single optimal solution, but a possibly very large set of Pareto-optimal solutions. Computing this set completely is in general very costly and often infeasible in practical applications.
We consider several methods that apply algorithms for soft CSP to this problem. We report on experiments… Mostrar más

We consider constraint satisfaction problems where so- lutions must be optimized according to multiple crite- ria. When the relative importance of different criteria cannot be quantified, there is no single optimal solution, but a possibly very large set of Pareto-optimal solutions. Computing this set completely is in general very costly and often infeasible in practical applications.
We consider several methods that apply algorithms for soft CSP to this problem. We report on experiments, both on random and real problems, that show that such algorithms can compute surprisingly good approxima- tions of the Pareto-optimal set. We also derive variants that further improve the performance. Mostrar menos

Many information systems are used in a problem solving context. Examples are travel planning systems, catalogs in electronic commerce, or agenda planning systems. They can be made more useful by integrating problem-solving capabilities into the information systems. This poses the challenge of scaleability: when hundreds of users access a server at the same time, it is important to avoid excessive computational load.
We present the concept of SmartClients: lightweight problem-solving agents… Mostrar más

Many information systems are used in a problem solving context. Examples are travel planning systems, catalogs in electronic commerce, or agenda planning systems. They can be made more useful by integrating problem-solving capabilities into the information systems. This poses the challenge of scaleability: when hundreds of users access a server at the same time, it is important to avoid excessive computational load.
We present the concept of SmartClients: lightweight problem-solving agents based on constraint satisfaction which can carry out the computation- and communication-intensive tasks on the user's computer. We present an example of an air travel planning system based on this technology. Mostrar menos

Many information systems are used in a problem solving context. Examples are travel planning systems, catalogs in electronic commerce, or agenda planning systems. They can be made more useful by integrating problem-solving capabilities into the information systems. This poses the challenge of scalability: when hundreds of users access a server at the same time, it is important to avoid excessive computational load.
We present the concept of smart clients: lightweight problem-solving agents… Mostrar más

Many information systems are used in a problem solving context. Examples are travel planning systems, catalogs in electronic commerce, or agenda planning systems. They can be made more useful by integrating problem-solving capabilities into the information systems. This poses the challenge of scalability: when hundreds of users access a server at the same time, it is important to avoid excessive computational load.
We present the concept of smart clients: lightweight problem-solving agents based on constraint satisfaction which can carry out the computation - and communication-intensive tasks on the user’s computer. We present an example of an air travel planning system based on this technology. Mostrar menos

When shall I fly to New York? Which airline should I choose? How are these related to which airport I arrive at, to how I might travel into the city and to where I choose to stay? Many current and potential applications of agents involve reasoning and communicating about multiple interrelated choices.To date however, most proposals for communication in agent systems have provided little or no direct support for the type of communication required by these applications. To address this need, this… Mostrar más

When shall I fly to New York? Which airline should I choose? How are these related to which airport I arrive at, to how I might travel into the city and to where I choose to stay? Many current and potential applications of agents involve reasoning and communicating about multiple interrelated choices.To date however, most proposals for communication in agent systems have provided little or no direct support for the type of communication required by these applications. To address this need, this paper describes the Constraint Choice Language (CCL) - an agent Content Language designed to support agent problem solving by providing explicit representations of choices and choice problems. Mostrar menos

In this work we address the problem of arranging meetings for several participants taking into con- sideration constraints for personal agendas and transportation schedules. We have implemented a multi-agent recommender system that solves the problem.
Building such applications implies to consider two main issues: collecting information from dif- ferent sources on the Internet, and solving the problem itself. We show that multi-agent systems that use constraint satisfaction for modelling and… Mostrar más

In this work we address the problem of arranging meetings for several participants taking into con- sideration constraints for personal agendas and transportation schedules. We have implemented a multi-agent recommender system that solves the problem.
Building such applications implies to consider two main issues: collecting information from dif- ferent sources on the Internet, and solving the problem itself. We show that multi-agent systems that use constraint satisfaction for modelling and solving problems can be very suitable for this kind of systems. Mostrar menos

This document gives the specification of version 2.01 of the Constraint Choice Language (CCL). CCL is designed as a language to be used for agent communication which directly supports the expression of choice and choice problems in the content of agent messages. This version of CCL has been adopted by FIPA as a standard FIPA-compliant content language named FIPA-CCL.

Manyinformation systems are used in a problem solving context. Examples are travel planning systems, catalogs in electronic commerce or agenda planning systems. They can be made more useful by integrating problem-solving capabilities into the information systems. This poses the challenge of scaleability: when hundreds of users access a server at the same time, it is important to avoid excessive computational load. We present the concept of smart clients: lightweight problem-solving agents based… Mostrar más

Manyinformation systems are used in a problem solving context. Examples are travel planning systems, catalogs in electronic commerce or agenda planning systems. They can be made more useful by integrating problem-solving capabilities into the information systems. This poses the challenge of scaleability: when hundreds of users access a server at the same time, it is important to avoid excessive computational load. We present the concept of smart clients: lightweight problem-solving agents based on constraint satisfaction which can carry out the computation -and communication- intensive tasks on the user’s computer. We present an example of an air travel planning system based on this technology. Mostrar menos

We address in this paper the adaptation of a case when a complete constraint model of the underlying problem is given. The idea is to apply methods from constraint based reasoning that allows the detection of "similar" solutions, which can be used to adapt a selected case to a new situation. We consider applications like configuration wherea complete constraint model is available. For these applications, it is often advantageous to use a case-based approach because constructing a solution from… Mostrar más

We address in this paper the adaptation of a case when a complete constraint model of the underlying problem is given. The idea is to apply methods from constraint based reasoning that allows the detection of "similar" solutions, which can be used to adapt a selected case to a new situation. We consider applications like configuration wherea complete constraint model is available. For these applications, it is often advantageous to use a case-based approach because constructing a solution from scratch requires often more interaction than taking a complete solution (case) and adapt it to fit the current requirements. Mostrar menos

Distributed problem solving on the web is becoming more and more important. Client-server architectures are often confronted with server overload. The pro- cess of browsing a large number of alternative solu- tions is particularly tedious. In this paper, we present a methodology for distributing the computation between server and client. The ides is to formalise the problem as a constraint satisfaction problem (CSP). This formalisation supports a natural decomposition of the task into two… Mostrar más

Distributed problem solving on the web is becoming more and more important. Client-server architectures are often confronted with server overload. The pro- cess of browsing a large number of alternative solu- tions is particularly tedious. In this paper, we present a methodology for distributing the computation between server and client. The ides is to formalise the problem as a constraint satisfaction problem (CSP). This formalisation supports a natural decomposition of the task into two subtasks: generation of the CSP by the server from its database, and generating and browsing the solutions on the client. In this way, the browsing process runs locally and can be very fast, while the server is only accessed once during the pro. ceas. We provide the Java Constraint library (JCL) for implementing the agent that solves the CSP on the client. We illustrate the concept on the example of planning air travel. Mostrar menos