Rule-based information filtering systems maintain user profiles where the profile consists of a set of filtering rules expressing the user’s information filtering policy. Filtering rules may refer to various attributes of the data items subject to the filtering process. In personal rulebased filtering systems, each user has his/her own personal filtering rules. In stereotype rule-based filtering systems, a user is assigned to a group of similar users (his/her stereotype) from which he/she inherits the stereotype’s filtering profile. This study compares the effectiveness of the two alternative rule-based filtering methods: stereotype-based rules versus personal rules. We conducted a comparison between filtering effectiveness when using the personal rules or when using the stereotype-based rules. Although, intuitively, personal filtering rules seem to be more effective because each user has his own tailored rules, our comparative study reveals that stereotype filtering rules yield more effective results. We believe that this is because users find it difficult to evaluate their filtering preferences accurately. The results imply that by using a stereotype it is possible not only to overcome the problem of user effort required to generate a manual rule-based profile, but at the same time even provide a better initial user profile.

Information filtering systems retrieve documents from document streams according to their users' long-term information interests represented by so-called profiles. The Profile Editor proposed in this article allows the interactive, direct manipulative construction of profiles. It takes a set of ranked queries and compiles them into a single profile by cropping and re-ranking the queries' results. The approach of manual profile generation is expected to lead to two advantages: a) Profile generation is expected to be much faster than feedback-based automatic profile generation and b) users' confidence in their profiles should be higher because they are in control of their profiles. The Profile Editor is currently being implemented in the context of an Internet TV program guide, in which it will be evaluated during the next months.

Email is the “killer network application”. Email is ubiquitous and pervasive. In a relatively short timeframe, the Internet has become irrevocably and deeply entrenched in our modern society primarily due to the power of its communication substrate linking people and organizations around the globe. Much work on email technology has focused on making email easy to use, permitting a wide variety of information and information types to be conveniently, reliably, and efficiently sent throughout the Internet. However, the analysis of the vast storehouse of email content accumulated or produced by individual users has received relatively little attention other than for specific tasks such as spam and virus filtering. As one paper in the literature puts it, ”the state of the art is still a messy desktop” (Denning,

A significant amount of information is stored in computer systems today, but people are struggling to manage their documents such that the information is easily found. XML is a de-facto standard for content publishing and data exchange. The proliferation of XML documents has created new challenges and opportunities for managing document collections. Existing technologies for automatically organizing document collections are either imprecise or based on only simple criteria. Since XML documents are self describing, it is now possible to automatically categorize XML documents precisely, according to their content. With the availability of the standard XML query languages, e.g. XQuery, much more powerful folder technologies are now feasible. To address this new challenge and exploit this new opportunity, this paper proposes a new and powerful dynamic folder mechanism, called Hubble. Hubble fully exploits the rich data model and semantic information embedded in the XML documents to build folder hierarchies dynamically and to categorize XML collections precisely. Besides supporting basic folder operations, Hubble also provides advanced features such as multi-path navigation and folder traversal across multiple document collections. Our performance study shows that Hubble is both efficient and scalable. Thus, it is an ideal technology for automating the process of organizing and categorizing XML documents.

This paper describes Mona , an email system that provides an automatic hypertext representation of conversational context. Mona is novel in that conversation facilities are provided without

This paper develops a conceptual framework for text filtering practice and research, and reviews present practice in the field. Text filtering is an information seeking process in which documents are selected from a dynamic text stream to satisfy a relatively stable and specific information need. A model of the information seeking process is introduced and specialized to define text filtering. The historical development of text filtering is then reviewed and case studies of recent work are used to highlight important design characteristics of modern text filtering systems. User modeling techniques drawn from information retrieval, recommender systems, machine learning and other fields are described. The paper concludes with observations on the present state of the art and implications for future research on text filtering.

incompatibility. We introduce Mona, an operational email system embodying this automatic approach. Mona establishes conversation context independently of user actions through the use of heuristic inferencing techniques which draw on information in standard email communications. We also discuss the applicability of Mona's design considerations to other areas of CSCW research. The value of electronic mail as a medium for collaborative and coordinated work can be enhanced by relating messages to conversations. While some groupware systems have offered such facilities, their ability to assess conversational context is dependent on explicit user action and the use of specific systems by all collaborators. This paper describes Mona, a novel conversation based email platform. Mona provides a hypertext representation of conversational context without requiring any additional effort from the user or the use of specific email systems by other collaborators. Mona's lack of requirements and inde...

: Email is much more than an essential communication technology. It is a rich source of quality and up-to-date information, and users need advanced facilities to store, organize and retrieve information. In this paper, we discuss the main features of a query language specifically designed to retrieve information in electronic mail environments. The language enables users to locate information independently from message and/or folder inspection, through the definition of ad hoc queries for retrieving messages, folders, information about messages and/or folders, or any combination of these. It makes use of full text indexing and information retrieval techniques in order to efficiently handle semi-structured messages. The main features of the language are presented and illustrated, and implementation issues are discussed. Key-Words: query language, full text indexing and retrieval, object-oriented modeling, electronic mail 1. Introduction Electronic mail (email) has become an essential ...

This paper summarizes the main contributions of [FER97], in which mechanisms to aid users in the classification and retrieval of large volume of messages, considering the use of electronic mail as information systems, were proposed. The retrieval mechanism enables to locate and retrieve messages, as well as to obtain information about messages and classification structures, through the definition of ad hoc queries using a specific-purpose language. This language makes use of text information indexing and retrieval techniques. The classification mechanism is based on the concept of virtual folder, which allows messages to be logically related to one or more folders, providing additional facilities and more flexibility to message classification procedures. A particularly interesting type of virtual folder is the automatic folder, defined by a query that retrieves a set of messages meeting a specified criterion. Automatic folder is thus similar to the concept of view in database systems. ...

In this paper, we propose and experiment a probabilistic approach to document classification. We consider the problem of automatically assigning a new article to a Usenet newsgroup. To model a newsgroup, we build a probabilistic language model which is supposed to generate articles for this newsgroup. When a new article is presented, we use a Maximum A Posteriori rule to decide if the message was generated by this newsgroup or not. We evaluate this approach and compare it to a classification based on keywords. On these cases, the probabilistic approach gives better recall and precision indicators.