Personal software assistants that help users with tasks like finding information, scheduling calendars, or managing work-flow will require significant customization to each individual user. For example, an assistant that helps schedule a particular user’s calendar will have to know that user’s scheduling preferences. This paper explores the potential of machine learning methods to automatically create and maintain such customized knowledge for personal software assistants. We describe the design of one particular learning assistant: a calendar manager, called CAP (Calendar APprentice), that learns user scheduling preferences from experience. Results are summarized from approximately five user-years of experience, during which CAP has learned an evolving set of several thousand rules that characterize the scheduling preferences of its users. Based on this experience, we suggest that machine learning methods may play an important role in future personal software assistants.

Two branches of the trend towards "agents" that are gaining currency are interface agents, software that actively assists a user in operating an interactive interface, and autonomous agents, software that takes action without user intervention and operates concurrently, either while the user is idle or taking other actions. These two branches are related, but not identical, and are often lumped together under the single term "agent". Much agent work can be classified as either being an interface agent, but not autonomous, or as an autonomous agent, but not operating directly in the interface. We show why it is important to have agents that are both interface agents and autonomous agents. We explore some design principles for such agents, and illustrate these principles with a description of Letizia, an autonomous interface agent that makes real-time suggestions for Web pages that a user might be interested in browsing. Keywords Agents, interface agents, autonomous agents, Web, browsi...

Analysis of text data using intelligent information retrieval, machine learning, natural language processing or other related methods is becoming an important issue for the development of intelligent agents. There are two frequently used approaches to the development of intelligent agents using machine learning techniques: a content-based and a collaborative approach. In the first approach, the content (eg., text) plays an important role, while in the second approach, the existence of several knowledge sources (eg., several users) is required. We can say that the usage of machine learning techniques on text databases (usually referred to as textlearning) is an important part of the content-based approach. Examples are agents for locating information on World Wide Web and Usenet news filtering agents. There are different research questions important for the development of text-learning intelligent agents. We focus on three of them: what representation is used for documents, how is the h...