Intelligent information-sharing systems

Abstract

Management in the 1990s is an industry and governmental agency supported research program. Its aim is to develop a better understanding of the managerial issues of the 1990s and how to deal most effectively with them, particularly as these issues revolve around anticipated advances in Information Technology.Assisting the work of the Sloan School scholars with financial support and as working partners in research are: Much recent work on the Information Lens system has been done by Kum-Yew Lai, Ramana Rao, and David Rosenblitt. Kevin Crowston implemented and maintains the mail gateway between our system and the outside network world, Chee-Seng Chow helped implement the first version of the central mail sorter, and Constance Perin collaborated in some of the early studies of information sharing in organizations. We gratefully acknowledge all these contributions. SummaryAs it becomes both technically and economically feasible to send electronic messages and other documents to large numbers of possible recipients, the problem of deciding who should receive a particular piece of information will become increasingly important. This paper focuses on the application of techniques from artificial intelligence, user interface design, and organizational science to help people share interesting and relevant information without being inundated by the potentially vast amount of less useful information. Much of the work that has been done on computer-mediated communication systems, such as electronic mail, computer conferencing, and electronic bulletin boards, has focused on technical capabilities and standards for transporting and storing messages. In this paper, we propose to shift our focus to a more general problem that we will refer to as the information sharing problem. The information sharing problem involves disseminating information such that the information reaches those people to whom it is valuable and yet does not interfere with the information processing of people who will find no value in its contents.We argue that this problem will become increasingly important with advances in communication technology and we will suggest three fundamental approaches to its solution. Our characterization of these approaches will be illustrated with (1) examples from a series of informal studies that we have conducted on how people share information in organizations, and (2) a description of an intelligent information sharing system, called the Information Lens, that we have developed.It is already a common experience in mature computer-based messaging communities for people to feel flooded with large quantities of electronic "junk mail" [Den82; Pal84; Wil84; Hil85], and the increasing availability of inexpensive communication technology has the potential to overwhelm people with even more messages that are of little or no value to them. In current systems, people often adopt crude methods--such as removing themselves entirely from distribution lists that are of occasional interest--in order to avoid being inundated with informa...

Intelligent Information Personalization Leveraging Constraint Satisfaction and Association Rule Methods

“…Our strategy to solve the PCSP is explained using an exemplar user profile (in Table 3) and dataset (in Table 4). 2 acquisition, crude, nat-gas t 10 bop, cpi, gnp, jobs t 3 acquisition, gold, lead, silver, zinc t 11 jobs, trade t 4 gas t 12 gnp, jobs t 5 CPI, crude, fuel, gas, nat-gas t 13 acquisition t 6 fuel, gas t 14 fuel, gas t 7 crude, gas t 15 jobs, trade t 8 GNP, income, ipi, retail, trade…”

“…From an AI perspective, a variety of techniques have been employed for pursuing IP. Foltz used latent semantic indexing (LSI) to perform information personalization [4]; Mooney and Roy developed a book recommending system based on a Bayesian text classifier [5]; Malone et al built a rule-based system to filter e-mail messages [6]; Jennings and Higuchi helped users get better access to news service using neural-networks [7]; Desjardins and Godin use genetic algorithms for personalization [8]. However, pursuing IP as a constraint satisfaction problem is a novel approach.…”

An evolutionary approach to prototyping pedagogical agents: from simulation to integrated system

“…The most common application of interface agents is integration with existing systems. A range of agent-enabled applications exists, including email (Malone et al, 1987), graphic editors (Lieberman, 1998), design environments (Fischer, McCall & Mørch, 1989), spreadsheets (Krogsaeter, To achieve a synergistic effect with agents in learning environments they must make their presence known at the right time. If they fail to do so, they may be ignored or perceived as disturbing ongoing activities.…”

Dynamic Web content filtering based on user's knowledge

“…Cognitive filtering, also called the content-based filtering, 'characterizes the contents of the message and the information needs of potential message recipients and then using these representations to intelligently match messages to receivers' [2]. Research based systems, such as SIFT [3], and the commercial systems (e.g., DansGuardian, iProtectYou, Parental Filter, Symatec Web Security, and We-Blocker) usually use URL or keywords to characterize content, while the social filtering systems use personal organizational interrelationships of individuals in a community.…”