Personalized information delivery: an analysis of information filtering methods

Abstract

With the increasing availability of information in electronic form, it becomes more important and feasible to have automatic methods to filter information.Research organizations generate large amounts of information, which can include departmental and technical memoranda, announcements of meetings and conferences, and minutes from meetings.This volume of information makes it difficult to keep employees apprised of all relevant work.Furthermore, only a small fraction of the available information will actually be relevant to any particular employee within an organization that covers a variety of areas.Thus, there is the problem of determining what information is of interest to the employee, while minimizing the amount of search through irrelevant information.This research tested several information-retrieval methods for filtering technical memos.Filtering of information is not a new concept, nor is it one that is limited to electronic documents.When we read standard paper texts, information filtering occurs.We only buy certain magazines, since other magazines may contain information that is redundant or irrelevant to our interests.In this way, we are filtering out some of the large amount of information to which we have access.Within any particular magazine, we also choose articles that appear relevant to our interests.Thus, when people are engaged in any sort of acquisition of information, they continually filter information.With the advent of electronic presentation of information, some of that filtering need no longer be done by us, but could be done automatically by the system that presents the information.