Views of Activity Monitoring

New developments in activity monitoring are receiving attention as the
volume of documents stored in databases has exploded. The concurrent
increase in information mined from databases has accelerated demand for
new approaches for organizing, summarizing and reporting results in a
manner convenient for the end-user. In some applications, data is being
generated at such a tremendous pace and the data sets are so massive
that accessing the data more than once may be prohibitively expensive.
In other applications, the discovery of inter- and intra-stream
correlations in a large collection of streams is the challenge. Quite
often the impetus for the development of new technologies is to reduce
costs and errors associated with human labor.
This report surveys methods for monitoring the contents and trends
in massive and dynamic data sets as well as data streams: new topic
detection and tracking, early problem detection, anomaly and change point
identification, sentiment and preference mining for targeted marketing,
and data stream mining. In some instances, activity monitoring technologies,
such as Web surfing pattern analysis and collaborative filtering, have
run amiss, as defiant, rogue programmers and scammers create and
disseminate malicious and invasive, covertly self-installing software.
We survey the development of malware and its impact beyond the
technical domain. This report begins with a summary of work in
information retrieval and traditional data mining that is closely related
to the historical development of modern activity monitoring.
Economic, sociological, and legal perspectives and issues will be
noted when relevant to the adoption, dissemination, and future development
of workable technologies.

By: Mei Kobayashi

Published in: RT0625 in 2007

This Research Report is not available electronically. Please request a copy from the contact listed below. IBM employees should contact ITIRC for a copy.

Questions about this service can be mailed to .