August 19, 2002
National Center of Sciences, Tokyo, Japan
Actually, in these 10 years, several techniques for data mining have been
developed, however, they can only show current or past tendencies.
Today, we are in the very changeable society. Therefore, we need to find
or to be shown a new event/situation that can be conceived either as
an opportunity or as a risk.
For all of us, it is very important to predict such an
event like the end of economic bubble (Japan, in 1990) or other
economic panics. Since these sorts of events can be thought of as rare
or novel events or exceptions. The conventional data mining techniques usually
ignore such exceptions. For example, the very famous Black-Scholes
equation (actually, this is not data mining, but its concept is like
that of induction or data mining) has no power for such situations.
Chance Discovery is such a research to study methodologies and theories to
show a new event/situation that can be conceived either as an
opportunity or as a risk.
It tries to deal with complex events in the real world. For example,
it will deal with
earth quake prediction, foretelling booms, risk management,
prediction of unpredictable change of stock price, etc.
For this purpose, Chance Discovery needs various techniques like
techniques to discover relationship between events, to suggest missing
events, knowledge in economy, knowledge in sociology, knowledge in
risk management, etc.
Therefore, now, it is very important to discuss ``Chance Discovery'' not
only from an AI's viewpoint but also from various viewpoints like economical
viewpoint, sociological viewpoint, and so on.
Also, this workshop will be an introduction to the following session
(KES2002)
and symposium (AAAI 2002 fall).
Chance Discovery is the discovery of chance, rather than discovery by chance. A ``chance" here means a new event/situation that can be conceived either as an opportunity or as a risk. The ``discovery" of chances is of crucial importance since it may have a significant impact on human decision making. Desirable effects of opportunities should be actively promoted, whereas preventive measures should be taken in the case of discovered risks. In other words, chance discovery aims to provide means for inventing or surviving the future, rather than predicting the future.
This workshop will discuss several problems in Chance Discovery. As
shown above, Chance Discovery is a research to study how to discover
rare or novel events causing potentially significant situation. Although
the event itself could not be significant.
Indeed, some data mining techniques can be applied to Chance
Discovery. However, they are not sufficient. Since, usually,
conventional data mining shows average events. Our main target is to
study how to discover rare or novel events. They are not average
matters but exceptions.
This workshop intends to discuss how to discover and suggest events
causing significant but hidden events.
In fact, we will deal with events in the real world, therefore, we
need to have knowledge about movement in society, behaviour of people,
as well as computational methods.
Therefore,
we would like to discuss from computational, cognitive, sociological,
economical and psychological viewpoints.
Topics to be discussed (will not be restricted to):
The papers can be submitted to: