Tóbiás, Roland (MTA-ELTE KKRK)
Analysis of Spectroscopic Networks
Networks are widely
employed for the description of complex phenomena occuring
in nature, society, and communication. A considerable amount of our chemical knowledge
is based on high-resolution spectroscopy measurements, detecting transitions
among discrete energy le¬vels the chemical systems
possess according to quantum mechanics. A network-based ap¬proach
to handle large spectroscopic data¬sets has been developed during the last decade and applied
successfully to the complex measured spectra of several molecules, some of astrophy¬sical interest and some related to the greenhouse
effect. In this presentation we will introduce the
concept of spectroscopic networks (SN) and seek answers to the following ques¬tions:
(1) How many components do
(measured and computed) spectroscopic networks have and what are their
characteristics?
(2) How can network theory
improve the information systems storing high-resolution spectro-scopic
data used by modellers, engineers, and scientists?
(3) What is the
relationship between cycles of SNs and the law of energy conservation (LEC)?
(4) How to use the LEC
equations to
(a) determine the term
(energy) values of energy levels participating in a transition as
accurately as possible,
(b) detect transitions with
incorrect ex¬perimental wa¬venumbers
or assignments, and
(c) characterize/improve
the measurement uncertainties?
(5) What is the effect of
bridges on the term values of energy levels in SNs?
Beyond these problems, some
open questions will also be introduced:
(1) How to treat parallel
edges (coincident transitions) in the light of the LEC?
(2) How to construct a
cycle basis which is most sensitive to the presence of outliers?
(3) What is the fastest way to approximate this cycle basis?
The talk is held in Hungarian!
Az elõadás magyar nyelven lesz megtartva!
Date: Nov 21, Tuesday 4:15pm
Place: BME, Building „Q”, Room QBF13