15
Software tool for analysis of problems whose abstraction is based on graphs
Authors: Ing. Alasia Delgado Saab, Ing. Orielisa J. Carapaica, Dra. Lourdes M. Ortiz Sosa

Number of views: 218
There are many problems that can be so/ved efficient/
y if they were approached through the use of
graphs. This article deals with the development of a
software application that allows to portray information
and its' relationships graphical/y, thus simp/ifying
data ana/ysis and supporting the decision making
process during prob/ems resolution. This project has
been developed under a tour phase methodology:
1) Research and information analysis, 2) Determination
of the conceptual model for the software,
through the relationship between the established
base theories and practica/ necessities, 3) Definition
of the software requirements in place, and 4) Development
of the application. Among the main resu!ts
accomplished from the development of this project,
there are: a) A classification of the algorithms u sed in
Graph Theory considering the perspectiva of severa/
authors and experts in the subject, b) An outline of
some of the most used a/gorithms in six ditferent
areas of app!ication, e) A conceptual model for a
scalable software application and d) An application
that implements that conceptual model considering
two specific areas of application. Among the most
relevant conclusions obtained from this project, there
are: a) Some problems easily handled by graphs do
not have a deterministic so/ution, because of this,
they must be so/ved through the use of heuristics,
which requires /onger processing time accordíng
to the problem complexíty, b) The applícatíon to
be defined and developed should be sca!able to
allow for the íncorporation of functíona!itíes and
dítferent application areas, e) Prevíous research ís a
mandatory activity in the development of any innovative
applícation, d) Software development should
be an incremental process determinad by qualíty,
e) The applicatíon developed may be extended in tho short term for a greater scientometríc analysís
addíng much value to scientifíc research, as we/1 as
extended to íncorporate new reports and analysis
algorithms.