Aves, Joshua Marl C Sept.

10,2019
COM 03
Fundamentals of Fuzzy Logic and Neural Networks using Python
The seminar kicked off by introducing Python’s background and history. It was
emphasized to us how advantageous it is to use python compared to other programming
languages. Advantages such as convenience when it comes to understanding the code
and having better functionality while having fewer lines of code put Python above complex
computer languages because it also makes it accessible to beginners. Then, we
proceeded to exploring python’s syntax and basic commands. Having experienced coding
other languages, most of the basic commands and functions (i.e. for loops, if else
statements, operators, while statement) had similar functionality as with other languages.
However, when I used python’s syntax, it seemed more readable and understandable to
a newbie’s eye and even with the experienced ones like myself. Variables were easier to
identify and name and functionalities became more pronounced.

An unfamiliar concept was introduced to us which was a very common and trendy
application of python in today’s technological world which was fuzzy logic. Fuzzy logic is
an approach to computing based on "degrees of truth" rather than the usual "true or false"
(1 or 0) Boolean logic on which the modern computer is based. Fuzzy logic was structured
to look at the world in imprecise terms, in much the same way that our brain takes in
information. The human brain can reason with uncertainties, vagueness, and judgments.
Computers can only manipulate precise valuations. Fuzzy logic is an attempt to combine
the two techniques. It was a whole new concept to me and I was amazed at how defining
and quantifying the world’s “uncertainties” could precisely render the necessary action to
fit the program’s functionality. We were taught how to use math to compartmentalize
values. For example, 5’6 when inserted into a basic logic metric could render you either
tall or short. However, when you plug in this value to a metric with fuzzy logic as its
structure, it would output 63% tall and 37% short depending on how you define your
values. This was furthered by assigning a necessary action to a set of predetermined
values to precisely address its need.

Then, another new concept of programming was introduced which was the artificial
neural networks. Artificial neural networks aim to mimic the real neural networks in our
brain in the sense that artificial neural networks figure out how our brain processes inputs
and gives off a certain output. This is very ideal to use when you don’t know how to code
the process but you know what your input and desired output is. There are said to be 3
layers: the input, output and the hidden layers where all the processing happens. This is
where the “neurons” are trained to adjust the weights of the multiple inputs until they
produce the desired output by the user.