\documentclass{article}
\begin{document}
\title{Research Proposal}
\author{AUTHOR NAME}
\maketitle
\newpage
\section{Introduction}
For my research project, I am mainly focusing on fuzzy logic. Fuzzy logic is interesting and important as it provides an alternative to boolean logic. Boolean logic basically states that things can either be true or false (have values of 0 or 1, nothing in between). This is the main system of logic we have been using in math, computer science, and other areas for hundreds of years. However, it is not always efficient. Say, for instance, you pour some water into a cup. Say the cup can hold 100 mL of water, and you poured in 30 mL. Is the cup full? It would be difficult to answer this question in boolean logic. There is some water in the cup, so it's not completely empty, but since it's not full to the very rim (and even then, you can't be too sure), you can't say it's full, either.
\section{Fuzzy Logic}
That's where fuzzy logic comes in. Fuzzy logic operates on degrees of truth; fuzzy truth represents membership in vaguely defined sets. So if you had two fuzzy sets, Empty and Full, then you could say about the aforementioned cup that it is 0.7 Empty and 0.3 Full (since 30 mL of the 100 mL contains water, which is 30%). This allows for such descriptions of the cup as “nearly empty” or “halfway full”, something boolean logic wouldn't allow. Boolean logic would only be applicable if the cup was 100% full or 100% empty, and since this is a real cup, we can't expect that to be useful to us in real world applications.
\section{Applications of FL}
There are many applications of fuzzy logic in the real world. It is mostly used for complex processes, when no simple mathematical model exists. However, it's not a good idea if an easily solvable and adequate mathematical model exists, or if the problem is simply not solvable. Some applications of fuzzy logic include:
\begin{itemize}
\item Control of robots
\item Camera aiming for sporting events
\item Temperature fluctuations in AC systems
\item Optimized planning of bus time-tables
\item Prediction of earthquakes
\item Cancer diagnosis
\item Handwriting recognition
\item Vacuum cleaner control
\item Washing machine control
\item Flight aid for helicopters
\end{itemize}
All these are things that depend on not just one simple variable, but a mixture of various measurements of varying degrees, all of which influence the decision made. For instance, fuzzy logic in washing machine control would sense the weight of the load and the amount of detergent, and then, without human input, would decide what setting to adjust itself to. The same goes for air conditioners; they would decide what setting to use by sensing the temperature, and then expressing it using fuzzy sets. It would be fairly cool, or really hot, or somewhat chilly, based on values from the sets Hot and Cold.
\section{Playlist Creation}
I'm sure most of us, at one point or another, have struggled with creating playlists for ourselves, or simply wished something could do it automatically. While there are playlist creators out there, most of them simply dump all the songs in a folder into a playlist, or they aren't very accurate. There are also sites that let you listen to songs and recommend new ones, such as Pandora (based on the Music Genome Project). From “About the Music Genome Project”:
\begin{quote}
Together we set out to capture the essence of music at the most fundamental level. We ended up assembling literally hundreds of musical attributes or "genes" into a very large Music Genome. Taken together these genes capture the unique and magical musical identity of a song - everything from melody, harmony and rhythm, to instrumentation, orchestration, arrangement, lyrics, and of course the rich world of singing and vocal harmony.
\end{quote}
One of my ideas was to create something like this. However, it wouldn't require the manual effort the creators of the MGP put into analyzing each song, as described here:
\begin{quote}
Since we started back in 2000, we've carefully listened to the songs of tens of thousands of different artists - ranging from popular to obscure - and analyzed the musical qualities of each song one attribute at a time. This work continues each and every day as we endeavor to include all the great new stuff coming out of studios, clubs and garages around the world.
\end{quote}
Instead, it would automatically analyze such features of song files (and individual settings of the player) as artist, album, genre, tempo, volume, number of times played, and average length of the song played (to see, perhaps, if the user just skips right past this song, or doesn't like the ending, which might be loud, in which case the program decides that the user doesn't like certain parts, and analyzes them and compares them with the other parts, to further help build a profile of the user's likes and dislikes).
If this idea was extended to a real-life mp3 player, with sufficient technology it could also measure the user's heart rate, body temperature, and walking/running speed. This would be useful for people who listen to music while they jog or do any kind of workout, as the program could adjust the type of song it plays based on what it thinks the user is doing (if it senses that the user is winding down a workout, for instance, it could switch to a slower, more calming song).
\section{Conclusion}
Since there are so many factors that would go into deciding what songs should go into the playlist the program creates, I think it would be best to use fuzzy logic to do this. I have not yet done extensive research as to what has been done with automatic playlist creation / song recommendation and fuzzy logic, but I have found something similar: http://www.wipo.int/pctdb/fr/ia.jsp?ia=US2005%2F016752&IA=US2005%2F016752&DISPLAY=CLAIMS
I'm also not sure how to approach this, but I expect that I'd have to become familiar with languages other than Java, such as Python.
Overall, I think I have a pretty solid (if somewhat general) idea of what I want to do, just not how exactly I'm going to do it.
\end{document}