this is for holding javascript data
Dylan Freedman added PossibleHarmonicComparisons.tex
about 9 years ago
Commit id: 06a157e58ae44ffae6d65d6d048fe75d57f015dd
deletions | additions
diff --git a/PossibleHarmonicComparisons.tex b/PossibleHarmonicComparisons.tex
new file mode 100644
index 0000000..e7086c2
--- /dev/null
+++ b/PossibleHarmonicComparisons.tex
...
\section{Comparing Chord Progressions}
\subsection{Simple Global Comparison}
\item Iterate through each chord one by one and compare if it's similar or not. Perhaps use a distance matrix or something of the sort.
\subsection{N-Gram Comparison}
\item Using a sliding window of n-grams (tried 4), can we see the distribution of top scores and make some sort of claim based on that.
\item Advantages: Compares localized alignments, that is alignments in different positions.
\item Disadvantages: Computationally expensive. Does not work for arbitrary window lengths without another exponent in computation time required. Does not account for marginal errors well.
\subsection{Minimum Edit Distance}
\item Global comparison, assign cost to different edit distances and find minimum required "transformation" between two songs. Uses dynamic programming and can be done in quadratic time.
\item Disadvantages: global alignments only.
\subsection{Smith-Waterman}
\item Localized comparison, dynamic programming, find minimum number of required "transformations" and optimal localized slice. Dynamic programming
\item Works well with inexact data, can deal with common pitfalls of chord extraction.
\item Isolates similar substructures.
\item Does not penalize closely related chords, i.e. Em7 and Gmaj
\item Difficulty in extracting multiple "best" options but good at finding one top contender
\item Difficulty in comparing scores
