From: Mart Lubbers <mart@martlubbers.net>
Date: Mon, 12 Sep 2016 18:30:19 +0000 (+0200)
Subject: update
X-Git-Url: https://git.martlubbers.net/?a=commitdiff_plain;h=8d38f5cb57c87fefced4d6c2c0dbc1cf491231ce;p=itlast1617.git

update
---

diff --git a/week3/eml.tex b/week3/eml.tex
index 5e13726..781f2fe 100644
--- a/week3/eml.tex
+++ b/week3/eml.tex
@@ -2,7 +2,6 @@
 
 \usepackage{amssymb}
 \usepackage{booktabs}
-\usepackage{float}
 \usepackage{geometry}
 
 \title{Exercise: Machine Learning}
@@ -12,7 +11,19 @@
 \begin{document}
 \maketitle
 \subsection*{Chapter 5: Machine Learning}
-\begin{table}[H]
+Table~\ref{t2} shows that there is some difference in classification when
+choosing different parameter sets. The results show that adding the focus word
+is very important knowledge since the percentage increases quite a bit.
+However, knowing only the focus word gives exceptionally low performance.
+
+Knowing the next words or the previous words gives some improvement but not a
+whole lot.
+
+Table~\ref{t1} shows that the method does not make a big difference.
+
+Using ten-fold cross validation decreases the percentage.
+		
+\begin{table}
 	\centering
 	\begin{tabular}{lll}
 		\toprule
@@ -24,38 +35,37 @@
 		J48 (10FCF) & $96.4352\%$ & $36.5122\%$\\
 		\bottomrule
 	\end{tabular}
-	\caption{Results for \texttt{P1D} and \texttt{FD}}
+	\caption{Results for \texttt{P1D} and \texttt{FD}\label{t1}}
 \end{table}
 
-\begin{table}[H]
+\begin{table}
 	\centering
 	\begin{tabular}{lllllll}
 		\toprule
 		\texttt{P2D} & \texttt{P1D} & \texttt{N2D} & \texttt{N1D} & \texttt{FW} & Correctly classified\\
 		\midrule
-		\checkmark{}	& \checkmark{}	& \checkmark{}	& \checkmark{}	& \checkmark{}	& $98.3225\%$\\
-		\checkmark{}	& \checkmark{}	& \checkmark{}	& \checkmark{}	&	& $97.3905\%$\\
-		\checkmark{}	& \checkmark{}	&	& \checkmark{}	& \checkmark{}	& $98.5555\%$\\
-		\checkmark{}	& \checkmark{}	&	& \checkmark{}	&	& $97.507\%$\\
-		\checkmark{}	& \checkmark{}	&	&	& \checkmark{}	& $98.0429\%$\\
-		\checkmark{}	& \checkmark{}	&	&	&	& $95.5732\%$\\
-
-			& \checkmark{}	& \checkmark{}	& \checkmark{}	& \checkmark{}	& $\%$\\
-			& \checkmark{}	& \checkmark{}	& \checkmark{}	&	& $\%$\\
-			& \checkmark{}	&	& \checkmark{}	& \checkmark{}	& $\%$\\
-			& \checkmark{}	&	& \checkmark{}	&	& $\%$\\
-			& \checkmark{}	&	&	& \checkmark{}	& $98.2992\%$\\
-			& \checkmark{}	&	&	&	& $96.6449\%$\\
-
-			&	& \checkmark{}	& \checkmark{}	& \checkmark{}	& $\%$\\
-			&	& \checkmark{}	& \checkmark{}	&	& $\%$\\
-			&	&	& \checkmark{}	& \checkmark{}	& $\%$\\
-			&	&	& \checkmark{}	&	& $\%$\\
-			&	&	&	& \checkmark{}	& $88.4436\%$\\
-
+		\checkmark{} & \checkmark{} & \checkmark{} & \checkmark{} & \checkmark{} & $98.3225\%$\\
+		\checkmark{} & \checkmark{} & \checkmark{} & \checkmark{} & & $97.3905\%$\\
+		\checkmark{} & \checkmark{} & & \checkmark{} & \checkmark{} & $98.5555\%$\\
+		\checkmark{} & \checkmark{} & & \checkmark{} & & $97.507\%$\\
+		\checkmark{} & \checkmark{} & & & \checkmark{} & $98.0429\%$\\
+		\checkmark{} & \checkmark{} & & & & $95.5732\%$\\
+		
+		& \checkmark{} & \checkmark{} & \checkmark{} & \checkmark{} & $98.6486\%$\\
+		& \checkmark{} & \checkmark{} & \checkmark{} & & $97.5769\%$\\
+		& \checkmark{} & & \checkmark{} & \checkmark{} & $98.5555\%$\\
+		& \checkmark{} & & \checkmark{} & & $97.973\%$\\
+		& \checkmark{} & & & \checkmark{} & $98.2992\%$\\
+		& \checkmark{} & & & & $96.6449\%$\\
+		
+		& & \checkmark{} & \checkmark{} & \checkmark{} & $91.8919\%$\\
+		& & \checkmark{} & \checkmark{} & & $85.5079\%$\\
+		& & & \checkmark{} & \checkmark{} & $92.579\%$\\
+		& & & \checkmark{} & & $85.2516\%$\\
+		& & & & \checkmark{} & $88.4436\%$\\
 		\bottomrule
 	\end{tabular}
-	\caption{NaiveBayes}
+	\caption{NaiveBayes on all sensible combinations\label{t2}}
 \end{table}
 
 \subsection*{Chapter 6: Exercises}
@@ -65,10 +75,11 @@
 		the preceding $n$ states. However, it is claimed that the algorithm
 		takes into account the whole sequence. Explain in your own words (at
 		most $100$) how the probability is influenced by the rest of the
-		sequence, i.e.\ both the positions more than n back and the following
+		sequence, i.e.\ both the positions more than $n$ back and the following
 		positions.}
 
 	\item\emph{Explain in your own words (at most 50) how the EM algorithm
 		works. I don't mean the mathematics, but the underlying concept.}
+
 \end{itemize}
 \end{document}