--- a
+++ b/results/output_knn.txt
@@ -0,0 +1,1528 @@
+[Creating the subsets for the values]
+[Creating the subsets for the labels "1"-"0"]
+
+[Optimization of the hyper-parameter k start]
+
+[Training the kNN model (with k=1) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6264594 
+
+    Area under curve (Davis & Goadrich):
+     0.6264609 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6028206 (worst possible: -1; best possible: +1)
+When k=1, the MCC value is 0.6028206	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=2) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6169076 
+
+    Area under curve (Davis & Goadrich):
+     0.616909 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.5923581 (worst possible: -1; best possible: +1)
+When k=2, the MCC value is 0.5923581	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=3) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6830796 
+
+    Area under curve (Davis & Goadrich):
+     0.6830816 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6636884 (worst possible: -1; best possible: +1)
+When k=3, the MCC value is 0.6636884	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=4) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6616313 
+
+    Area under curve (Davis & Goadrich):
+     0.661633 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6385756 (worst possible: -1; best possible: +1)
+When k=4, the MCC value is 0.6385756	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=5) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6620504 
+
+    Area under curve (Davis & Goadrich):
+     0.6620522 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6413557 (worst possible: -1; best possible: +1)
+When k=5, the MCC value is 0.6413557	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=6) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6755082 
+
+    Area under curve (Davis & Goadrich):
+     0.6755101 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6569899 (worst possible: -1; best possible: +1)
+When k=6, the MCC value is 0.6569899	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=7) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6837836 
+
+    Area under curve (Davis & Goadrich):
+     0.6837855 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6655734 (worst possible: -1; best possible: +1)
+When k=7, the MCC value is 0.6655734	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=8) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6800778 
+
+    Area under curve (Davis & Goadrich):
+     0.6800797 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6621069 (worst possible: -1; best possible: +1)
+When k=8, the MCC value is 0.6621069	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=9) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6723174 
+
+    Area under curve (Davis & Goadrich):
+     0.6723192 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6527535 (worst possible: -1; best possible: +1)
+When k=9, the MCC value is 0.6527535	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=10) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6811248 
+
+    Area under curve (Davis & Goadrich):
+     0.6811267 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.660644 (worst possible: -1; best possible: +1)
+When k=10, the MCC value is 0.660644	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=11) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6818313 
+
+    Area under curve (Davis & Goadrich):
+     0.6818332 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6625213 (worst possible: -1; best possible: +1)
+When k=11, the MCC value is 0.6625213	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=12) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6823706 
+
+    Area under curve (Davis & Goadrich):
+     0.6823725 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6646654 (worst possible: -1; best possible: +1)
+When k=12, the MCC value is 0.6646654	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=13) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6832644 
+
+    Area under curve (Davis & Goadrich):
+     0.6832663 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6634545 (worst possible: -1; best possible: +1)
+When k=13, the MCC value is 0.6634545	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=14) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.677557 
+
+    Area under curve (Davis & Goadrich):
+     0.6775589 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6568918 (worst possible: -1; best possible: +1)
+When k=14, the MCC value is 0.6568918	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=15) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6834522 
+
+    Area under curve (Davis & Goadrich):
+     0.6834542 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6632248 (worst possible: -1; best possible: +1)
+When k=15, the MCC value is 0.6632248	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=16) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6765703 
+
+    Area under curve (Davis & Goadrich):
+     0.6765722 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6584004 (worst possible: -1; best possible: +1)
+When k=16, the MCC value is 0.6584004	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=17) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6905335 
+
+    Area under curve (Davis & Goadrich):
+     0.6905355 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6735008 (worst possible: -1; best possible: +1)
+When k=17, the MCC value is 0.6735008	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=18) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.687534 
+
+    Area under curve (Davis & Goadrich):
+     0.687536 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6690656 (worst possible: -1; best possible: +1)
+When k=18, the MCC value is 0.6690656	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=19) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6841468 
+
+    Area under curve (Davis & Goadrich):
+     0.6841488 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6650927 (worst possible: -1; best possible: +1)
+When k=19, the MCC value is 0.6650927	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=20) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6793496 
+
+    Area under curve (Davis & Goadrich):
+     0.6793515 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6601956 (worst possible: -1; best possible: +1)
+When k=20, the MCC value is 0.6601956	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=21) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.680415 
+
+    Area under curve (Davis & Goadrich):
+     0.6804169 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6616029 (worst possible: -1; best possible: +1)
+When k=21, the MCC value is 0.6616029	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=22) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6828979 
+
+    Area under curve (Davis & Goadrich):
+     0.6828998 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6639265 (worst possible: -1; best possible: +1)
+When k=22, the MCC value is 0.6639265	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=23) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6859101 
+
+    Area under curve (Davis & Goadrich):
+     0.6859121 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6683802 (worst possible: -1; best possible: +1)
+When k=23, the MCC value is 0.6683802	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=24) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6790152 
+
+    Area under curve (Davis & Goadrich):
+     0.6790171 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6606992 (worst possible: -1; best possible: +1)
+When k=24, the MCC value is 0.6606992	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=25) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6823706 
+
+    Area under curve (Davis & Goadrich):
+     0.6823725 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6646654 (worst possible: -1; best possible: +1)
+When k=25, the MCC value is 0.6646654	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=26) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6814796 
+
+    Area under curve (Davis & Goadrich):
+     0.6814815 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6630094 (worst possible: -1; best possible: +1)
+When k=26, the MCC value is 0.6630094	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=27) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6793496 
+
+    Area under curve (Davis & Goadrich):
+     0.6793515 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6601956 (worst possible: -1; best possible: +1)
+When k=27, the MCC value is 0.6601956	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=28) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6770507 
+
+    Area under curve (Davis & Goadrich):
+     0.6770526 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.657628 (worst possible: -1; best possible: +1)
+When k=28, the MCC value is 0.657628	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=29) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6807639 
+
+    Area under curve (Davis & Goadrich):
+     0.6807658 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6611152 (worst possible: -1; best possible: +1)
+When k=29, the MCC value is 0.6611152	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=30) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6732314 
+
+    Area under curve (Davis & Goadrich):
+     0.6732333 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6544314 (worst possible: -1; best possible: +1)
+When k=30, the MCC value is 0.6544314	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=31) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6700402 
+
+    Area under curve (Davis & Goadrich):
+     0.6700421 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6501895 (worst possible: -1; best possible: +1)
+When k=31, the MCC value is 0.6501895	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=32) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6700402 
+
+    Area under curve (Davis & Goadrich):
+     0.6700421 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6501895 (worst possible: -1; best possible: +1)
+When k=32, the MCC value is 0.6501895	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=33) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6755082 
+
+    Area under curve (Davis & Goadrich):
+     0.6755101 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6569899 (worst possible: -1; best possible: +1)
+When k=33, the MCC value is 0.6569899	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=34) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6781162 
+
+    Area under curve (Davis & Goadrich):
+     0.6781181 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6590371 (worst possible: -1; best possible: +1)
+When k=34, the MCC value is 0.6590371	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=35) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6688331 
+
+    Area under curve (Davis & Goadrich):
+     0.6688349 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6490422 (worst possible: -1; best possible: +1)
+When k=35, the MCC value is 0.6490422	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=36) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6680522 
+
+    Area under curve (Davis & Goadrich):
+     0.668054 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.647093 (worst possible: -1; best possible: +1)
+When k=36, the MCC value is 0.647093	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=37) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6706786 
+
+    Area under curve (Davis & Goadrich):
+     0.6706804 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6524217 (worst possible: -1; best possible: +1)
+When k=37, the MCC value is 0.6524217	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=38) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6685574 
+
+    Area under curve (Davis & Goadrich):
+     0.6685592 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6495905 (worst possible: -1; best possible: +1)
+When k=38, the MCC value is 0.6495905	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=39) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6672332 
+
+    Area under curve (Davis & Goadrich):
+     0.667235 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.648737 (worst possible: -1; best possible: +1)
+When k=39, the MCC value is 0.648737	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=40) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6684234 
+
+    Area under curve (Davis & Goadrich):
+     0.6684253 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6498704 (worst possible: -1; best possible: +1)
+When k=40, the MCC value is 0.6498704	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=41) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6681633 
+
+    Area under curve (Davis & Goadrich):
+     0.6681652 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6504415 (worst possible: -1; best possible: +1)
+When k=41, the MCC value is 0.6504415	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=42) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6604042 
+
+    Area under curve (Davis & Goadrich):
+     0.660406 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6413865 (worst possible: -1; best possible: +1)
+When k=42, the MCC value is 0.6413865	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=43) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6613514 
+
+    Area under curve (Davis & Goadrich):
+     0.6613532 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6431107 (worst possible: -1; best possible: +1)
+When k=43, the MCC value is 0.6431107	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=44) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6615746 
+
+    Area under curve (Davis & Goadrich):
+     0.6615763 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6425109 (worst possible: -1; best possible: +1)
+When k=44, the MCC value is 0.6425109	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=45) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6605155 
+
+    Area under curve (Davis & Goadrich):
+     0.6605173 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6410885 (worst possible: -1; best possible: +1)
+When k=45, the MCC value is 0.6410885	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=46) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6648667 
+
+    Area under curve (Davis & Goadrich):
+     0.6648685 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6464784 (worst possible: -1; best possible: +1)
+When k=46, the MCC value is 0.6464784	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=47) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6628686 
+
+    Area under curve (Davis & Goadrich):
+     0.6628704 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6433473 (worst possible: -1; best possible: +1)
+When k=47, the MCC value is 0.6433473	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=48) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6593459 
+
+    Area under curve (Davis & Goadrich):
+     0.6593476 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6399632 (worst possible: -1; best possible: +1)
+When k=48, the MCC value is 0.6399632	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=49) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6615746 
+
+    Area under curve (Davis & Goadrich):
+     0.6615763 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6425109 (worst possible: -1; best possible: +1)
+When k=49, the MCC value is 0.6425109	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=50) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6629903 
+
+    Area under curve (Davis & Goadrich):
+     0.6629921 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6430603 (worst possible: -1; best possible: +1)
+When k=50, the MCC value is 0.6430603	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=51) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6571225 
+
+    Area under curve (Davis & Goadrich):
+     0.6571242 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6374154 (worst possible: -1; best possible: +1)
+When k=51, the MCC value is 0.6374154	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=52) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6571225 
+
+    Area under curve (Davis & Goadrich):
+     0.6571242 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6374154 (worst possible: -1; best possible: +1)
+When k=52, the MCC value is 0.6374154	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=53) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6571225 
+
+    Area under curve (Davis & Goadrich):
+     0.6571242 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6374154 (worst possible: -1; best possible: +1)
+When k=53, the MCC value is 0.6374154	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=54) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6550037 
+
+    Area under curve (Davis & Goadrich):
+     0.6550054 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6345626 (worst possible: -1; best possible: +1)
+When k=54, the MCC value is 0.6345626	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=55) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6559627 
+
+    Area under curve (Davis & Goadrich):
+     0.6559644 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6362946 (worst possible: -1; best possible: +1)
+When k=55, the MCC value is 0.6362946	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=56) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.65453 
+
+    Area under curve (Davis & Goadrich):
+     0.6545317 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6361237 (worst possible: -1; best possible: +1)
+When k=56, the MCC value is 0.6361237	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=57) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.65778 
+
+    Area under curve (Davis & Goadrich):
+     0.6577817 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6400827 (worst possible: -1; best possible: +1)
+When k=57, the MCC value is 0.6400827	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=58) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6537498 
+
+    Area under curve (Davis & Goadrich):
+     0.6537515 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6337486 (worst possible: -1; best possible: +1)
+When k=58, the MCC value is 0.6337486	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=59) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6591333 
+
+    Area under curve (Davis & Goadrich):
+     0.659135 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.64057 (worst possible: -1; best possible: +1)
+When k=59, the MCC value is 0.64057	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=60) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6525102 
+
+    Area under curve (Davis & Goadrich):
+     0.6525119 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6329479 (worst possible: -1; best possible: +1)
+When k=60, the MCC value is 0.6329479	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=61) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6534768 
+
+    Area under curve (Davis & Goadrich):
+     0.6534785 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6346958 (worst possible: -1; best possible: +1)
+When k=61, the MCC value is 0.6346958	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=62) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6500728 
+
+    Area under curve (Davis & Goadrich):
+     0.6500745 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6313858 (worst possible: -1; best possible: +1)
+When k=62, the MCC value is 0.6313858	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=63) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6534768 
+
+    Area under curve (Davis & Goadrich):
+     0.6534785 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6346958 (worst possible: -1; best possible: +1)
+When k=63, the MCC value is 0.6346958	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=64) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6570203 
+
+    Area under curve (Davis & Goadrich):
+     0.657022 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6377206 (worst possible: -1; best possible: +1)
+When k=64, the MCC value is 0.6377206	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=65) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6533903 
+
+    Area under curve (Davis & Goadrich):
+     0.653392 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6350186 (worst possible: -1; best possible: +1)
+When k=65, the MCC value is 0.6350186	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=66) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6501507 
+
+    Area under curve (Davis & Goadrich):
+     0.6501524 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6310562 (worst possible: -1; best possible: +1)
+When k=66, the MCC value is 0.6310562	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=67) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6425675 
+
+    Area under curve (Davis & Goadrich):
+     0.6425691 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6220184 (worst possible: -1; best possible: +1)
+When k=67, the MCC value is 0.6220184	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=68) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6481208 
+
+    Area under curve (Davis & Goadrich):
+     0.6481225 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6278665 (worst possible: -1; best possible: +1)
+When k=68, the MCC value is 0.6278665	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=69) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6436853 
+
+    Area under curve (Davis & Goadrich):
+     0.643687 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6231183 (worst possible: -1; best possible: +1)
+When k=69, the MCC value is 0.6231183	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=70) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6535655 
+
+    Area under curve (Davis & Goadrich):
+     0.6535673 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6343766 (worst possible: -1; best possible: +1)
+When k=70, the MCC value is 0.6343766	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=71) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6524229 
+
+    Area under curve (Davis & Goadrich):
+     0.6524246 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6332672 (worst possible: -1; best possible: +1)
+When k=71, the MCC value is 0.6332672	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=72) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6514541 
+
+    Area under curve (Davis & Goadrich):
+     0.6514558 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6315184 (worst possible: -1; best possible: +1)
+When k=72, the MCC value is 0.6315184	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=73) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6491761 
+
+    Area under curve (Davis & Goadrich):
+     0.6491778 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6292987 (worst possible: -1; best possible: +1)
+When k=73, the MCC value is 0.6292987	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=74) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6493398 
+
+    Area under curve (Davis & Goadrich):
+     0.6493415 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6286567 (worst possible: -1; best possible: +1)
+When k=74, the MCC value is 0.6286567	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=75) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.645934 
+
+    Area under curve (Davis & Goadrich):
+     0.6459356 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6253258 (worst possible: -1; best possible: +1)
+When k=75, the MCC value is 0.6253258	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=76) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6492569 
+
+    Area under curve (Davis & Goadrich):
+     0.6492585 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6289759 (worst possible: -1; best possible: +1)
+When k=76, the MCC value is 0.6289759	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=77) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6516325 
+
+    Area under curve (Davis & Goadrich):
+     0.6516342 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6308905 (worst possible: -1; best possible: +1)
+When k=77, the MCC value is 0.6308905	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=78) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6504839 
+
+    Area under curve (Davis & Goadrich):
+     0.6504856 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6297723 (worst possible: -1; best possible: +1)
+When k=78, the MCC value is 0.6297723	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=79) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6481208 
+
+    Area under curve (Davis & Goadrich):
+     0.6481225 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6278665 (worst possible: -1; best possible: +1)
+When k=79, the MCC value is 0.6278665	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=80) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6438215 
+
+    Area under curve (Davis & Goadrich):
+     0.6438231 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6224554 (worst possible: -1; best possible: +1)
+When k=80, the MCC value is 0.6224554	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=81) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6460845 
+
+    Area under curve (Davis & Goadrich):
+     0.6460862 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6246768 (worst possible: -1; best possible: +1)
+When k=81, the MCC value is 0.6246768	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=82) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6450257 
+
+    Area under curve (Davis & Goadrich):
+     0.6450273 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6232419 (worst possible: -1; best possible: +1)
+When k=82, the MCC value is 0.6232419	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=83) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6405828 
+
+    Area under curve (Davis & Goadrich):
+     0.6405844 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6184729 (worst possible: -1; best possible: +1)
+When k=83, the MCC value is 0.6184729	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=84) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.64164 
+
+    Area under curve (Davis & Goadrich):
+     0.6416416 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6199112 (worst possible: -1; best possible: +1)
+When k=84, the MCC value is 0.6199112	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=85) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.64164 
+
+    Area under curve (Davis & Goadrich):
+     0.6416416 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6199112 (worst possible: -1; best possible: +1)
+When k=85, the MCC value is 0.6199112	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=86) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6404597 
+
+    Area under curve (Davis & Goadrich):
+     0.6404613 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6191429 (worst possible: -1; best possible: +1)
+When k=86, the MCC value is 0.6191429	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=87) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6372388 
+
+    Area under curve (Davis & Goadrich):
+     0.6372403 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6151631 (worst possible: -1; best possible: +1)
+When k=87, the MCC value is 0.6151631	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=88) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6381885 
+
+    Area under curve (Davis & Goadrich):
+     0.6381901 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6172943 (worst possible: -1; best possible: +1)
+When k=88, the MCC value is 0.6172943	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=89) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6384067 
+
+    Area under curve (Davis & Goadrich):
+     0.6384083 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6159269 (worst possible: -1; best possible: +1)
+When k=89, the MCC value is 0.6159269	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=90) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6395249 
+
+    Area under curve (Davis & Goadrich):
+     0.6395265 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6170336 (worst possible: -1; best possible: +1)
+When k=90, the MCC value is 0.6170336	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=91) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6438927 
+
+    Area under curve (Davis & Goadrich):
+     0.6438943 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6221291 (worst possible: -1; best possible: +1)
+When k=91, the MCC value is 0.6221291	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=92) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6428339 
+
+    Area under curve (Davis & Goadrich):
+     0.6428355 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6206924 (worst possible: -1; best possible: +1)
+When k=92, the MCC value is 0.6206924	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=93) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6438927 
+
+    Area under curve (Davis & Goadrich):
+     0.6438943 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6221291 (worst possible: -1; best possible: +1)
+When k=93, the MCC value is 0.6221291	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=94) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6394047 
+
+    Area under curve (Davis & Goadrich):
+     0.6394063 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6177038 (worst possible: -1; best possible: +1)
+When k=94, the MCC value is 0.6177038	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=95) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6415759 
+
+    Area under curve (Davis & Goadrich):
+     0.6415776 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6202445 (worst possible: -1; best possible: +1)
+When k=95, the MCC value is 0.6202445	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=96) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6438215 
+
+    Area under curve (Davis & Goadrich):
+     0.6438231 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6224554 (worst possible: -1; best possible: +1)
+When k=96, the MCC value is 0.6224554	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=97) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6406475 
+
+    Area under curve (Davis & Goadrich):
+     0.6406491 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6181429 (worst possible: -1; best possible: +1)
+When k=97, the MCC value is 0.6181429	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=98) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.638528 
+
+    Area under curve (Davis & Goadrich):
+     0.6385296 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6152634 (worst possible: -1; best possible: +1)
+When k=98, the MCC value is 0.6152634	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=99) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6405828 
+
+    Area under curve (Davis & Goadrich):
+     0.6405844 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6184729 (worst possible: -1; best possible: +1)
+When k=99, the MCC value is 0.6184729	 (worst possible: -1; best possible: +1)
+
+[Training the kNN model (with k=100) on training set & applying the kNN model to validation set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.642631 
+
+    Area under curve (Davis & Goadrich):
+     0.6426326 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6216818 (worst possible: -1; best possible: +1)
+When k=100, the MCC value is 0.6216818	 (worst possible: -1; best possible: +1)
+
+The best k value is 17, corresponding to MCC=0.673500820316828
+[Optimization end]
+
+[Training the kNN model (with the OPTIMIZED hyper-parameter k=17) on training set & applying the kNN to the test set]
+
+  Precision-recall curve
+
+    Area under curve (Integral):
+     0.6743471 
+
+    Area under curve (Davis & Goadrich):
+     0.674349 
+
+    Curve not computed ( can be done by using curve=TRUE )
+
+MCC = 0.6472466 (worst possible: -1; best possible: +1)
+
+f1_score = 0.7272727 (worst: 0.0; best: 1.0)
+accuracy = 0.8678794 (worst: 0.0; best: 1.0)
+
+true positive rate = recall = 0.6594982 (worst: 0.0; best: 1.0)
+true negative rate = specificity = 0.9438276 (worst: 0.0; best: 1.0)
+