; verbose (-v: yes -v-: no) -v - ; keep intermediary files (-x: yes -x-: no) -x - ; flex rules (input file, binary format) ;-b (not specified, not doing actions with already created flex rules.) ; Word list (Optional if -b is specified. Otherwise N/A) (-I filename) ;-I (N/A) ; Output, lemmas of words in input (-I option) ;-O (N/A) ; word/lemma list -i Label-Delaf_pt_v4_1.tab.ph ; extra file name affix -e ziggurat ; suffix only (-s: yes -s-: no) -s - ; make rules with infixes less prevalent(-A: yes -A-: no) -A - ; columns (1 or F or W=word,2 or B or L=lemma,3 or T=tags,0 or O=other) -n FBO ; max recursion depth when attempting to create candidate rule -Q 1 ; flex rules (output, binary format, can be left unspecified) ;-o (Not specified, autogenerated) ; temp dir (including separator at end!) -j tmp/ ; penalties to decide which rule survives (4 or 6 floating point numbers: R=>R;W=>R;R=>W;W=>W[;R=>N/A;W=>NA], where R=#right cases, W=#wrong cases, N/A=#not applicable cases, previous success state=>success state after rule application) -D 0.1326332335;-0.6279433806;0.7548159800;-0.0161485532;0.1326242114;0.0223252152; ; compute parms (-p: yes -p-: no) -p ; expected optimal pruning threshold (only effective in combination with -XW) -C -1 ; tree penalty (-XC: constant -XD: more support is better -XE: higher entropy is better -XW: Fewer pattern characters other than wildcards is better) -X S ; current parameters (-P filename) -P parms.txt ; best parameters (-B filename) -B best_ziggurat.txt ; start training with minimal fraction of training pairs (-Ln: 0.0 < n <= 1.0) -L 0.053281 ; end training with maximal fraction of training pairs (-Hn: 0.0 < n <= 1.0) -H 1.000000 ; number of differently sized fractions of trainingdata (natural number) -K 20 ; number of iterations of training with same fraction of training data when fraction is minimal (positive number) -N 100.000000 ; number of iterations of training with same fraction of training data when fraction is maximal (positive number) -M 10.000000 ; competition function (deprecated) ;-f (N/A) ; redo training after homographs for next round are removed (-R: yes -R-: no) ;-R - (N/A) ; max. pruning threshold to evaluate -c 5 ; test with the training data (-T: yes -T-: no) -T ; test with data not used for training (-t: yes -t-: no) -t ; create flexrules using full training set (-F: yes -F-: no) -F ; Number of clusters found in word/lemma list: 81813 ; Number of lines found in word/lemma list: 938418 ; Evaluation: ; ----------- ; Lemmatization results for all data in the training set. ; For pruning threshold 0 there may be no errors (diff%%). ; prun. thrshld. 0 1 2 3 4 5 ; rules 70558.000000 39645.000000 13066.000000 7183.000000 5143.000000 4115.000000 ; rules% 7.731392 4.344101 1.431707 0.787077 0.563544 0.450901 ; same% 95.383277 93.461332 92.121668 91.646441 91.406253 91.217784 ; ambi1% 2.267874 2.213963 1.751447 1.455485 1.280384 1.178917 ; ambi2% 2.267874 2.010482 1.425351 1.167302 1.028361 0.946728 ; ambi3% 0.080976 0.039447 0.011725 0.006794 0.005917 0.005917 ; diff% 0.000000 2.274777 4.689810 5.723978 6.279085 6.650654 ; same%stdev 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ; ambi1%stdev 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ; ambi2%stdev 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ; ambi3%stdev 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ; diff%stdev 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ; ;Evaluation of prediction of ambiguity (whether a word has more than one possible lemma) ;--------------------------------------------------------------------------------------- ; amb.rules% 4.616723 4.392423 3.438573 2.901217 2.590572 2.408020 ; false_amb% 0.000000 1.182752 1.475866 1.431159 1.338020 1.270741 ; false_not_amb% 0.000000 1.407052 2.654016 3.146665 3.364171 3.479444 ; true_amb% 4.616723 3.209671 1.962707 1.470058 1.252552 1.137279 ; true_not_amb% 95.383277 94.200524 93.907411 93.952118 94.045257 94.112536 ; precision 1.000000 0.575708 0.399376 0.339319 0.318830 0.309147 ; recall 1.000000 0.695227 0.425130 0.318420 0.271308 0.246339 ; Evaluation: ; ----------- ; Lemmatization results for data that is not part of the training data. ; prun. thrshld. 0 1 2 3 4 5 ; rules 69625.750000 39176.500000 12861.750000 7076.000000 5108.500000 4091.000000 ; rules% 7.733924 4.351667 1.428664 0.785991 0.567445 0.454422 ; same% 89.981988 90.200563 90.332112 90.518306 90.443424 90.394851 ; ambi1% 1.882172 1.783004 1.485499 1.281091 1.179899 1.149542 ; ambi2% 1.445022 1.438951 1.234543 1.048350 1.028111 1.001801 ; ambi3% 0.002024 0.002024 0.008095 0.004048 0.000000 0.002024 ; diff% 6.688794 6.575459 6.939750 7.148206 7.348566 7.451782 ; same%stdev 0.534325 0.386069 0.543349 0.443342 0.519027 0.497025 ; ambi1%stdev 0.115133 0.179237 0.141316 0.041886 0.079162 0.041188 ; ambi2%stdev 0.067831 0.060530 0.085982 0.081800 0.109519 0.106152 ; ambi3%stdev 0.003836 0.003836 0.008840 0.004730 0.000000 0.003836 ; diff%stdev 0.530171 0.427546 0.632535 0.520470 0.518248 0.549971 ; ;Evaluation of prediction of ambiguity (whether a word has more than one possible lemma) ;--------------------------------------------------------------------------------------- ; amb.rules% 3.778511 3.667200 3.159216 2.703851 2.568254 2.483253 ; false_amb% 0.595009 0.580842 0.465483 0.433102 0.366315 0.354172 ; false_not_amb% 0.627391 0.647629 0.639534 0.688106 0.680011 0.675963 ; true_amb% 0.267147 0.246909 0.255004 0.206432 0.214527 0.218575 ; true_not_amb% 21.895934 21.910101 22.025460 22.057841 22.124628 22.136771 ; precision 0.183333 0.175287 0.215017 0.192453 0.226496 0.235808 ; recall 0.298643 0.276018 0.285068 0.230769 0.239819 0.244344 ; ; Power law relating the number of rules in the decision tree to the number of examples in the training data ;---------------------------------------------------------------------------------------------------------- ; #rules = 0.633*N^0.844 0.439*N^0.829 0.226*N^0.795 0.254*N^0.742 0.259*N^0.716 0.268*N^0.697 ; Postscriptum ; The number of rules can be estimated from the number of training examples by ; a power law. See the last line in the table above, which is based on 7 ; different samples from the total available training data mass varying in size ; from 1.54 % to 98.56 %