Lexical Tools

Results of Proposed Rules - 2020

I. Results

11 non-duplicated SD-Rules are proposed to be added to the SD-Rule for evaluation. The results from the optimal set are described as follows:

SD-RuleRankPrecisionInstancesSourceDecomposeResults
Good Rules
ion$|noun|ory$|adj5100.00%471ORG_DRoot-ParentGood SD-Rule
ability$|noun|ible$|adj10100.00%127NOM_DRoot-ParentGood SD-Rule
sable$|adj|zability$|noun1297.85%186NOM_DRoot-ParentGood SD-Rule
sability$|noun|zable$|adj1392.49%253NOM_DRoot-ParentGood SD-Rule
sis$|noun|ze$|verb4190.00%100NOM_DRoot-ParentGood SD-Rule
$|adj|s$|noun4989.80%98ORG_DRoot-ParentGood SD-Rule
al$|noun|e$|verb9290.00%100NOM_DRoot-ParentGood SD-Rule
Bad Rules
$|verb|age$|noun10061.29%93ORG_DRoot-ParentBad SD-Rule
$|noun|ial$|adj10650.00%18ORG_DRoot-ParentBad SD-Rule
$|noun|oid$|noun10447.62%42EXP_SUGRoot-ParentBad SD-Rule
$|adj|oid$|adj11047.62%42EXP_SUGRoot-ParentBad SD-Rule
  • Good SD-Rules: 7 of them are evaluated as good rules in the optimized set
  • Bad Sd-Rules: 4 are bad rules.

  • Experts' suggestion: 0% (0/2) is good.
  • Computation rules: 78% (7/9) is good.
    • NOM_D: 100% (5/5) is good
    • ORG_D: 50% (2/4) is good

  • Also, in the optimized set, 4 child rules are used to replace proposed root-parent rules
    • bility$|noun|ble$|adj from ility$|noun|le$|adj
    • nce$|noun|nt$|adj from ce$|noun|t$|adj
    • enesis$|noun|enic$|adj from esis$|noun|ic$|adj
    • ic$|adj|is$|noun from c$|adj|s$|noun

II. Further Observation on NOM_D

The top SD-Rules generated from NOM_D are added and evaluated (${SUFFIX_D}/data/${YEAR}/dataR/SdRulesFromSdPairs/nomD/sdRulesFromSdPairs.rpt.${YEAR}).

IDSD-RuleRankNotes
Added in 2015: Freq. > 200, Coverage > 1.00% , Accum. Coverage > 80.0%
1$|adj|ness$|noun1Good
2bility$|noun|ble$|adj2Good (ility$|noun|le$|adj)
3se$|verb|zation$|noun3Good
4sation$|noun|ze$|verb4Good
5iness$|noun|y$|adj16Good
6ation$|noun|e$|verb21Good
7nce$|noun|nt$|adj25Good (ce$|noun|t$|adj)
8e$|verb|ion$|noun26Good
9cy$|noun|t$|adj27Good
10$|verb|ment$|noun28Good
11ication$|noun|y$|verb29Good
12ed$|adj|ion$|noun30Good
13$|adj|ity$|noun32Good
14e$|adj|ity$|noun35Good
15$|verb|ion$|noun49Good
16$|verb|ing$|noun53Good
17$|verb|ation$|noun61Good
Added in 2016: Freq. > 100, coverage > 0.40% , Accum. Coverage > 83.36%)
18e$|verb|is$|noun43Good
19ation$|noun|ed$|adj50Good
20e$|verb|ing$|noun60Good
21$|adj|ism$|noun62Good
22e$|adj|ion$|noun100Bad
Added in 2017: Freq. > 70, Coverage > 0.30% , Accum. Coverage > 85.00%)
23sation$|noun|zed$|adj7Good
24sed$|adj|zation$|noun8Good
25sity$|noun|us$|adj65Good (osity$|noun|ous$|adj)
26e$|verb|tion$|noun63Good
27ous$|adj|y$|noun116Bad (exit in 2013)
Added in 2020: Freq. > 50, Coverage > 0.20% , Accum. Coverage > 87.41%)
28ability$|noun|ible$|adj10Good
29sable$|adj|zability$|noun12Good
30sability$|noun|zable$|adj13Good
31sis$|noun|ze$|verb41Good
32al$|noun|e$|verb92Good

The results shows 93.75% (30/32) are good SD-Rules, more SD-Rules from nomD should be added and evaluated in the future releases.

III. Further Observation on ORG_D

The top SD-Rules generated from ORG_D are added and evaluated (${SUFFIX_D}/data/${YEAR}/dataR/SdRulesFromSdPairs/orgFacts/sdRulesFromSdPairs.rpt.${YEAR}).

IDSD-RuleRankNotes
Added in 2015: Freq. > 40, Coverage > 1.00% , Accum. Coverage > 11.50%
1$|noun|less$|adj17Good
2$|adj|ally$|adv23Good
3ist$|noun|y$|noun45Good
4$|verb|ion$|noun49Good, also in NOM_D
5c$|adj|s$|noun57Good (ic$|adj|is$|noun)
6$|noun|ful$|adj64Good
Added in 2016: Freq. >= 35; Accu. coverage: > 16.00% Ind Coverage: > 0.80%
7sia$|noun|tic$|adj47Good
8e$|noun|ic$|adj56Good
9on$|noun|ve$|adj58Good
10$|noun|ship$|noun79Good
11$|noun|age$|noun114Bad
Added in 2017: Freq. >= 30; Accu. coverage: > 19.00% Ind Coverage: > 0.70%
12$|noun|tous$|adj33Good
13$|noun|ish$|adj94Bad
14$|noun|y$|noun101Bad
15$|noun|fully$|adv128Bad
Added in 2020: Freq. >= 25; Accu. coverage: > 23.00% Ind Coverage: > 0.60%
16ion$|noun|ory$|adj5Good
17$|adj|s$|noun49Good
18$|verb|age$|noun100Bad
19$|noun|ial$|adj106Bad

The results shows 68.42% (13/19) are good SD-Rules, more SD-Rules from orgD should be added in the future releases.

V. Future Work

Evaluated more SD-Rules from NOM_D and ORG_D down the list.