Results of Proposed Rules - 2024 updates
I. Results16 new non-duplicated SD-Rules are proposed to be added to the SD-Rule for evaluation. The good rule threshold is at 105 in the optimal set. The results are described as follows:
SD-Rule | Rank | Precision | Instances | Source | Decompose | Results |
---|---|---|---|---|---|---|
Good Rules (6) | ||||||
$|noun|free$|adj | 43 | 95.83%% | 24 | EXP_SUG | Root-Parent | Good SD-Rule |
ier$|noun|y$|verb | 56 | 93.33% | 75 | WORDNET | Root-Parent | Good SD-Rule |
ize$|verb|y$|noun | 76 | 87.37% | 95 | WORDNET | CHILD: mize$|verb|my$|noun | Good SD-Rule |
$|verb|per$|noun | 81 | 85.96% | 57 | WORDNET | CHILD: p$|verb|pper$|noun | Good SD-Rule |
$|verb|ter$|noun | 92 | 80.00% | 55 | WORDNET | CHILD: t$|verb|tter$|noun | Good SD-Rule |
$|noun|ly$|adj | 98 | 75.74% | 136 | WORDNET | Root-Parent | Good SD-Rule |
Bad Rules (10) | ||||||
$|noun|ian$|ad | 108 | 69.55% | 243 | WORDNET | Root-Parent | Bad SD-Rule |
e$|verb|ive$|adj | 109 | 68.94% | 425 | WORDNET | Root-Parent | Bad SD-Rule |
$|noun|ise$|verb | 114 | 65.61% | 410 | WORDNET | Root-Parent | Bad SD-Rule |
$|verb|ive$|adj | 118 | 59.68% | 248 | WORDNET | Root-Parent | Bad SD-Rule |
$|noun|er$|noun | 120 | 58.12% | 1237 | WORDNET | Root-Parent | BAD SD-Rule |
e$|verb|ory$|adj | 121 | 57.77% | 251 | WORDNET | Root-Parent | Bad SD-Rule |
$|noun|an$|adj | 135 | 47.11% | 121 | WORDNET | Root-Parent | Bad SD-Rule |
asm$|noun|astic$|adj | 144 | 36.11% | 36 | WORDNET | SLEF | Bad SD-Rule |
$|noun|ical$|adj | 149 | 19.51% | 164 | WORDNET | Root-Parent | Bad SD-Rule |
al$|adj|s$|noun | 152 | 9.09% | 264 | WORDNET | Root-Parent | Bad SD-Rule |
Proposed PARENT rule | CHILD Rule used | Notes | ||||||
---|---|---|---|---|---|---|---|---|
ize$|verb|y$|noun | mize$|verb|my$|noun | Good Rule | ||||||
$|verb|per$|noun | p$|verb|pper$|noun | Good Rule | ||||||
$|verb|ter$|noun | t$|verb|tter$|noun | Good Rule
PARENT rule | Proposed CHILD Rule used | Notes
| m$|noun|tic$|adj asm$|noun|astic$|adj SELF rule, Bad Rule
| |
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}).
ID | SD-Rule | Rank | Notes |
---|---|---|---|
Added in 2015: Freq. > 200, Coverage > 1.00% , Accum. Coverage > 80.0% | |||
1 | $|adj|ness$|noun | 1 | Good |
2 | bility$|noun|ble$|adj | 2 | Good (ility$|noun|le$|adj) |
3 | se$|verb|zation$|noun | 3 | Good |
4 | sation$|noun|ze$|verb | 4 | Good |
5 | iness$|noun|y$|adj | 16 | Good |
6 | ation$|noun|e$|verb | 21 | Good |
7 | nce$|noun|nt$|adj | 25 | Good (ce$|noun|t$|adj) |
8 | e$|verb|ion$|noun | 26 | Good |
9 | cy$|noun|t$|adj | 27 | Good |
10 | $|verb|ment$|noun | 28 | Good |
11 | ication$|noun|y$|verb | 29 | Good |
12 | ed$|adj|ion$|noun | 30 | Good |
13 | $|adj|ity$|noun | 32 | Good |
14 | e$|adj|ity$|noun | 35 | Good |
15 | $|verb|ion$|noun | 49 | Good |
16 | $|verb|ing$|noun | 53 | Good |
17 | $|verb|ation$|noun | 61 | Good |
Added in 2016: Freq. > 100, coverage > 0.40% , Accum. Coverage > 83.36%) | |||
18 | e$|verb|is$|noun | 43 | Good |
19 | ation$|noun|ed$|adj | 50 | Good |
20 | e$|verb|ing$|noun | 60 | Good |
21 | $|adj|ism$|noun | 62 | Good |
22 | e$|adj|ion$|noun | 100 | Bad |
Added in 2017: Freq. > 70, Coverage > 0.30% , Accum. Coverage > 85.00%) | |||
23 | sation$|noun|zed$|adj | 7 | Good |
24 | sed$|adj|zation$|noun | 8 | Good |
25 | sity$|noun|us$|adj | 65 | Good (osity$|noun|ous$|adj) |
26 | e$|verb|tion$|noun | 63 | Good |
27 | ous$|adj|y$|noun | 116 | Bad (exit in 2013) |
Added in 2020: Freq. > 50, Coverage > 0.20% , Accum. Coverage > 87.41%) | |||
28 | ability$|noun|ible$|adj | 10 | Good |
29 | sable$|adj|zability$|noun | 12 | Good |
30 | sability$|noun|zable$|adj | 13 | Good |
31 | sis$|noun|ze$|verb | 41 | Good |
32 | al$|noun|e$|verb | 92 | Good |
Added in 2021: Freq. > 40, Coverage > 0.17% , Accum. Coverage > 89.27%) | |||
33 | ability$|noun|eable$|adj | 34 | Good |
34 | c$|adj|sm$|noun | 134 | Bad |
35 | er$|verb|ration$|noun | 15,29 | Good |
36 | $|verb|nce$|noun | 74 | Good |
37 | ed$|adj|ment$|noun | 85 | Bad |
38 | ity$|noun|y$|adj | 68 | Good |
39 | $|adj|y$|noun | 145 | Bad |
40 | able$|adj|eability$ | 13 | Good |
41 | e$|verb|ition$|noun | 71 | Good |
42 | d$|verb|sion$|noun | 56 | Good |
The results shows 88.09% (37/42) are good SD-Rules, more SD-Rules from nomD should be added and evaluated in the future releases.
III. Further Observation on ORG_FACT
The top SD-Rules generated from ORG_FACT are added and evaluated (${SUFFIX_D}/data/${YEAR}/dataR/SdRulesFromSdPairs/orgFacts/sdRulesFromSdPairs.rpt.${YEAR}).
ID | SD-Rule | Rank | Notes | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Added in 2015: Freq. > 40, Coverage > 1.00% , Accum. Coverage > 11.50% | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1 | $|noun|less$|adj | 17 | Good | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
2 | $|adj|ally$|adv | 23 | Good | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
3 | ist$|noun|y$|noun | 45 | Good | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
4 | $|verb|ion$|noun | 49 | Good, also in NOM_D | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
5 | c$|adj|s$|noun | 57 | Good (ic$|adj|is$|noun) | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
6 | $|noun|ful$|adj | 64 | Good | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Added in 2016: Freq. >= 35; Accu. coverage: > 16.00% Ind Coverage: > 0.80% | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
7 | sia$|noun|tic$|adj |
The results shows 60.87% (14/23) 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_FACT down the list.
ORG_FACT is closed to the limit, maybe review 1 more year until there is no good rules can be found.