Retrieve New SD-Rules from NomD, 2016

I. Description

A set of computer programs (FindSdRulesFromDPairs.java) are developed to find the SD-Rules from a set of suffixD pairs. It identifies and eliminates the same starting characters of a SD-pair and then generates the SD-Rules automatically. Please note that only root-parent SD-Rules is generated in this program. Two sets of SD-pairs are used for this task. This page details the new SD-Rules selected from nomD.

II. Procedures

• Directory: ${SUFFIXD_DIR}
• Programs:
  • shell>cd ${SUFFIXD_DIR}/bin
  • shell>GetSdRule ${YEAR}
2
nomD

  • shell>GetSdRule ${YEAR}
3
nomD

  • shell>GetSdRule ${YEAR}
5
${YEAR}
rule

III. Results

• From: ./data/${YEAR}/dataR/SdRulesFromSdPairs/nomD/sdRulesFromSdPairs.rpt
• These are SD-Pairs from nominalizations of Lexicon.2016
• The file of ${NOM_D_DIR}/data/nomD.yes.S.data is used as input
• There are 23,485 SD-Pairs to generate 1,026 SD-Rules,
• All generated SD-Rules are root-parent rules (without parent-rule).
• Rules with following criteria are selected:
  • 2015 release:
    • frequency: >= 200:
    • Accumulate coverage: 80.34% (> 80.00%)
    • Individual coverage: 1.26% (> 1.00%)
  • 2016 release:
    • frequency: >= 100:
    • Accumulate coverage: 83.36%
    • Individual coverage: 0.40%

  • SD-Rules meet above criteria (total instance No. 23,485):

    SD-Rules	Instances No.	Accu. No.	Notes
    $|adj|ness$|noun	2734 (11.64%)	2734 (11.64%)	2013-, existing rule
    ation$|noun|e$|verb	2491 (10.61%)	5225 (22.25%)	2013-, existing rule
    e$|verb|ion$|noun	2299 (9.79%)	7524 (32.04%)	2015, has child rules exist
    $|adj|ity$|noun	2037 (8.67%)	9561 (40.71%)	2013-, existing rule
    ility$|noun|le$|adj	1612 (8.67%)	11173 (47.58%)	2015, has child rules exist
    se$|verb|zation$|noun	1108 (4.72%)	12281 (52.29%)	2015, has no child rules exist
    sation$|noun|ze$|verb	1072 (4.56%)	13353 (56.86%)	2015, has no child rules exist
    ce$|noun|t$|adj	843 (3.59%)	14196 (60.45%)	2015, has child rules exist
    e$|adj|ity$|noun	833 (3.55%)	15029 (63.99%)	2013-, existing rule
    ed$|adj|ion$|noun	691 (2.94%)	15720 (66.94%)	2013-, existing rule
    $|verb|ment$|noun	575 (2.45%)	16295 (69.38%)	2013-, existing rule
    iness$|noun|y$|adj	545 (2.32%)	16840 (71.71%)	2013-, existing rule
    $|verb|ion$|noun	536 (2.28%)	17376 (73.99%)	2013-, existing rule
    $|verb|ing$|noun	480 (2.04%)	17856 (76.03%)	2013-, existing rule
    cy$|noun|t$|adj	401 (1.71%)	18257 (77.74%)	2015, has child rules exist
    $|verb|ation$|noun	307 (1.31%)	18564 (79.05%)	2013-, existing rule
    ication$|noun|y$|verb	295 (1.26%)	18859 (80.30%)	2013-, existing rule
    2015: Frequency > 200, Instance coverage > 1.00% , Accum. Coverage > 80.0%
    e$|verb|ing$|noun	191 (0.81%)	19050 (81.12%)	2013-, existing rule
    ation$|noun|ed$|adj	158 (0.67%)	19208 (81.79%)	2016, has no child rules exist
    $|adj|ism$|noun	133 (0.57%)	19341 (82.35%)	2016, has no child rules exist
    e$|adj|ion$|noun	123 (0.52%)	19464 (82.88%)	2016, has no child rules exist
    e$|verb|is$|noun	113 (0.48%)	19577 (83.36%)	2016, has child rules exist
    2016: Frequency > 100, Instance coverage > 0.40% , Accum. Coverage > 83.36%)

  • New SD-Rules with childred rules

    New rules	Instances	Examples	Child-rules
    e$|verb|is$|noun	113	diagnose|verb|E0022275|diagnosis|noun|E0022276	ose$|verb|osis$|noun|2013|ORG_RULE|CHILD

  • New SD-Rules without child-rule

    New rules	Instances	Examples	Child-rules
    ation$|noun|ed$|adj	158	excitation|noun|E0026541|excited|adj|E0218417	N/A
    $|adj|ism$|noun	133	lyric|adj|E0421373|lyricism|noun|E0589144	N/A
    e$|adj|ion$|noun	123	opposite|adj|E0540612|opposition|noun|E0044001	N/A