Treebank Statistics: UD_Marathi-CMUPAN: POS Tags: NUM
There are 633 NUM lemmas (4%), 737 NUM types (2%) and 3619 NUM tokens (3%).
Out of 14 observed tags, the rank of NUM is: 5 in number of lemmas, 5 in number of types and 9 in number of tokens.
The 10 most frequent NUM lemmas: एक, दोन, एका, तीन, चार, हजार, पाच, २२, दोन्ही, लाख
The 10 most frequent NUM types: एक, दोन, एका, तीन, हजार, चार, पाच, २२, दोन्ही, सहा
The 10 most frequent ambiguous lemmas: एक (NUM 587, DET 4, PROPN 2, ADJ 1, NOUN 1), एका (NUM 151, DET 1), चार (NUM 107, VERB 6, ADJ 1, NOUN 1), हजार (NUM 103, NOUN 5, ADV 2), २२ (NUM 73, NOUN 1, PROPN 1), दोन्ही (NUM 60, NOUN 1), लाख (NUM 60, NOUN 6, ADV 1), सहा (NUM 57, PROPN 1), १० (NUM 56, NOUN 1), कोटी (NUM 48, NOUN 9)
The 10 most frequent ambiguous types: एक (NUM 517, DET 2, PROPN 1), एका (NUM 169, DET 2), २२ (NUM 72, NOUN 1, PROPN 1), १० (NUM 54, NOUN 1), लाख (NUM 52, NOUN 2), कोटी (NUM 44, NOUN 1), सात (NUM 34, NOUN 1), २५ (NUM 27, NOUN 1), ८ (NUM 22, NOUN 1), एकाच (NUM 17, DET 1)
- एक
- एका
- २२
- १०
- लाख
- कोटी
- सात
- २५
- ८
- एकाच
Morphology
The form / lemma ratio of NUM is 1.164297 (the average of all parts of speech is 1.720393).
The 1st highest number of forms (9) was observed with the lemma “२”: २, २,११६, २,१३३, २,२८६, २,४१२, २,४३९, २,७००, २.२५, २.७.
The 2nd highest number of forms (7) was observed with the lemma “एक”: एक, एकच, एकही, एका, एकाच, एकाने, एकाही.
The 3rd highest number of forms (7) was observed with the lemma “एका”: एका, एकाच, एकाचे, एकापेक्षा, एकामागे, एकास, एकाही.
NUM occurs with 4 features: NumType (3475; 96% instances), Number (103; 3% instances), Gender (74; 2% instances), Case (64; 2% instances)
NUM occurs with 8 feature-value pairs: Case=Nom, Gender=Fem, Gender=Masc, Gender=Neut, NumType=Card, NumType=Ord, Number=Plur, Number=Sing
NUM occurs with 29 feature combinations.
The most frequent feature combination is NumType=Card (3400 tokens).
Examples: एक, दोन, एका, तीन, हजार, चार, पाच, २२, दोन्ही, सहा
Relations
NUM nodes are attached to their parents using 14 different relations: nummod (2849; 79% instances), nmod (171; 5% instances), obl (164; 5% instances), compound (151; 4% instances), root (81; 2% instances), dep (52; 1% instances), amod (51; 1% instances), conj (38; 1% instances), nsubj (33; 1% instances), obj (14; 0% instances), xcomp (10; 0% instances), acl (2; 0% instances), iobj (2; 0% instances), acl:relcl (1; 0% instances)
Parents of NUM nodes belong to 12 different parts of speech: NOUN (2672; 74% instances), NUM (397; 11% instances), VERB (251; 7% instances), PROPN (149; 4% instances), (81; 2% instances), ADJ (23; 1% instances), ADP (15; 0% instances), ADV (11; 0% instances), DET (11; 0% instances), AUX (4; 0% instances), PRON (4; 0% instances), SCONJ (1; 0% instances)
2778 (77%) NUM nodes are leaves.
484 (13%) NUM nodes have one child.
158 (4%) NUM nodes have two children.
199 (5%) NUM nodes have three or more children.
The highest child degree of a NUM node is 8.
Children of NUM nodes are attached using 15 different relations: punct (389; 24% instances), dep (279; 17% instances), nummod (264; 16% instances), compound (219; 14% instances), nmod (170; 11% instances), case (117; 7% instances), conj (56; 3% instances), nsubj (37; 2% instances), cc (30; 2% instances), cop (29; 2% instances), amod (17; 1% instances), acl (5; 0% instances), det (2; 0% instances), mark (2; 0% instances), advmod (1; 0% instances)
Children of NUM nodes belong to 13 different parts of speech: NUM (397; 25% instances), PUNCT (389; 24% instances), NOUN (306; 19% instances), SCONJ (135; 8% instances), PROPN (111; 7% instances), ADV (90; 6% instances), ADP (74; 5% instances), DET (35; 2% instances), ADJ (32; 2% instances), AUX (29; 2% instances), VERB (10; 1% instances), PRON (6; 0% instances), PART (3; 0% instances)