📦 @bntk/pos

UniversalPOSTag

Defined in: index.ts:11

Universal POS tags based on Universal Dependencies (UD) v2 specification Source: https://universaldependencies.org/u/pos/

Enumeration Members

Enumeration Member	Value	Description	Defined in
ADJ	"ADJ"	Adjectives are words that typically modify nouns Example বড়/boro, সুন্দর/sundor, নতুন/notun	index.ts:14
ADP	"ADP"	Adpositions are prepositions and postpositions Example মধ্যে/moddhe, থেকে/theke, উপরে/upore	index.ts:17
ADV	"ADV"	Adverbs are words that typically modify verbs, adjectives or other adverbs Example খুব/khub, ধীরে/dhire, ভালোভাবে/bhalobhabe	index.ts:20
AUX	"AUX"	Auxiliary verbs are used to form tenses, moods, etc. Example আছে/ache, হয়/hoi, হবে/hobe	index.ts:23
CCONJ	"CCONJ"	Coordinating conjunctions connect words, phrases, clauses of equal status Example এবং/ebong, কিন্তু/kintu, অথবা/othoba	index.ts:26
DET	"DET"	Determiners are words that modify nouns or noun phrases Example এই/ei, সেই/sei, কোন/kon	index.ts:29
INTJ	"INTJ"	Interjections are exclamatory words Example ওহ/oh, বাহ/bah, হায়/hay	index.ts:32
NOUN	"NOUN"	Nouns are words denoting all physical objects and materials Example বই/boi, মানুষ/manush, বাড়ি/bari	index.ts:35
NUM	"NUM"	Numerals represent numbers, quantities, etc. Example এক/ek, দুই/dui, প্রথম/prothom	index.ts:38
PART	"PART"	Particles are function words that must be associated with another word Example না/na, তো/to, কি/ki	index.ts:41
PRON	"PRON"	Pronouns substitute for nouns or noun phrases Example আমি/ami, তুমি/tumi, সে/se	index.ts:44
PROPN	"PROPN"	Proper nouns are names of specific persons, places, organizations Example ঢাকা/dhaka, রবীন্দ্রনাথ/robindronath, বাংলাদেশ/bangladesh	index.ts:47
PUNCT	"PUNCT"	Punctuation marks Example ।, ?, !, ,	index.ts:50
SCONJ	"SCONJ"	Subordinating conjunctions link dependent clauses to independent ones Example যদি/jodi, কারণ/karon, যখন/jokhon	index.ts:53
SYM	"SYM"	Symbols represent currency, math operators, etc. Example ৳, +, =	index.ts:56
VERB	"VERB"	Verbs denote actions and processes Example যাই/jai, খাই/khai, পড়ি/pori	index.ts:59
X	"X"	Other words that don't fit into above categories Example ইত্যাদি/ittyadi, প্রভৃতি/probhriti	index.ts:62

---

TaggedWord

Defined in: index.ts:68

Represents a word with its POS tag and additional linguistic features

Properties

Property	Type	Description	Defined in
features?	Record<string, string>	Additional linguistic features (e.g., gender, number, case)	index.ts:74
tag	UniversalPOSTag	The POS tag for the word	index.ts:72
word	string	The original word	index.ts:70

---

getPOSTagFrequencies()

function getPOSTagFrequencies(taggedWords): Map<UniversalPOSTag, number>;

Defined in: index.ts:196

Gets the most common POS tags in a sequence of tagged words

Parameters

Parameter	Type	Description
taggedWords	TaggedWord[]	Array of tagged words

Returns

Map<UniversalPOSTag, number>

Map of POS tags to their frequencies

Description

This function analyzes a sequence of tagged words and returns a map showing how many times each POS tag appears.

Example

const tagged = [
  { word: "আমি", tag: UniversalPOSTag.PRON },
  { word: "বাংলায়", tag: UniversalPOSTag.ADP },
  { word: "গান", tag: UniversalPOSTag.NOUN },
  { word: "গাই", tag: UniversalPOSTag.VERB },
];
const frequencies = getPOSTagFrequencies(tagged);
console.log(frequencies);
// Output: Map(4) {
//   UniversalPOSTag.PRON => 1,
//   UniversalPOSTag.ADP => 1,
//   UniversalPOSTag.NOUN => 1,
//   UniversalPOSTag.VERB => 1
// }

---

tagText()

function tagText(text): TaggedWord[];

Defined in: index.ts:162

Tags a Bangla text string with parts of speech

Parameters

Parameter	Type	Description
text	string	The Bangla text to tag

Returns

TaggedWord[]

Array of words with their POS tags and features

Description

This function first tokenizes the input text into words and then performs POS tagging on the resulting word sequence.

Example

const text = "আমি বাংলায় গান গাই";
const tagged = tagText(text);
console.log(tagged);
// Output: [
//   { word: "আমি", tag: UniversalPOSTag.PRON },
//   { word: "বাংলায়", tag: UniversalPOSTag.ADP },
//   { word: "গান", tag: UniversalPOSTag.NOUN },
//   { word: "গাই", tag: UniversalPOSTag.VERB }
// ]

---

tagWord()

function tagWord(word): TaggedWord;

Defined in: index.ts:102

Tags a single Bangla word with its part of speech

Parameters

Parameter	Type	Description
word	string	The Bangla word to tag

Returns

TaggedWord

The word with its POS tag and features

Description

This function performs POS tagging for a single Bangla word. It uses a combination of rules and dictionary lookup to determine the most likely POS tag for the given word.

Examples

const result = tagWord("বাংলা");
console.log(result);
// Output: { word: "বাংলা", tag: UniversalPOSTag.NOUN }

const result = tagWord("সুন্দর");
console.log(result);
// Output: { word: "সুন্দর", tag: UniversalPOSTag.ADJ }

---

tagWords()

function tagWords(words): TaggedWord[];

Defined in: index.ts:134

Tags a sequence of Bangla words with their parts of speech

Parameters

Parameter	Type	Description
words	string[]	Array of Bangla words to tag

Returns

TaggedWord[]

Array of words with their POS tags and features

Description

This function performs POS tagging for a sequence of Bangla words. It takes into account the context of surrounding words to improve tagging accuracy.

Example

const words = ["আমি", "বাংলায়", "গান", "গাই"];
const tagged = tagWords(words);
console.log(tagged);
// Output: [
//   { word: "আমি", tag: UniversalPOSTag.PRON },
//   { word: "বাংলায়", tag: UniversalPOSTag.ADP },
//   { word: "গান", tag: UniversalPOSTag.NOUN },
//   { word: "গাই", tag: UniversalPOSTag.VERB }
// ]