Kalmasoft PoS Tagset

Oveview

Part of speech tagging is the process of selecting the most likely sequence of syntactic categories for the words in a sentence. It determines grammatical characteristics of the words, such as part of speech, grammatical number, gender, person, etc. In the case of complex morphology languages, this task is not trivial since most of the words are ambiguous as a result of the absence of vowels eg Arabic, Hebrew and Syriac or multilevel syntactic affixation eg Amharic and Geez; the notion of "word" may not be fully applicable to any surface form found in those languages, some forms are actually very compact full sentences. KTagset system is developed to address such complexities and so designed for these purposes.

KTagset

A KTag is string of thirty [30] characters grouped in three 4-characters sets and one 18-characters set [4-18-4-4], the 18 characters set ends with five place holders reserved to represent optional language variety or dialect while the rest of characters have their different grammatical notation, each letter may have different representation depending on the set it resides in and the specific position it occupies, the tag string is basically case sensitive with optional lower case or Greek letters for different levels of detail and extended grammatical features, KTagset is broadly categorized as follows {syntax}{morphology}{semantics}{ontology}.

"يعطيانكموها" => {VINA}{•10T3DM2PM3SF04501}{••••}{••••}

ترميز كلماسوفت (KTagset) يستخدم سلسلة من ثلاثين حرفاً موزعة إلى أربعة مجموعات الثانية منها تضم ثمانية عشر حرفاً تنتهي بخمسة أرقام أو حروف تبين مستوى اللغة بينما البقية تتألف من أربعة لكل مجموعة، أي حرف في الترميز يأخذ وظيفته بحسب المجموعة وترتيبه داخلها، الترميز مرتب إلى الوظائف التالية (النحو، الإملاء، الدلالة، التكوين).

---