Assignment Task

 

Fuzzy Kurdish Semantic Similarity Measures 
  
Semantic Similarity Measures are designed to calculate the similarity between words or short texts. They were mainly designed for use with the English language, of which there exists a language rich resource. The most establish measured is known as STASIS [8] which was reliant on the Word Net repository, and was developed for the specific purpose of  representing the level of syntactic and semantic similarity between short texts through the use of ontological relations between words. Recently,  theoretical research has been undertaken to develop Arabic word and semantic similarity algorithms [1,2,9]. These measures have been tested on pilot datasets and had shown success but have not been deployed in any real application. The Arabic semantic similarity measure developed by  [1] included the use of Arabic nouns and verbs. 
Semantic similarity measures take advantage of the techniques from Natural Language Processing (NLP), in order to identify the similarity of any two words, even if they are not exact matches. 
In  recent research, the Kurdish language too has entered the era of NLP. A KLPT – Kurdish Language Processing Toolkit (KLTP) has been published as an open resource. 
The latter can be beneficial as it will have services for NLP such as Parsing, stemming, lementisation, stop words, clauses, parts of speech (nouns)  etc. The toolkit also provides the processes and codes for Kurdish language in the NLP. 
Although, Kurdish language has descended from Aryan languages, meaning that it is a sister to the Persian language and is written in Arabic-based script, yet it has its own morphology and grammatic rules [10]. It is as low resource language and there are no theoretical research developments for a such semantic similarity algorithm, nor real applications have been deployed as of today. [4,5 ]. An initial  survey of literature, has found that no word or short text similiarty measures have been developed for the Kurdish language. 
  
Regardless of language, semantic similarity measures do not typically identify or include “fuzzy words” in the actual similarity calculation. A fuzzy word is one which has subjective definition i.e. “hot”, “cold etc. One known fuzzy similarity measure, known as FAST [3], exists which was created for a portion of the English language. This work established that fuzzy words have a significant impact on non-fuzzy words in a short text when calculating the similarity in that the measures were closer to that of humans.  No work has been conducted on modelling fuzzy words in the context of modern Kurdish language. In the context of traditional measures such as STASIS [8], if the word ‘hot’ appeared in a pair of sentences it would always match as being identical, however, it does not model the fuzziness (human perception) of the word ‘hot’ in the context of it being used. Neither does it apply the implication of fuzzy linguistic variables such as ‘very’. 
The aim of this project is to develop a Fuzzy Kurdish Semantic Similarity Measure  which is suitable for application in a number of  technologies from search queries to tweet analysis or for using it in dialogue systems. The proposed research will attempt to answer the following research questions:   

  • 1Is it possible to construct a framework for developing a fuzzy short text semantic similarity measure for Kurdish Language? 
  • 2.Is it possible to measure the semantic similarity between a pair of Kurdish fuzzy words in relation to other words present in the short text? 
  • 3.Can the use of fuzzy words allow disambiguation of  all words in a Kurdish short text? 

 


This English Assignment has been solved by our English Experts at UniLearnO. Our Assignment Writing Experts are efficient to provide a fresh solution to this question. We are serving more than 10000+Students in Australia, UK & US by helping them to score HD in their academics. Our Experts are well trained to follow all marking rubrics & referencing style.

Be it a used or new solution, the quality of the work submitted by our assignment Experts remains unhampered. You may continue to expect the same or even better quality with the used and new assignment solution files respectively. There’s one thing to be noticed that you could choose one between the two and acquire an HD either way. You could choose a new assignment solution file to get yourself an exclusive, plagiarism (with free Turnitin file), expert quality assignment or order an old solution file that was considered worthy of the highest distinction.

  • Uploaded By : Roman
  • Posted on : April 23rd, 2019
  • Downloads : 422

Whatsapp Tap to ChatGet instant assistance