Özet:

Studies in the area of language dictionary are focused on automatic dictionary creation. In this article, an English word is given as a reference and an automatic creation of the dictionary of the article subject is provided. The first dictionary word is derived from this English word which is given as a starting point for the system. Web search is then performed in the Azure Web Cognitive Web Search system by using the first seed word added to the dictionary. The first document from the search result has meaningful words with the Helmholtz Principle as applied to the reference document. Among the meaningful words found, the word with the highest value is added to the dictionary. Thus, as a result of processing a document obtained from the Web, the most meaningful word for that document is added to the dictionary only. Then, the word added to the dictionary is searched on the Web. The documents obtained as a result of web search are put into the system and the meaningful words of these documents are calculated. The search cycle on the web is repeated in this way and finally ends when the desired number of words for the dictionary is reached. In order to measure the performance of the dictionary, Hash similarity value was calculated. Dictionaries with a similarity of 40.46% can be created in the tests performed on the reference words given in different subjects.

Anahtar Kelimeler: