The purpose of this paper is to introduce various attempts to quantify meaning, review their significance, and then find ways to supplement methods to quantify meaning. Until now, meaning has been quantified based on the word frequency in the document...
The purpose of this paper is to introduce various attempts to quantify meaning, review their significance, and then find ways to supplement methods to quantify meaning. Until now, meaning has been quantified based on the word frequency in the document and the word frequency of search words and their collocates. The former is a method that does not consider the order of appearance of words, and the latter considers only adjacent words. To elaborate the quantification of meaning, efforts are needed to reduce the importance of relationships between adjacent but syntactically unrelated words and to pay attention to relationships between distant but syntactically related words. If meaning is quantified in this way, vectorization of words or sentences can become more sophisticated, and based on this, more efficient language models can be developed.