MySQL Natural Language Full-text searches are directed to search queries against the body to search for the most relevant matches. Therefore, suppose we have an article that contains “I love cake” and we have documents d1, d2, d3 (the database in your case). Documents 1 and 2 relate to sport and religion, respectively, and document 3 refers to food. Your request,
SELECT * FROM articles WHERE MATCH (name, body) AGAINST ('database' IN NATURAL LANGUAGE MODE);
Will return d3 and then d2, d1 (random order d2, d1, depending on which is larger than the article), because d3 best fits the article.
The main algorithm that MYSQL uses is probably the tf-idf algorithm, where tf denotes the time frequency and idf for the inverse frequency of the document. tf, he says, is simply the number of times the word w in an article occurs in document A. idf is based on how many documents the word has. Therefore, the words that are found in many documents do not contribute to the adoption of the most representative document. The product tf * idf gives an estimate, the higher the better the word is a document. Thus, “pie” will only appear in d3 document and thus will have high tf and high idf (since it is inverse). Whereas, "he will have a high tf, but a low idf that will exit tf and give a low score.
MYSQL's natural language mode also comes with a set of stop words (a, a, some, etc.) and deletes words that are less than 4 letters. This can be seen in the link you provided.
Some words are ignored in full-text search:
Any word that is too short is ignored. The default minimum length of words that are found by full-text searches is three characters for
InnoDB search indexes or four characters for MyISAM. You can control circumcision by setting a configuration parameter before creating index: the innodb_ft_min_token_size configuration parameter for finding InnoDB indexes or ft_min_word_len for MyISAM.
Words in the stopword list are ignored. A stopword is a word such as "the" or "some" that is so common that it is considered to have
zero semantic value. There is a built-in list of notes, but it can be overridden by a custom list. Note lists and related configuration options are different for InnoDB and MyISAM search indexes. Interrupt handling is controlled by the configuration of options innodb_ft_enable_stopword, innodb_ft_server_stopword_table, and innodb_ft_user_stopword_table for InnoDB and ft_stopword_file search indexes for MyISAM.