<![CDATA[The song is a rhythmic variety of sounds and is one of the entertainment facilities that is popular among today's modern society. The desire of a society to listen to a song can be based on various things such as music (music genres), singers, years of songs to the lyrics of a song, singers, song years to song lyrics. In considering this matter, a system is needed that can facilitate the community in discussing a song based on the basics, one of which is a system of song contributions based on the lyrics. One method that can be used in building a song contribution system is Word Embedding which in this study uses Word2Vec. The stages of the system discussion of songs that contain data sets that produce a collection of words Word2Vec and TF-IDF, then the system will perform the processes of the lyrics of input songs by making expansion requests in accordance with the words words from TF-IDF close words the results of Word2Vec training. The process carried out by the system will produce 10 song titles that have similar lyrics to the input song. Results The aiming of the word parameters with TF-IDF taken and the close word test taken is the highest precision @ 10 value for the highest TF-IDF word taken = 25 and for the close word taken = 3.]]>
Copyrights © 2019
