Institusion
Institut Teknologi Perusahaan Listrik Negara
Author
Caesar, Miko
Praptini, Puji Catur Siswi
Djunaidi, Karina
Subject
Teknik Informatika
Datestamp
2023-06-27 08:25:47
Abstract :
The current condition of students currently does not have a guide in choosing
specialization courses that are suitable for work in the Information Technology (IT) field,
especially students of the IT-PLN Informatics Engineering Study Program. In
determining students, it is necessary to know what job categories are in accordance with
job vacancies that are increasingly rapidly developing. Students also have to adapt to the
needs of the industry and can choose specialization courses that have been provided by
the campus. Therefore, this study analyzes a job vacancy website that is in accordance
with industry needs and Semester Learning Plans (RPS) using the cosine similarity
method to produce similarities between study materials for specialization courses and job
vacancies in the field of information technology. Cosine similarity can be done and get
an average of 7.5% similarity between specialization courses and job vacancies starting
from the stage of collecting data from study materials in the Semester Learning Plan
(RPS), then processed in preprocessing text consisting of case folding, stopword removal,
lematizer, stemming, and tokenizing. followed by analyzing the text, namely the
countvectorizer, TF-IDF data weighting, cosine similarity classification, and k-fold cross
validation data validation. The datasets used are 15 RPS for specialization courses and
the results of the classification of 10 types of work. The results of this test have an
accuracy value of 25% using k-fold cross validation.