ANALISIS SIMILARITAS LOWONGAN PEKERJAAN DENGAN PROFIL PELAMAR MENGGUNAKAN ALGORITMA TF-IDF DAN COSINE SIMILARITY

Hidayatulloh, Akhmad Nur (2023) ANALISIS SIMILARITAS LOWONGAN PEKERJAAN DENGAN PROFIL PELAMAR MENGGUNAKAN ALGORITMA TF-IDF DAN COSINE SIMILARITY. Undergraduate thesis, Politeknik Negeri Jember.

[img] Text (ABSTRACT)
ABSTRACT.pdf - Submitted Version
Available under License Creative Commons Attribution Share Alike.

Download (480kB)
[img] Text (BAB 1)
BAB 1.pdf - Submitted Version
Available under License Creative Commons Attribution Share Alike.

Download (630kB)
[img] Text (DAFTAR PUSTAKA)
DAFTAR PUSTAKA.pdf - Submitted Version
Available under License Creative Commons Attribution Share Alike.

Download (699kB)
[img] Text (LAPORAN LENGKAP)
LAPORAN LENGKAP - TANPA LAMPIRAN.pdf
Restricted to Registered users only

Download (4MB) | Request a copy

Abstract

Talent Acquisition (TA) is a type of job that handles the employee recruitment process. The conventional method will take quite a long time, because there are relatively many job candidates taken in one recruitment. Based on the above problems, this research will analyze the process to find the level of similarity between job vacancies and job applicant profiles using the TF-IDF and Cosine Similarity algorithms. In this research, 5 job vacancies in Jobstreet were taken, each with different categories. These categories include Backend Developer, Frontend Developer, UI/UX Designer, Android Developer, and DevOps Engineer. Job applicant data in this study was collected as a total of 100 Linkedin account profiles and divided into 20 accounts for each category. This research obtained the highest results in the second experiment with a threshold of the top 20 ranks, the results were an average precision of 54%, recall 54% and F1-score of 54%. The level of similarity is strongly affected by the length of the data and content in the document, as the quality of the corpus used in the filtering process. The better of the corpus quality, the similarity results significantly increased.

Item Type: Thesis (Undergraduate)
Contributors:
ContributionContributorsNIDN/NIDK
Thesis advisorGumilang, Mukhamad AnggaNIDN0012089401
Uncontrolled Keywords: Jobstreet, Linkedin, Job Vacancy, Applicant, Cosine Similarity
Subjects: 410 - Rumpun Ilmu Teknik > 450 - Teknik Elektro dan Informatika > 458 - Teknik Informatika
Divisions: Jurusan Teknologi Informasi > Prodi D4 Teknik Informatika > Tugas Akhir
Depositing User: Akhmad Nur Hidayatulloh
Date Deposited: 31 Jul 2023 09:05
Last Modified: 31 Jul 2023 09:07
URI: https://sipora.polije.ac.id/id/eprint/27181

Actions (login required)

View Item View Item