IMPLEMENTASI MODEL NATURAL LANGUAGE PROCESSING (NLP) PADA SISTEM REKOMENDASI PEKERJAAN MENGGUNAKAN CONVOLUTIONAL NEURAL NETWORKS ( CNN) DAN LONG SHORT TERM MEMORY (LSTM)

FIKRI, MHD. HAIDIR and Sutarno, Sutarno (2024) IMPLEMENTASI MODEL NATURAL LANGUAGE PROCESSING (NLP) PADA SISTEM REKOMENDASI PEKERJAAN MENGGUNAKAN CONVOLUTIONAL NEURAL NETWORKS ( CNN) DAN LONG SHORT TERM MEMORY (LSTM). Undergraduate thesis, Sriwijaya University.

	Text RAMA_56201_09011282025064.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (3MB) | Request a copy
	Text RAMA_56201_09011282025064_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (9MB) | Request a copy
	Text RAMA_56201_09011282025064_0201117802_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (1MB)
	Text RAMA_56201_09011282025064_0201117802_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (853kB) | Request a copy
	Text RAMA_56201_09011282025064_0201117802_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_56201_09011282025064_0201117802_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_56201_09011282025064_0201117802_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (415kB) | Request a copy
	Text RAMA_56201_09011282025064_0201117802_06_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (356kB) | Request a copy
	Text RAMA_56201_09011282025064_0201117802_07_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (668kB) | Request a copy

Abstract

Advances in technology present challenges in matching individual skills with appropriate job recommendations. This research develops a Natural Language Processing (NLP) based job recommendation system using Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) models. The research data is obtained from the Hugging Face and Kaggle public datasets, which include descriptions of work experience and skills. The data was processed through pre-processing stages such as stopword removal, regex, stemming, tokenization, padding, as well as balancing techniques such as SMOTE and undersampling. The dataset is split for model training and testing, with performance measured through accuracy, loss, and validation metrics. Results show that the CNN model on the Hugging face dataset provides the best results, with 95% accuracy, 0.1 loss, 72% validation accuracy, 1.87 validation loss, 79% F1-score, 90% precision, and 72% recall, outperforming LSTM which tends to overfitting. Overall, CNN proved to be more stable and reliable in generalization, especially in handling complex data. This research also confirms that CNN are more effective in processing complex data than LSTM, which tend to be more prone to overfitting, especially on long text data.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Sistem rekomendasi, Natural Language Processing (NLP), CNN, LSTM, Text Classification.
Subjects:	Q Science > Q Science (General) > Q300-390 Cybernetics > Q325.5 Machine learning Q Science > Q Science (General) > Q334-342 Computer science. Artificial intelligence. Algorithms. Robotics. Automation.
Divisions:	09-Faculty of Computer Science > 56201-Computer Systems (S1)
Depositing User:	Mhd. Haidir Fikri
Date Deposited:	09 Jan 2025 08:15
Last Modified:	09 Jan 2025 08:15
URI:	http://repository.unsri.ac.id/id/eprint/162088

Actions (login required)

View Item