MODEL EVALUASI KINERJA DOSEN PADA PROSES BELAJAR MENGAJAR MENGGUNAKAN OPINION MINING BERBASIS EMBEDDING TEXT-TO-SEQUENCE

PURBA, MARIANA and Ermatita, Ermatita and Abdiansah, Abdiansah (2023) MODEL EVALUASI KINERJA DOSEN PADA PROSES BELAJAR MENGAJAR MENGGUNAKAN OPINION MINING BERBASIS EMBEDDING TEXT-TO-SEQUENCE. Doctoral thesis, Sriwijaya University.

	Text RAMA_21001_03013682126010.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (10MB) | Request a copy
	Text RAMA_21001_03013682126010_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (15MB) | Request a copy
	Text RAMA_21001_03013682126010_0013096707_0001108401_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (1MB)
	Text RAMA_21001_03013682126010_0013096707_0001108401_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_21001_03013682126010_0013096707_0001108401_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (600kB) | Request a copy
	Text RAMA_21001_03013682126010_0013096707_0001108401_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (549kB) | Request a copy
	Text RAMA_21001_03013682126010_0013096707_0001108401_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (681kB) | Request a copy
	Text RAMA_21001_03013682126010_0013096707_0001108401_06.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (346kB) | Request a copy
	Text RAMA_21001_03013682126010_0013096707_0001108401_07_ref.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (387kB) | Request a copy
	Text RAMA_21001_03013682126010_0013096707_0001108401_08_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (8MB) | Request a copy

Abstract

To support the evaluation of the teaching and learning process in higher education institutions, it is necessary to build an opinion mining. This research will focus on how to conduct opinion mining on feedback on the learning process at private universities and develop appropriate optimization models to support the processing of the dataset. This research has five main steps, including literature study, dataset collection, opinion mining model development, and publication writing. Data retrieval is performed in Universitas Sjakhyakirti, Institut Teknologi dan Bisnis Palcomtech, Universitas Muhammadiyah Palembang, Universitas Bina Darma, AMIK Bina Sriwijaya and Politeknik Darusalam. The development of the questionnaire resulted in a learning process evaluation for higher educational institution (Learn-EV) consisting of one question item for each evaluation of pedagogic competence, professional competence, personality competence and social competence. The composition of datasets with detailed data with positive labels amounted to 9,104, data with negative labels amounted to 12,764 and data with neutral labels amounted to 2,788. With a dataset division of 16,519 for training data and 8137 for test data. The performance of the TELSTMLR model from the results of optimization of long short-term memory (LSTM) using text-to-sequence (T2S) and embedding layer (EL) using the leaky-relu (LReLU) function to classify datasets consisting of positive labels totaling 9,104, data with negative labels totaling 12,764 and data with neutral labels totaling 2,788 had the best performance among the models conducted experimentally, namely 86.14%. The second-best method is TELSTM to get an accuracy value of 85.64%.

Item Type:	Thesis (Doctoral)
Uncontrolled Keywords:	knowledge sharing, opinion mining, private university, LSTM, text-to-sequence, LReLU
Subjects:	T Technology > T Technology (General) > T58.6-58.62 Management information systems > T58.6 General works Industrial engineering Information technology. Information systems (General) Management information systems -- Continued
Divisions:	03-Faculty of Engineering > 21001-Engineering Science (S3)
Depositing User:	MARIANA PURBA
Date Deposited:	28 Nov 2023 05:20
Last Modified:	28 Nov 2023 05:20
URI:	http://repository.unsri.ac.id/id/eprint/131307

Actions (login required)

View Item