Pemodelan Regresi Teks Data Perhotelan Menggunakan Neural Network Dan Regresi Logistik

Maryana, Indah (2021) Pemodelan Regresi Teks Data Perhotelan Menggunakan Neural Network Dan Regresi Logistik. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

Text 06211740000014-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2023. Download (1MB) | Request a copy

Abstract

Persaingan ketat pada sektor perhotelan semenjak pandemi secara tidak langsung menuntut hotel untuk lebih memuaskan pengunjung. Oleh karena itu, perlu dimanfaatkan data ulasan yang kini tersedia pada situs seperti TripAdvisor untuk mendalami perspektif pengunjung. Perspektif pengunjung berguna untuk mengetahui kelebihan hotel, kekurangan hotel, serta faktor-faktor apa saja yang memengaruhi kepuasan dan ketidakpuasan pengunjung. Pada penelitian ini, hotel yang diambil ulasannya ialah YELLO Hotel Harmoni, HARRIS Vertu Hotel Harmoni, dan Aone Hotel. Data ulasan dari ketiga hotel ini diberlakukan text preprocessing dan pembobotan kata sedemikian hingga datanya siap untuk diterapkan analisis sentimen, topic modeling, dan regresi teks. Hasil sentimen menggunakan TextBlob berhasil mengkategorikan ulasan ke dalam kelas positif dan negatif, di mana setelah itu dilakukan topic modeling menggunakan Latent Semantic Analysis pada masing-masing ulasan positif dan negatif. Kelompok topik yang dihasilkan sejumlah 3 topik positif dan 4 topik negatif. Total sebanyak 7 topik dimodelkan dengan variabel independen rekomendasi editor, kategori hotel, dan tujuan pengunjung menggunakan neural network dan regresi logistik. Hasil yang didapatkan adalah variabel dependen topik positif, cenderung lebih baik menggunakan model neural network. Sebaliknya yang berupa topik negatif, cenderung lebih baik dimodelkan dengan regresi logistik apabila melihat dari nilai akurasi, nilai F1, dan hasil uji kesesuaian model regresi logistik.
====================================================================================================
Strict rivalry in the hospitality sector since the pandemic has indirectly demanded hotels to satisfy visitors more. Therefore, it is necessary to take advantage of online reviews that are now available on sites such as TripAdvisor to deepen the perspective of visitors. The visitor's perspective is useful to know the advantages of the hotel, the disadvantages of the hotel, and factors that affect the satisfaction and dissatisfaction of visitors. In this study, the hotels that were used were YELLO Hotel Harmoni, HARRIS Vertu Hotel Harmoni, and Aone Hotel. The reviews from these hotels are subjected to text preprocessing and term weighting so that the data is ready to be applied to sentiment analysis, topic modeling, and text regression. Sentiment results using TextBlob managed to categorize reviews into positive and negative classes, then topic modeling was carried out using Latent Semantic Analysis on each positive and negative reviews. The resulting group of topics are 3 positive topics and 4 negative topics. A total of 7 topics were modeled with the independent variables of editor's recommendation, hotel category, and visitor’s purpose using neural network and logistic regression. The results obtained are positive topic dependent variables, tend to be better using the neural network model. On the other hand, negative topics tend to be better modeled with logistic regression when looking at the metrics of accuracy, F1 score, and the result of the goodness of fit test of logistic regression model.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	analisis sentimen, regresi teks, topic modeling, sentiment analysis, text regression, topic modeling.
Subjects:	Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines. Q Science > Q Science (General) > Q325.78 Back propagation Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Indah Maryana
Date Deposited:	13 Sep 2021 01:36
Last Modified:	13 Sep 2021 01:36
URI:	http://repository.its.ac.id/id/eprint/91970

Actions (login required)

View Item