ANALISIS KOMPARASI PEMODELAN TOPIK LATENT DIRICHLET ALLOCATION DAN LATENT SEMANTIC ANALYSIS PADA ULASAN RESTORAN DI YOGYAKARTA

Ulfa Mulya, NIM.:16650028 (2020) ANALISIS KOMPARASI PEMODELAN TOPIK LATENT DIRICHLET ALLOCATION DAN LATENT SEMANTIC ANALYSIS PADA ULASAN RESTORAN DI YOGYAKARTA. Skripsi thesis, UIN SUNAN KALIJAGA YOGYAKARTA.

[img]
Preview
Text (ANALISIS KOMPARASI PEMODELAN TOPIK LATENT DIRICHLET ALLOCATION DAN LATENT SEMANTIC ANALYSIS PADA ULASAN RESTORAN DI YOGYAKARTA)
16650028_BAB I_BAB PENUTUP_DAFTAR PUSTAKA.pdf

Download (1MB) | Preview
[img] Text (ANALISIS KOMPARASI PEMODELAN TOPIK LATENT DIRICHLET ALLOCATION DAN LATENT SEMANTIC ANALYSIS PADA ULASAN RESTORAN DI YOGYAKARTA)
16650028_BAB II SAMPAI_BAB IV.pdf
Restricted to Registered users only

Download (1MB) | Request a copy

Abstract

Restaurants are the most promising business prospects in Yogyakarta. To find out things that are of concern to the community, a survey is needed. As for one way that can be done is to take advantage of content reviews from Google Maps users. Therefore, research is carried out to save costs, time and energy. This research aims to find hidden topic trends in a collection of restaurant reviews automatically to facilitate reading and understanding data using topic modeling methods namely Latent Dirichlet Allocation (LDA) and Latent Semantic Analysis (LSA). Both methods were tested with 1010 review data to find the output of the best topic model that represents human interpretation. The results of the evaluation by measuring Cv coherence produce a value of 0.552396 with an output model of 7 topics on the LDA algorithm and produce a value of 0.55132 with an output model of 4 topics on the LSA algorithm. In this research the LDA algorithm is proven to work better, and 7 topics become the best topics, i.e. “harga”, “suasana”, “menu”, “unik”, “lesehan”, “instagramable”, “nongkrong”.

Item Type: Thesis (Skripsi)
Additional Information: Pembimbing: M. Didik Rohmad Wahyudi, S.T., MT.
Uncontrolled Keywords: topic koherence; text mining; google maps; machine learning
Subjects: Tehnik Informatika
Sains
Divisions: Fakultas Sains dan Teknologi > Teknik Informatika (S1)
Depositing User: Muchti Nurhidaya edt
Date Deposited: 19 Apr 2022 11:39
Last Modified: 19 Apr 2022 11:39
URI: http://digilib.uin-suka.ac.id/id/eprint/50297

Share this knowledge with your friends :

Actions (login required)

View Item View Item
Chat Kak Imum