Wijayanto, Riko (2019) Implementasi Social Media Mining untuk Pengambilan Keputusan dalam Perencanaan Produk berdasarkan Pemodelan Topik dan Analisis Sentimen pada Twitter. Other thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
06111540000010-Undergraduate_Theses.pdf Download (4MB) | Preview |
Abstract
Perkembangan teknologi informasi dan komunikasi, khususnya dalam bidang media sosial, membawa suatu perubahan besar pada bidang perencanaan produk dari suatu perusahaan yang menaungi. Saat ini, perusahaan dapat mengidentifikasi peluang untuk perencanaan produk dari opini pelanggan yang disampaikan melalui posting Twitter. Peluang ini dapat dimanfaatkan untuk meningkatkan daya saing terhadap kompetitor. Metode utama yang digunakan adalah social media mining melalui pemodelan topik dan analisis sentimen. Topik yang sedang ramai dibicarakan pelanggan diidentikkan dengan derajat importance, sedangkan hasil analisis sentimen diidentikkan dengan derajat satisfaction. Pemodelan topik dengan Latent Semantic Analysis dan K-Means serta analisis sentimen berdasarkan daftar lexicon dari C. J. Hutto dan Eric Gilbert berhasil diterapkan dengan baik serta dikemas dalam sebuah perangkat lunak berbasis bahasa pemrograman Python. Studi kasus pada McDonald’s dengan memanfaatkan 2000 data tweet, didapatkan suatu hasil bahwa “pelayanan di pagi hari” memiliki opportunity tertinggi dan berada dalam wilayah under-served. Ini memberikan arti bahwa “pelayanan di pagi hari” merupakan salah satu layanan yang memiliki peluang bagus di pasar. Dengan demikian, McDonald’s dapat mempertimbangkan perencanaan pelayanan di pagi hari.
================================================================================================================================
The development of information and communication technology, especially in the social media, brought an important change of product planning from an overshadowed company. Recently, using the the text mining and social media mining, the companies can identify opportunities for product planning based on customer’s opinion from Twitter’s postings to increase their competitiveness with competitors. The hotest topic and the customers’ satisfaction can be identified by topic modelling and sentiment analysis. The hotest topic is acquisied as the importance degree and the result of sentiment analysis is acquisied as the satisfaction degree. Topic modelling with Latent Semantic Analysis and K-Means cluster, also sentiment analysis based on lexicon list from C. J. Hutto and Eric Gilbert, succesfully implemented and packaged in a build-based using Python programming language. Case studies at McDonald’s by utilizing 2000 tweets, a result was obtained that “operational in morning day” had the highest opportunity value in under-served area. It means that “operational in morning day” has a nice opportunity at the market. As such, McDonald’s can consider following up on the proposes thing that is “operational in morning day”.
| Item Type: | Thesis (Other) |
|---|---|
| Additional Information: | RSMa 006.312 Wij i-1 2019 |
| Uncontrolled Keywords: | Social Media Mining, Pemodelan Topik, Latent Semantic Analysis (LSA), K-Means, Analisis Sentimen, Opportunity Mining |
| Subjects: | Q Science > QA Mathematics > QA278.55 Cluster analysis Q Science > QA Mathematics > QA336 Artificial Intelligence Q Science > QA Mathematics > QA76.9.D343 Data mining. Querying (Computer science) Q Science > QA Mathematics > QA76.9.I52 Information visualization |
| Divisions: | Faculty of Mathematics, Computation, and Data Science > Mathematics > 44201-(S1) Undergraduate Thesis |
| Depositing User: | Riko Wijayanto |
| Date Deposited: | 08 Jun 2023 02:41 |
| Last Modified: | 08 Jun 2023 02:41 |
| URI: | http://repository.its.ac.id/id/eprint/64065 |
Actions (login required)
 |
View Item |