Permana, Satya Wira (2023) Perbandingan metode Lexicon-Based dan Support Vector Machine untuk analisis sentimen pinjaman online pada media sosial Twitter. Sarjana thesis, UIN Sunan Gunung Djati Bandung.
|
Text (COVER)
1_cover.pdf Download (87kB) | Preview |
|
|
Text (ABSTRAK)
2_abstrak.pdf Download (66kB) | Preview |
|
|
Text (DAFTAR ISI)
3_daftarisi.pdf Download (2MB) | Preview |
|
|
Text (BAB I)
4_bab1.pdf Download (374kB) | Preview |
|
Text (BAB II)
5_bab2.pdf Restricted to Registered users only Download (355kB) | Request a copy |
||
Text (BAB III)
6_bab3.pdf Restricted to Registered users only Download (325kB) | Request a copy |
||
Text (BAB IV)
7_bab4.pdf Restricted to Registered users only Download (2MB) | Request a copy |
||
Text (BAB V)
8_bab5.pdf Restricted to Registered users only Download (134kB) | Request a copy |
||
Text (DAFTAR PUSTAKA)
9_daftarpustaka.pdf Restricted to Registered users only Download (144kB) | Request a copy |
Abstract
INDONESIA : Berkembangnya pinjaman online memunculkan banyak reaksi, pendapat, perbincangan, dan perdebatan masyarakat yang diungkapkan di berbagai media sosial, salah satunya Twitter. Berbagai hal yang disampaikan melalui Tweet Twitter dapat dimanfaatkan untuk mengetahui bagaimana pandangan pengguna media sosial tersebut terhadap suatu isu atau permasalahan melalui kegiatan analisis sentimen atau klasifikasi. Algoritma Lexicon-Based dan Support Vector Machine (SVM) digunakan dengan pendekatan metodologi CRISP-DM untuk proses data mining. Hasil pengujian model menggunakan 610 data uji menghasilkan tingkat akurasi analisis sentimen Pinjaman Online pada media sosial Twitter menggunakan Lexicon-Based sebesar 65% dan SVM sebesar 85%. Hasil prediksi metode SVM menggunakan 9.710 data tweet menghasilkan 5.255 data tweet hasil pembersihan. sebanyak 2.885 data tweet bersifat netral, sedangkan tweet bersifat positif sebanyak 1.190 data, dan tweet bersifat negatif sebanyak 1.180 data, serta 4.455 data bersifat duplikasi dan tweet kosong. Dengan hasil tersebut didapatkan kecenderungan dari masyarakat Indonesia pengguna media sosial twitter memiliki pandangan atau opini yang netral terhadap pinjaman online. Hasil perbandingan dua metode berdasarkan perhitungan performa metrik menujukkan metode SVM memiliki performa yang lebih baik bila dibandingkan dengan metode Lexicon-Based, dengan nilai presisi tertinggi pada metode SVM kernel linear untuk kelas positif sebesar 92%. ENGLISH : The development of online lending has led to many reactions, opinions, discussions and public debates expressed on various social media, one of which is Twitter. Various things conveyed through Twitter Tweets can be used to find out how social media users view an issue or problem through sentiment analysis or classification activities. Lexicon-Based Algorithms and Support Vector Machine (SVM) are used with the CRISP-DM methodology approach for the data mining process. The results of testing the model using 610 test data produce an accuracy level of sentiment analysis of Online Loans on social media Twitter using Lexicon-Based by 65% and SVM by 85%. The predicted results of the SVM method use 9,710 tweet data to produce 5,255 clean tweet data. 2,885 tweets are neutral, while 1,190 data are positive tweets, and 1,180 data are negative tweets, and 4,455 data are duplicates and tweets are empty. With these results, it was found that the tendency of Indonesian people to use Twitter social media has views or opinions that are neutral towards online loans. The results of a comparison of the two methods based on performance metric calculations show that the SVM method has better performance when compared to the Lexicon-Based method, with the highest precision value in the linear kernel SVM method for the positive class of 92%.
Item Type: | Thesis (Sarjana) |
---|---|
Uncontrolled Keywords: | Analisis Sentimen; Twitter; Lexicon-Based; SVM; Pinjaman Online; CRISP-DM |
Subjects: | Data Processing, Computer Science |
Divisions: | Fakultas Sains dan Teknologi > Program Studi Teknik Informatika |
Depositing User: | Satya Wira Permana |
Date Deposited: | 15 Feb 2023 06:01 |
Last Modified: | 15 Feb 2023 06:01 |
URI: | https://digilib.uinsgd.ac.id/id/eprint/64524 |
Actions (login required)
View Item |