Hidayat, Firman (2019) Penerapan Algoritma Cosine Similarity Dan Pembobotan TF-IDF Pada Sistem Chatbot Untuk Calon Mahasiswa ITS. Other thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
05211540000149-Undergraduate_Theses.pdf Download (2MB) | Preview |
Abstract
Akhir-akhir ini penggunaan chatbot sedang marak digunakan pada organisasi. Hal ini dikarenakan chatbot merupakan salah satu solusi untuk menjawab pertanyaan yang sering ditanyakan oleh pelanggan, atau frequently asked questions (FAQ) bagi suatu organisasi. Dengan adanya chatbot ini membuat kemampuan CRM pada suatu organisasi berjalan dan bisa digunakan untuk berhubungan dengan customer secara realtime dan aktif selama 24/7. Salah satu cara mengembangkan aplikasi chatbot adalah dengan menggunakan pendekatan information retrieval yaitu menggunakan metode pembobotan Term Frequency – Inverse Document Frequency dan Algoritma Cosine Similarity. Berdasarkan hasil penelitian pada tugas akhir ini, metode pembobotan TF-IDF dan algoritma Cosine Similarity memiliki kemampuan yang baik untuk digunakan dalam aplikasi chatbot karena sistem chatbot dapat menfidentifikasi pertanyaan dengan rata – rata kemiripan 0.662550216 untuk pertanyaan yang belum pernah ditanyakan maupun yang pernah ditanyakan sebelumnya.
=================================================================================================================================
On this day, the use of chatbots is being used in much organizations. This is because chatbots are one of the solution to answering frequently asked questions (FAQs) from customers, for an organization. Chatbot makes CRM capabilities in an organization can running and can be used to connect with customers in real time and be active 24/7. One of ways to develop chatbot applications is using an information retrieval approach that is using the Term Frequency weighting method - Inverse Document Frequency and the Cosine Similarity Algorithm. Based on the results of this research, the weighting method of TF-IDF and the Cosine Similarity algorithm have good ability to be used in chatbot applications because chatbot can specify questions with an average similarity of 0.662550216 for questions that have never been asked whether they have been asked before.
| Item Type: | Thesis (Other) |
|---|---|
| Additional Information: | RSSI 004.019 Hid p-1 2019 |
| Uncontrolled Keywords: | Chatbot, Information retrieval, Term Frequency, Inverse Document Frequency, FAQ, Cosine Similarity |
| Subjects: | T Technology > T Technology (General) > T57.83 Dynamic programming |
| Divisions: | Faculty of Information Technology > Information System > 57201-(S1) Undergraduate Thesis |
| Depositing User: | Firman Hidayat |
| Date Deposited: | 30 May 2024 02:02 |
| Last Modified: | 30 May 2024 02:02 |
| URI: | http://repository.its.ac.id/id/eprint/64618 |
Actions (login required)
 |
View Item |