Customer support entails multi-faceted benefits for IT businesses. Presently, the business depends upon on conventional channels like e-mail, customer care and web interface to provide customer sup ...

Server-Less Rule-Based Chatbot Using Deep Neural Network

Membangun sebuah chatbot tidak lepas dari basis pengetahuan. Basis pengetahuan bisa diperoleh dari data yang telah diberi label oleh pengembang, dokumen yang telah di konversi menjadi pre-processing data, atau informasi yang diambil dari media sosial. Dalam hal ini, data yang digunakan sebagai pengetahuan adalah riwayat obrolan. Dialam riwayat obrolan tersebut tentu banyak sekali variasi jawaban dan menungkinkan satu pertanyaan menimbulkan banyak jawaban. Untuk mengatasi multi respon tersebut maka dibuatlah pemeringaktan respon. Adanya pemeringkatan, tentu respon yang diinginkan oleh pengguna akan lebih sesuai. Tantangan dalam pemeringkatan adalah bagaimana mendapatkan inti dari sebuah pertanyaan dan menemukan pasangan tanya jawab dari data. Hal tersebut bisa diselesaikan dengan model sequence to sequence. Namun, permasalahan yang akan muncul adalah konsistensi jawaban. Adanya banyak riwayat obrolan dari layanan travel tentu menimbulkan banyak jawaban mesi inti dari pertanyaan adalah sama. Untuk itu Algoritma CNN sebagai solusi untuk permasalahan tersebut. Penelitian ini menggunakan convolutional sequence to sequence yang akan diterapkan untuk respon ranking. Kami memandingkan efisiensi dari model ini. Dengan melakukan perbandingan, model ini menunjukan hasil akurasi sebesar 86.7%

/pdf/multi-respon-ranking-pada-percakapan-layanan-travel-3pkj0ape4j.pdf

Multi Respon Ranking Pada Percakapan Layanan Travel Berdasarkan Riwayat Obrolan

Most prospective new students want to get fast and accurate information. Prospective new students certainly want to get a lot of information about the procedure for registering new students at each tertiary institution, especially Indo Global Mandiri University. Information on new student admissions at Indo Global Mandiri University only uses websites and social pages as general information. To overcome this problem, a new student acceptance chatbot application was created at Indo Global Mandiri University to provide information that can summarize well and display the information to users. To create this chatbot using the Python programming language using Deep learning. The deep learning algorithm model that is applied in making chatbots uses the Artificial Neural Network model. The application of the Artificial Neural Network model can recognize question patterns quickly and accurately so as to get an appropriate response. The model was then tested with 15 different conversations and successfully answered with a chatbot accuracy value of 86% and 13% error

http://ejournal.uigm.ac.id/index.php/JINIG/article/download/3073/1896

Aplikasi chatbot untuk penerimaan mahasiswa baru universitas indo global mandiri menggunakan deep learning

The new capital city (IKN) of the Republic of Indonesia has been ratified and inaugurated by President Joko Widodo since January 2022. Unfortunately, there are still many Indonesian citizens who do not understand all the information regarding the determination of the new capital city. Even though the Indonesian Government has created an official website regarding the new capital city (www.ikn.go.id) the information is still not optimal because web page visitors are still unable to interact actively with the required information. Therefore, the development of the Chatting Robot (Chatbot) application is deemed necessary to become an interactive component in obtaining information needed by users related to new capital city. In this study, a chatbot application was developed by applying Natural Language Processing (NLP) using the Term Frequency-Inverse Document Frequency (TF-IDF) method for term weighting and the Cosine-Similarity algorithm to calculate the similarity of the questions asked by the user. The research successfully designed and developed a chatbot application using the Cosine-Similarity algorithm. The testing phase of the chatbot model uses several scenarios related to the points of NLP implementation. The test results show that all scenarios of questions asked can be responded well by the chatbot.

Application of Natural Language Processing Using Cosine-Similarity Algorithm in Making Chatbot Information on the New Capital City of the Republic of Indonesia

The level of children’s health is one of the problems that the Indonesian government pays attention to, especially in the health sector. Data from the Indonesian Demographic and Health Survey (IDHS) in 2017 showed that the infant mortality rate (IMR) reached 24 deaths out of 1000 live births. Diseases and health problems that cause child mortality such as perinatal problems, infectious disorders, and malnutrition problems are difficult problems to handle at the health care center level in remote areas due to the lack of diagnostic equipment and medicines. Along with the rapid development of technology today, there are many applications that function to simplify the process of managing and finding information, one of which is chatbot technology. Like a pediatrician, later the chatbot will diagnose the disease from a series of questions submitted by the child’s parents. This final project research aims to develop an android-based chatbot application to help identify diseases in children using a back-propagation artificial neural network. The data used in this research is based on the Integrated Management of Sick Toddler (IMCI) chart book. Based on the results obtained in this final project, the best BPNN model for children’s disease problems in a chatbot application with the number of neurons (128, 64), epochs as many as 800, dropout rate 0.5, optimizer Adam, getting training loss values of 0.11, training accuracy of 96%, validation loss of 1.05 and has a validation accuracy of 64%.

Android-Based Chatbot Application Using Back Propagation Neural Network to Help the First Treatment of Children’s Diseases

Kemajuan teknologi membuat banyak otomatisasi dalam dunia industri, Salah satunya penerapan Chatbot pada industri pendidikan. Dengan teknologi otomatisasi ini, Memudahkan universitas untuk melayani mahasiswa kapanpun waktunya dan dimanapun tempatnya. Kurangnya literasi mahasiswa terhadap fungsi dan penggunaan web kuliah dalam melaksanakan kuliah online menyebabkan banyaknya pertanyaan yang sama secara berulang kepada pihak universitas yang sebenarnya pertanyaan yang sering ditanyakan sudah ditulis dalam daftar pertanyaan yang sering muncul(Faq), seperti: pengumpulan tugas, lupa password , kuliah online, kuliah video conference dan aplikasi web kuliah di android. Dengan menggunakan Chatbot secara otomatis akan menjawab pertanyaan mahasiswa dilaman web kuliah dengan memberikan informasi dan mengarahkan sesuai dengan jawaban pertanyaan. Dalam penelitian ini akan dikembangkan model Chatbot berbasis teks dengan menggunakan algoritma Artificial neural network (ANN). Dataset yang digunakan untuk melakukan pelatihan Chatbot merupakan data pertanyaan sering timbul (Faq) didalam web kuliah yaitu 25 pertanyaan beserta jawaban yang dibagi kedalam 16 label atau kelas. Pengujian dilakukan dengan menggunakan 110 percakapan yang berbeda dengan dataset tetapi mempunyai maksud yang sama. Dari 110 percakapan Chatbot berhasil menjawab dengan tepat sebanyak 107 percakapan dan jawaban salah sebanyak 3 percakapan. Hasil pengujian menunjukkan hasil yang baik yaitu mempunyai akurasi 97,27 % dan kesalahan 2,72 %. Kata Kunci: Chatbot , Artificial Neural Network, ANN, Berbasis teks.

/pdf/algoritma-artificial-neural-network-pada-text-based-chatbot-4wxz7evvg0.pdf

Feri Mustakim

Papers

Algoritma Artificial Neural Network pada Text-based Chatbot Frequently Asked Question (FAQ) Web Kuliah Universitas Nasional