Hasil Pencarian  ::  Simpan CSV :: Kembali

"Penelitian Part-of-Speech tagger (POS tagger) untuk bahasa Indonesia telah banyak dikembangkan. Sayangnya, sejauh ini baru Polyglot yang menggunakan POS tag menurut pedoman anotasi Universal Dependencies (UD). Namun, Polyglot sendiri masih mempunyai kekurangan karena belum dapat mengatasi klitik dan kata ulang yang terdapat dalam bahasa Indonesia. Tujuan penelitian ini adalah mengembangkan POS tagger untuk bahasa Indonesia yang tidak hanya sesuai dengan ketentuan anotasi UD, tapi juga sudah mengatasi kekurangan Polyglot. POS tagger ini akan dikembangkan dengan metode deep learning menggunakan arsitektur yang merupakan versi modifikasi dari Recurrent Neural Network (RNN), yaitu Bidirectional Long Short-Term Memory (Bi-LSTM). Dataset yang digunakan untuk mengembangkan POS tagger adalah sebuah dependency treebank bahasa Indonesia yang terdiri dari 1.000 kalimat dan 19.401 token. Hasil eksperimen dengan menggunakan Polyglot sebagai pembanding menunjukkan bahwa POS tagger yang dikembangkan lebih baik dengan tingkat akurasi POS tagging yang meningkat sebesar 6,69% dari 84,82% menjadi 91,51%.

---

There have been many studies that have developed Part-of-Speech tagger (POS tagger) for Indonesian language. Unfortunately, so far only Polyglot that has used POS tag according to Universal Dependencies (UD) annotation guidelines. However, Polyglot itself still has shortcomings since it has not been able to overcome clitics and reduplicated words in Indonesian language. The purpose of this study is to develop POS tagger for Indonesian language which is not only in accordance with UD annotation guidelines, but also has overcome Polyglot’s shortcomings. This POS tagger will be developed under deep learning method by using modified version of Recurrent Neural Network (RNN) architecture, Bidirectional Long Short-Term Memory (Bi-LSTM). The dataset used to develop POS tagger is an Indonesian dependency treebank consisting of 1.000 sentences and 19.401 tokens. Result of experiment using Polyglot as baseline shows that the developed POS tagger is better. This is indicated by increased accuracy POS tagging by 6,69% from 84,82% to 91,51%."

"This study aims to analyze the determinant factors that affect issues development within APEC, map out those issues during the period 1993-2010, and show the relation of those issues with the APEC Summit Agenda 2013 in Indonesia. The analysis is based on secondary data, literature review of APEC meeting documents, interviews, and focus group discussions. Some interesting findings suggest that, lirstly, issues development in APEC has been shaped by responses of APEC to opportunities and challenges related to economic, social, political and security conditions within APEC and the world. It is not only government agencies that are involved in issues development but other agents as well, such as the Pacific Economic Cooperation Council, the Association Southeast Asian Nations, the World Trade Organization, APEC Business Advisory Council, and APEC Study Centers Consortium. In the past and at present, the Eminent Persons Group and the Paciiic Business Forum, which were set up for a specific time by APEC, continue to play vital and influential roles. Secondly, this study iinds that there are four big groups involved in issues development in APEC. All issues are part of the development issues in APEC economies. Even though the issues are very broad, encompassing economic and non-economic matters, these are nonetheless focused on economic integration of APEC, with Bogor Goals being in the nucleus of issues. The development of the range of issues, which APEC has pursued to respond to challenges and opportunities in the APEC economies, is intended to support and secure economic integration. Thirdly, the Indonesian APEC Summit Agenda 2013 emphasized three specific agenda items: attaining the Bogor Goals, sustainable and inclusive growth, and connectivity. All these are inter-related issues of developments that have been discussed since the establishment of APEC."

"The southern coast of the North Sea is mainly protected by a single line of defence against flooding. The traditional means for heightening and strengthening the existing coastal barriers are limited due to the increased enormous resource costs, i.e. building materials, funding and space. On the other hand, further interests and needs such as tourism, nature conservation and agriculture demand an integrated planning process according to the principle of a sustainable coastal development. The spatial protection concepts discussed in the book provide new options for sustainable coastal development in line with the integration of several types of land use. Furthermore, the multifunctional coastal protection zones offer new opportunities for the application of different options for coastal protection, i.e. holding the line, accommodation or retreat. This book also provides a sound basis for an integrated planning process in coastal zones combining participatory methods with adapted --tools from socio-economic and ecologic evaluation"

"Sistem Penilaian Esai Otomatis (SIMPLE-O) dikembangkan oleh Departemen Teknik Elektro Fakultas Teknik Universitas Indonesia untuk ujian bahasa Indonesia. Skripsi ini akan membahas mengenai pengembangan SIMPLE-O untuk penilaian ujian bahasa Indonesia menggunakan metode Siamese Manhattan Long Short-Term Memory (LSTM) dan bahasa pemrograman Python. Terdapat dua dokumen yang akan menjadi input, yaitu jawaban esai dari peserta ujian dan jawaban referensi dari penguji. Kedua jawaban diproses dengan layer LSTM yang sama. Selanjutnya, kemiripan antara keduanya dihitung dengan fungsi persamaan. Pengujian dengan dataset jawaban dummy mendapatkan nilai MAE dan RMSE sebesar 0.0254 dan 0.0346. Kemudia, pengujian dengan dataset jawaban asli mendapatkan nilai MAE dan RMSE terbaik sebesar 0.1596 dan 0.2190. Rata-rata nilai akurasi yang didapatkan adalah 92.82 untuk fase training dan 84.03 untuk validasi.

---

The Automatic Essay Assessment System (SIMPLE-O) was developed by the Department of Electrical Engineering, Faculty of Engineering, University of Indonesia for the Indonesian language test. This thesis will discuss the development of SIMPLE-O for the assessment of Indonesian language tests using the Siamese Manhattan Long Short-Term Memory (LSTM) method and the Python programming language. There are two documents that will be input, essay answers from examinees and answer answers from examiners. Both answers are processed with the same LSTM layer. Next, the similarity between the two is calculated by the similarity function. Testing with dummy answer dataset produces MAE and RMSE values of 0.0254 and 0.0346. Then, testing with the real answer dataset produces MAE and RMSE values of 0.1596 and 0.2190. The average accuracy value obtained was 92.82 for the training phase and 84.03 for validation.

"

"Machine Learning (ML) dan Deep Learning merupakan bidang yang populer pada masa kini. Salah satu ranah tersebut yang menantang untuk diteliti adalah tentang mendeteksi emosi pada teks. Interaksi antara komputer dan manusia dapat menjadi lebih baik apabila komputer dapat mendeteksi emosi, menginterpretasikan emosi tersebut, dan memberikan umpan balik yang sesuai dengan apa yang manusia inginkan. Oleh karena itu, penelitian ini bertujuan untuk membuat sistem pendeteksi emosi pada teks Bahasa Indonesia. Pada penelitian ini, terdapat 2 macam algoritma Deep Learning yang digunakan, yaitu Convolutional Neural Network (CNN) dan Long Short-Term Memory (LSTM). Convolutional Neural Network merupakan salah satu algoritma Deep Learning dimana karakteristik utamanya menggunakan operasi matriks konvolusi. Long ShortTerm Memory merupakan salah satu algoritma Deep Learning dimana merupakan perkembangan dari algoritma Recurrent Neural Network (RNN). Kedua algoritma tersebut akan didukung dengan Word Embedding Bahasa Indonesia dari fastText dan Polyglot. Package text2emotion akan digunakan sebagai data tambahan untuk evaluasi. Input dataset yang digunakan untuk Deep Learning adalah dataset cerita dongeng yang memiliki emosi "Senang", "Sedih", "Marah", "Takut", "Terkejut", dan "Jijik". Input dataset tersebut akan melalui tahap preprocessing berupa Case Normalization, Stopword Removal, Stemming, Tokenizer, dan Padding. Setelah itu, proses training dijalankan dengan menggunakan RandomizedSearchCV sebagai hyperparameter tuning. Hasil akan dibandingkan dan dianalisis berdasarkan nilai Evaluation Metrics Accuracy, Precision, Recall, dan F1-Score. Sistem berhasil dirancang dengan mencapai hasil Accuracy sebesar 91,60%, Precision sebesar 92,48%, Recall sebesar 91,60%, dan F1- Score sebesar 91,68%.

---

Machine Learning (ML) and Deep Learning is a popular region to be used right now. One of the scopes that challenging to research is about emotion recognition on text. Interaction between computer and human can be better if the computer can recognize the emotion, interpret it, and giving a suitable feedback with the human’s need. Therefore, this research has goal to make an emotion recognition on Indonesian text language. On this research, there’s 2 kind of Deep Learning algorithm that used, that is Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM). Convolutional Neural Network is one of Deep Learning algorithm that its main characteristic is using convolution matrix operation. Long Short-Term Memory is one of Deep Learning algorithm which is an improvement from Recurrent Neural Network (RNN) algorithm. Both algorithms will be supported with Indonesian Word Embedding from fastText and Polyglot. Text2emotion package is used for additional data for evaluation. The input dataset that will be used on this Deep Learning is a fairy tale dataset which have “Happy”, “Sad”, “Anger”, “Fear”, “Surprised”, and “Disgust” emotion. That input dataset will be passed to preprocessing stage that consist of Case Normalization, Stop-word Removal, Stemming, Tokenizer, and Padding. After that, training process started with using RandomizedSearchCV as hyperparameter tuning. The result will be compared and analyzed based on Accuracy, Precision, Recall, and F1- Score Evaluation Metrics. System is made with reaching 91.60% Accuracy, 92,48% Precision, 91,60% Recall, and 91,68% F1-Score."