I Nyoman Prayana Trisna
Department Of Computer Science And Electronics, FMIPA UGM, Yogyakarta

Published : 3 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 3 Documents
Search

Single document keywords extraction in Bahasa Indonesia using phrase chunking I Nyoman Prayana Trisna; Arif Nurwidyantoro
TELKOMNIKA (Telecommunication Computing Electronics and Control) Vol 18, No 4: August 2020
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/telkomnika.v18i4.14389

Abstract

Keywords help readers to understand the idea of a document quickly. Unfortunately, considerable time and effort are often needed to come up with a good set of keywords manually. This research focused on generating keywords from a document automatically using phrase chunking. Firstly, we collected part of speech patterns from a collection of documents. Secondly, we used those patterns to extract candidate keywords from the abstract and the content of a document. Finally, keywords are selected from the candidates based on the number of words in the keyword phrases and some scenarios involving candidate reduction and sorting. We evaluated the result of each scenario using precision, recall, and F-measure. The experiment results show: i) shorter-phrase keywords with string reduction extracted from the abstract and sorted by frequency provides the highest score, ii) in every proposed scenario, extracting keywords using the abstract always presents a better result, iii) using shorter-phrase patterns in keywords extraction gives better score in comparison to using all phrase patterns, iv) sorting scenarios based on the multiplication of candidate frequencies and the weight of the phrase patterns offer better results.
Hate Speech Detection in Indonesian Twitter using Contextual Embedding Approach Guntur Budi Herwanto; Annisa Maulida Ningtyas; I Gede Mujiyatna; Kurniawan Eka Nugraha; I Nyoman Prayana Trisna
IJCCS (Indonesian Journal of Computing and Cybernetics Systems) Vol 15, No 2 (2021): April
Publisher : IndoCEISS in colaboration with Universitas Gadjah Mada, Indonesia.

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22146/ijccs.64916

Abstract

Hate speech develops along with the rapid development of social media. Hate speech is often issued due to a lack of public awareness of the difference between criticism and statements that might contribute to this crime. Therefore, it is very important to do early detection of sentences that will be written before causing a criminal act due to public ignorance. In this paper, we use the advancement of deep neural networks to predict whether a sentence contains a hate speech and abusive tone. We demonstrate the robustness of different word and contextual embedding to represent the semantic of hate speech words. In addition, we use a document embedding representation via a recurrent neural networks with gated recurrent unit as the main architecture to provide richer representation. Compared to syntactic representation of the previous approach, the contextual embedding in our model proved to give a significant boost on the performance by a significant margin.
Perbandingan penghitungan jarak pada k-nearest neighbour dalam klasifikasi data tekstual Wahyono Wahyono; I Nyoman Prayana Trisna; Sarah Lintang Sariwening; Muhammad Fajar; Danur Wijayanto
Jurnal Teknologi dan Sistem Komputer Volume 8, Issue 1, Year 2020 (January 2020)
Publisher : Department of Computer Engineering, Engineering Faculty, Universitas Diponegoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.14710/jtsiskom.8.1.2020.54-58

Abstract

One algorithm to classify textual data in automatic organizing of documents application is KNN, by changing word representations into vectors. The distance calculation in the KNN algorithm becomes essential in measuring the closeness between data elements. This study compares four distance calculations commonly used in KNN, namely Euclidean, Chebyshev, Manhattan, and Minkowski. The dataset used data from Youtube Eminem’s comments which contain 448 data. This study showed that Euclidian and Minkowski on the KNN algorithm achieved the best result compared to Chebycev and Manhattan. The best results on KNN are obtained when the K value is 3.