Irwan Oyong

Mr. Irwan Oyong

Universitas Amikom Yogyakarta, Yogyakarta, Indonesia
orcid

Research Focus & Metrics

Academic Career Length

10Years (2017-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
20 April 2022 in Applied Sciences
Verified
Journal Article
Open access
PDF file
Setyanto AriefSetyanto Arief
Setyanto Arief
Arif Dwi LaksitoArif Dwi Laksito
Arif Dwi Laksito
FAFawaz Khaled Alarfaj
Fawaz Khaled Alarfaj
MAMohammed Alrashoodi
Mohammed Alrashoodi
KKKusrini Kusrini
Kusrini Kusrini
LKLilis Kurniasari
Lilis Kurniasari
Arabic is one of the official languages recognized by the United Nations (UN) and is widely used in the middle east, and parts of Asia, Africa, and other countries. Social media activity currently dominates the textual communication on the Internet and potentially represents people’s views about specific issues. Opinion mining is an important task for understanding public opinion polarity towards an issue. Understanding public opinion leads to better decisions in many fields, such as public services and business. Language background plays a vital role in understanding opinion polarity. Variation is not only due to the vocabulary but also cultural background. The sentence is a time series signal; therefore, sequence gives a significant correlation to the meaning of the text. A recurrent neural network (RNN) is a variant of deep learning where the sequence is considered. Long short-term memory (LSTM) is an implementation of RNN with a particular gate to keep or ignore specific word signals during a sequence of inputs. Text is unstructured data, and it cannot be processed further by a machine unless an algorithm transforms the representation into a readable machine learning format as a vector of numerical values. Transformation algorithms range from the Term Frequency–Inverse Document Frequency (TF-IDF) transform to advanced word embedding. Word embedding methods include GloVe, word2vec, BERT, and fastText. This research experimented with those algorithms to perform vector transformation of the Arabic text dataset. This study implements and compares the GloVe and fastText word embedding algorithms and long short-term memory (LSTM) implemented in single-, double-, and triple-layer architectures. Finally, this research compares their accuracy for opinion mining on an Arabic dataset. It evaluates the proposed algorithm with the ASAD dataset of 55,000 annotated tweets in three classes. The dataset was augmented to achieve equal proportions of positive, negative, and neutral classes. According to the evaluation results, the triple-layer LSTM with fastText word embedding achieved the best testing accuracy, at 90.9%, surpassing all other experimental scenarios.

31 October 2019 in Journal of King Saud University Computer and Information Sciences
Verified
Journal Article
Open access
PDF file
EUEma Utami
Ema Utami
AHAnggit Dwi Hartanto
Anggit Dwi Hartanto
SASumarni Adi
Sumarni Adi
Irwan OyongIrwan Oyong
Irwan Oyong
SRSuwanto Raharjo
Suwanto Raharjo
MIMihuandayani
Mihuandayani
Data in the timeline of social media users consists of data in the form of text, images, audio, and video. Large and unstructured data in social media can be processed using various techniques such as text processing or image processing. In this study, the processed text data is used to classify Twitter users’ personality based on the DISC framework. Out of the initial collected 292 users, we semi-automatically filtered them for only personal accounts with Indonesian language posts. For being able to observe and assess a user’s personality out of their tweets choice of words, we made relevant keyword vocabularies corresponding to DISC framework and theory. There are four experiment scenarios done in this study, with variations on whether the keywords and text data are stemmed or not, and the keywords frequency calculation being weighted or not. Weighting the keywords using the current number in calculation based on their level does not show positive results, neither does stemming as the best results are shown by the not stemmed and not weighted scenario. This study is a preliminary research for an automatic profiling system which employs a combination of Natural Language Processing and Machine Learning approaches.

7 September 2021 in Applied Computing and Informatics
Verified
Journal Article
Open access
PDF file
EUEma Utami
Ema Utami
Irwan OyongIrwan Oyong
Irwan Oyong
SRSuwanto Raharjo
Suwanto Raharjo
AHAnggit Dwi Hartanto
Anggit Dwi Hartanto
SASumarni Adi
Sumarni Adi
Purpose: Gathering knowledge regarding personality traits has long been the interest of academics and researchers in the fields of psychology and in computer science. Analyzing profile data from personal social media accounts reduces data collection time, as this method does not require users to fill any questionnaires. A pure natural language processing (NLP) approach can give decent results, and its reliability can be improved by combining it with machine learning (as shown by previous studies).Design/methodology/approach: In this, cleaning the dataset and extracting relevant potential features “as assessed by psychological experts” are essential, as Indonesians tend to mix formal words, non-formal words, slang and abbreviations when writing social media posts. For this article, raw data were derived from a predefined dominance, influence, stability and conscientious (DISC) quiz website, returning 316,967 tweets from 1,244 Twitter accounts “filtered to include only personal and Indonesian-language accounts”. Using a combination of NLP techniques and machine learning, the authors aim to develop a better approach and more robust model, especially for the Indonesian language.Findings: The authors find that employing a SMOTETomek re-sampling technique and hyperparameter tuning boosts the model’s performance on formalized datasets by 57% (as measured through the F1-score).Originality/value: The process of cleaning dataset and extracting relevant potential features assessed by psychological experts from it are essential because Indonesian people tend to mix formal words, non-formal words, slang words and abbreviations when writing tweets. Organic data derived from a predefined DISC quiz website resulting 1244 records of Twitter accounts and 316.967 tweets.

1 July 2018
Verified
Conference Paper
Closed access
Irwan OyongIrwan Oyong
Irwan Oyong
EUEma Utami
Ema Utami
ELEmha Taufiq Luthfi
Emha Taufiq Luthfi
The shame and disgrace of depression cause people not to seek help for the problem. Contemporary social media technologies such as Twitter provide an opportunity for people to be able to express their feelings in an anonymous and confidential environment. In this study symptoms lexicon terms are used with the frequency lexicon for weighting and perform depression score calculations from text materials that have undergone Natural Language Processing. Through the scraping process, 55 relevant users were found who netted depression-related keywords. For each user, a data search is made for a week before and after the initial tweet. Tweet extensions generate a total of 6055 tweets from 55 users in question. The sum of the scores becomes the label determinant of user depression, which is then compared to the labels that have been given manually by psychologists based on clinical screening standards. Based on comparative and evaluation results, the same F1 score of 0.47 is obtained for standard text processing and text-specific processing for Twitter, and a Sensitivity value of 0.89 at the threshold value of 0.5. A slightly better F1 Score value of 0.50 is obtained by text-specific processing on the threshold value of 0.8. Research shows differences in minor results between standard text processing and text-specific processing of Twitter. Seen some of the advantages of text-specific Twitter processing when handling non-standard text, thereby enhancing the fmdings of the Part-of-Speech tagging process.

1 July 2019
Verified
Conference Paper
Closed access
SASumarni Adi
Sumarni Adi
AHAnggit Dwi Hartanto
Anggit Dwi Hartanto
EUEma Utami
Ema Utami
SRSuwanto Raharjo
Suwanto Raharjo
Irwan OyongIrwan Oyong
Irwan Oyong
Twitter have a tremendous number of tweets from its user. The bigger of the data’s volume and more complete the tweet’s data that gathered, the more accurate the personality analysis will be. This study is sought to analyze and measure which acquisition technique that viable to be used on the personality profile analysis and to know whether the acquisition technique will always return a huge amount of data volume in a longer time period with a faster time or not. The most commonly used acquisition technique is crawling, streaming and scrapping. This study makes uses of those three acquisition methods to measure the compatibility toward the personality profile analysis. Based on the previous study, the usage of scrapping acquisition method is proved to be an arguably good method as it can acquire a more completed tweet data in a huge amount with bigger computation time and the acquisition time is faster. It caused by the big volume of the data and it need more computational time, so the acquisition speed of the tweet data is faster. Based on that result, scrapping technique is said to be a good option to analyze someone’s personality profile.

30 September 2022 in Jurnal Informatika
Verified
Journal Article
Open access
PDF file
AMAlva Hendi Muhammad
Alva Hendi Muhammad
KKKusrini Kusrini
Kusrini Kusrini
Irwan OyongIrwan Oyong
Irwan Oyong
Revisiting the challenges and surveys in text similarity matching and detection methods