Indonesian News Research Articles

MODEL HYBRID NONLINEAR REGRESSION LOGISTIC (NLR) –DOUBLE EXPONENSIAL SMOOTHING (DES) DAN PENERAPANNYA PADA JUMLAH KASUS KUMULATIF COVID-19 DI INDONESIA DAN BELANDA

The economic relationship between Indonesia and the Netherlands is a good trade relationship, but the spread of COVID-19 disrupts the two countries' economies. Both countries need to have an explanation regarding the condition of COVID-19 to raise economic market sentiment. Based on this, Hybrid and non-hybrid models are used to predict the dispersion conditions and compare them through the MAPE value. The double-exponential nonlinear logistic regression hybrid model on the cumulative number of COVID-19 is not suitable for use in the Netherlands COVID-19 cases but is suitable for use in the cumulative number of COVID-19 cases Indonesia. The hybrid nonlinear regression logistic-double exponential model is one way to optimize MAPE, especially in training data. Based on the hybrid non-client regression logistic model, the peak incidence of Covid-19 in the Netherlands is estimated at 22 November 2020, and the hybrid nonlinear regression logistic-Double exponential model predicts that the peak of Covid-19 occurs in Indonesia on 28 November 2020. the Netherlands wave is around 2.83 percent and Indonesia 1.62 percent. Therefore the decline in Indonesia is predicted to be faster, but the Netherlands will reach the peak of the Indonesian news wave.

Author Obfuscation on Indonesian News Articles Using Genetic Algorithms

Authorship attribution is a method for identifying the author of a text from a group of potential authors and can solve the anonymity of unknown authors. Such method threatens anyone’s privacy, especially those who wish to write anonymously. To address this issue, author obfuscation is proposed to modify a text to disguise its author.In this research, a genetic algorithm-based author obfuscation model was created to modify Indonesian news articles to avoid identification from authorship attribution while keeping its semantics. The model iteratively changed some words in the article using crossover and mutation techniques guided by a fitness function which involve identification probability and similarity to the original article.The model is evaluated based on safety, soundness, and sensibleness parameter. The model has good safety since it can reduce the given authorship attribution model's accuracy by 0.3018 but drops to 0.1179 when tested on different models. Its soundness is pretty good since the similarity of the modified to the original articles reaches 0.7817. The model obtained a score of 2.571 on a scale of 0 to 4 in terms of sensibleness which indicates that some articles are acceptable in terms of grammar, but not a few are messy.

Translation Strategies in <i>Tempo</i> English Magazine

This research aims at identifying and examining the translation strategies applied in Tempo English magazine, the English edition of Tempo magazine. The data were gathered from 10 selected articles of the source magazine, Tempo, and its translated magazine, Tempo English, both of which belonged to the edition of April 30 – May 6, 2018. The analysis and categorization were based on the news translation theory proposed by Bielsa and Bassnett and supported by Joan Cutting’s context theory as well. The result of this research shows that the translators of Tempo English magazine used all of the news translation strategies in translating the Indonesian news articles to English. There were 199 data found in the research with 77 data of the Addition strategy, 68 data of the Elimination, 25 data of the strategy of Summarizing Information, 15 data of the Change in the Order of Paragraphs, and the other 14 data using the Change of Title and Lead. The research analysis concludes that in the strategy of the Change in the Order of Paragraphs, Tempo English magazine tended to apply the change of order at the phrase and sentence level rather than at the paragraph level.

Fake news classification for Indonesian news using Extreme Gradient Boosting (XGBoost)

Fake news or commonly known as a hoax has become one of the most visible cybercrime. Hoax news dissemination harms the social community, such as raising hatred towards something both individuals and groups. This paper is to classify amongst hoaxes and valid news utilizing Extreme Gradient Boosting (XGBoost) method in this research based on Indonesian news. The dataset used is Indonesian news about Indonesia itself and the world from 2015 to early 2020. The study used 500 news data including 250 valid news and 250 hoax news, divided into 80% training data and 20% test data. The result of this study shows that the machine learning model created using XGBoost has an accuracy value of 89%, with the precision value of 90% and recall value 80%.

STUDENTS' RECEPTION OF COVID-19 NEWS ON CNN INDONESIA "GOOD MORNING" PROGRAM

The news of the coronavirus or Covid-19 has become a hot topic of discussion in the last few months. This article investigates the student's reception of the Corona Virus Alert segment broadcasted through the CNN Indonesia news program, "Good Morning." Using Stuart Hall's reception studies model, we conducted in-depth interviews with five students who watched Covid-19 news intensively on "Good Morning," CNN Indonesia. We found that students decode differently the meaning of messages conveyed by the information. Some coded the story in a hegemonic or dominant position --it means they agree with what and how the media frame the events and phenomena. On the other hand, some interpreted the meaning conveyed through the news with a different perspective, indicating they are in the oppositional position. They criticized the media for providing inaccurate, even factual, information. This study is vital in the midst of the Covid-19 pandemic information storm conveyed by both the media and the public. It becomes a reference for how the audience can interpret news differently and can share their interpretations with the community according to what they receive. To cite this article (7th APA style):Rachma, S. & Kholisoh, N. (2020). Students' Reception of Covid-19 News on CNN Indonesia "Good Morning" Program. Journal Communication Spectrum: Capturing New Perspectives in Communication, 10(2), xx-xx. http://dx.doi.org/10.36782/jcs.v10i2.2104

Utilizing Indonesian Universal Language Model Fine-tuning for Text Classification

Inductive transfer learning technique has made a huge impact on the computer vision field. Particularly, computer vision applications including object detection, classification, and segmentation, are rarely trained from scratch; instead, they are fine-tuned from pretrained models, which are products of learning from huge datasets. In contrast to computer vision, state-of-the-art natural language processing models are still generally trained from the ground up. Accordingly, this research attempts to investigate an adoption of the transfer learning technique for natural language processing. Specifically, we utilize a transfer learning technique called Universal Language Model Fine-tuning (ULMFiT) for doing an Indonesian news text classification task. The dataset for constructing the language model is collected from several news providers from January to December 2017 whereas the dataset employed for text classification task comes from news articles provided by the Agency for the Assessment and Application of Technology (BPPT). To examine the impact of ULMFiT, we provide a baseline that is a vanilla neural network with two hidden layers. Although the performance of ULMFiT on validation set is lower than the one of our baseline, we find that the benefits of ULMFiT for the classification task significantly reduce the overfitting, that is the difference between train and validation accuracies from 4% to nearly zero.

Political Scandal and Public Figure

Indonesian society is familiar with the terms 'public figures' and 'celebrities', but the distinction is often not understood properly. The public's interest in content that focuses on entertainment, lifestyles, and gossip, as well as the presence of a media that facilitates such content, makes the process of 'celebrating' common. This process has resulted in the private space of public figures being transformed into objects of public consumption.Scandals are often quite popularly discussed among the public, especially when their subject is a public figure. However, studies of how scandals affect the public and its political behaviour have not been widely documented. In 2018, Indonesian news media began widely covering the divorce of well-known politician Basuki Tjahaja Purnama (BTP/Ahok) from his ex-wife Veronica Tan, and this brought questions of extramarital affairs to the surface in the midst of a heated local election atmosphere. This situation was divisive, and received various public responses. Previous studies have shown that scandals tend to negatively affect popular attitudes towards the politicians involved in them. In Indonesia, scandals have been common, widely recognised by the public, but their effects are never discussed in depth. Therefore, this study, which involved around 400 respondents, seeks to provide an overview of how the Indonesian public responds to politicians involved in scandals and how such scandals affect politicians' electability. The results of this study show that scandals do affect the public's political attitude, but not in the ways suggested by existing studies.

Development of Web Crawler to Build Indonesian Text Corpus

Recent improvement in natural language understanding research can be attributed to the availability of large scale datasets. Those datasets are mainly in English. In this work, we develop a web crawler with the purpose of extracting Indonesian news content from the DetikNews website and building a large dataset of texts. The web crawler is developed by following the waterfall model using Python, Scrapy, and BeautifulSoup4. It collects more than 790k news from DetikNews, spanning from 2011 to 2020, which consists of a total number of more than 190 million words, almost 2 million unique words, and more than 14 million sentences.

Event Geoparser with Pseudo-Location Entity Identification and Numerical Argument Extraction Implementation and Evaluation in Indonesian News Domain

Geoparser is a fundamental component of a Geographic Information Retrieval (GIR) geoparser, which performs toponym recognition, disambiguation, and geographic coordinate resolution from unstructured text domain. However, geoparsing of news articles which report several events across many place-mentions in the document are not yet adequately handled by regular geoparser, where the scope of resolution is either toponym-level or document-level. The capacity to detect multiple events and geolocate their true coordinates along with their numerical arguments is still missing from modern geoparsers, much less in Indonesian news corpora domain. We propose an event geoparser model with three stages of processing, which tightly integrates event extraction model into geoparsing and provides precise event-level resolution scope. The model casts the geotagging and event extraction as sequence labeling and uses LSTM-CRF inferencer equipped with features derived using Aggregated Topic Model from a large corpus to increase the generalizability. Throughout the proposed workflow and features, the geoparser is able to significantly improve the identification of pseudo-location entities, resulting in a 23.43% increase for weighted F1 score compared to baseline gazetteer and POS Tag features. As a side effect of event extraction, various numerical arguments are also extracted, and the output is easily projected to a rich choropleth map from a single news document.

Strategies of ideological polarisation in the online news media: A social actor analysis of Megawati Soekarnoputri

This article examines strategies of ideological polarisation in the discourse of the Indonesian online news media site, Kompas.com. Applying Van Leeuwen’s model of social actor analysis and van Dijk’s concept of the ideological square, the study focuses on the representation of Megawati Soekarnoputri, leader of the Indonesian Democratic Party of Struggle (PDI-P) as an icon of ideological contestation during the 2014 presidential election. Situated in the era of digital platform convergence, the analysis uncovers a pattern of strategically ambiguous representations of Megawati and her apparently transgressive actions and interactions. This practice entices readers to ‘read between the lines’ and activate their ideological repertoire to determine in-group and out-group members. It also enables Kompas.com to pursue commercial objectives and navigate journalistic constraints by obscuring explicitly ideological content. The implications are discussed in terms of the impact of online news media discourse upon democratic political engagement, and women’s political participation.

Curing the Infected Wound: Metaphor of State-Owned Enterprises in News Headlines

The current condition of Indonesian State Own Enterprises (SOEs) has become the focus of various studies across fields. The studies extend from an economic perspective to legal aspects, but the study to address the SOEs condition from linguistics perspective has been overlooked. On the other hand, the study of conceptual metaphor in media discourse puts little attention on this particular issue, and plenty of studies of conceptual metaphor do not account for the effect of contextual aspects on the additional inferences of metaphor. This study attempts to fill the gaps by examining the news headlines reports on the current conditions of SOEs using metaphorical expressions. It will take a closer look at the conceptual metaphor STATE-OWNED ENTERPRISES ARE PATIENTS in Indonesian news headlines and conducting further investigation on how contextual aspects influence the additional metaphorical inferences. The study is qualitative with purposive sampling method; the data taken are ten news headlines from online media news which discuss the current improvement process and condition of SOEs using metaphorical expressions. Conceptual Metaphor Theory (Lakoff & Johnson, 1980) is the conceptual framework of this study. From ten news headlines analysed, three metaphorical framings to speak about the current condition of SOEs as if SOEs are patient, i.e.: SOEs is a sick patient with particular sickness, the government is a medical practitioner, and after treatment, SOEs is a healthy patient or a dying patient. The underlying conception of STATE-OWN ENTERPRISES ARE PATIENTS implies that there are still plenty outstanding actions to be taken to improve the SOEs' performance, not only to provide remedies but most importantly is to prevent the sickness. Furthermore, additional inferences may emerge from the same metaphorical expressions in a different situational context. The implication is, media or journalists should consider that unintended inferences may arise from a particular metaphorical expression which are not initially meant to.

CLICK-ID: A novel dataset for Indonesian clickbait headlines.

News analysis is a popular task in Natural Language Processing (NLP). In particular, the problem of clickbait in news analysis has gained attention in recent years [1, 2]. However, the majority of the tasks has been focused on English news, in which there is already a rich representative resource. For other languages, such as Indonesian, there is still a lack of resource for clickbait tasks. Therefore, we introduce the CLICK-ID dataset of Indonesian news headlines extracted from 12 Indonesian online news publishers. It is comprised of 15,000 annotated headlines with clickbait and non-clickbait labels. Using the CLICK-ID dataset, we then developed an Indonesian clickbait classification model achieving favourable performance. We believe that this corpus will be useful for replicable experiments in clickbait detection or other experiments in NLP areas.

Indonesian news classification using convolutional neural network

<span>Every language has unique characteristics, structures, and grammar. Thus, different styles will have different processes and result in processed in Natural Language Processing (NLP) research area. In the current NLP research area, Data Mining (DM) or Machine Learning (ML) technique is popular, especially for Deep Learning (DL) method. This research aims to classify text data in the Indonesian language using Convolutional Neural Network (CNN) as one of the DL algorithms. The CNN algorithm used modified following the Indonesian language characteristics. Thereby, in the text pre-processing phase, stopword removal and stemming are particularly suitable for the Indonesian language. The experiment conducted using 472 Indonesian News text data from various sources with four categories: ‘hiburan’ (entertainment), ‘olahraga’ (sport), ‘tajuk utama’ (headline news), and ‘teknologi’ (technology). Based on the experiment and evaluation using 377 training data and 95 testing data, producing five models with ten epoch for each model, CNN has the best percentage of accuracy around 90,74% and loss value around 29,05% for 300 hidden layers in classifying the Indonesian News data.</span>

Hybrid Conditional Random Fields and K-Means for Named Entity Recognition on Indonesian News Documents

Hybrid Conditional Random Fields and K-Means for Named Entity Recognition on Indonesian News Documents

Users’ Review on Converged News: Indonesia News Apps Case

Review is important for users to adapt, adopt, and accept new mobile applications. The digital and Internet have changed the way of users viewing and reading news. Since news outlet has transformed from print to website, mobile website, and mobile application, the users’ testimony could make them install, use, and be loyal. At the developers’ perspective, the review could make new ways to improve the features of news apps. The news apps that have better features could give better impression to readers. Moreover, news apps could make so much easier for users or readers to keep up with the latest news and information. It makes a lot of interest in the world of news on mobile platforms especially the news outlet that concern on the behaviour of users such as Kompas and Republika. By collecting the reviews and categorizing them with rating appreciation, PACMAD, EMA, TAM, and Nielsen’s Acceptability, this study found that not all categories were relevant and considered perfectly fit in news apps. The new construct has been developed about user appreciation of news apps. The elaboration of measurement gives new implication toward converged news media. The implication of this study would be also discussed and would give theoretical and methodological suggestions for further study

Comparative Analysis of Text Mining Results With Tf ldf Features and SQL Like Operator in Indonesian News Search

Research on the implementation of text mining uses the TF IDF method to be used in the Information retrieval / Indonesian news search feature. The dataset used was sourced from NewsAPI and built a Codeigniter based website named "News Plus Six Dua". This study also uses the Vertor Space Model (VSM) method to overcome the weaknesses of the TF IDF method at the time of the sorting process. The results of this study explain that the search by the TF IDF method has higher accuracy when compared to SQL like operators. TF IDF produces a percentage of precision 100% and recall (sensitivity) 66.7% on searches with the keyword "Indonesian soccer schedule" while SQL like operators do not display search results or equal to 0%. But the TF IDF method has the disadvantage of running slower than SQL like operators. This has been tested using either the number of words or terms entered, the number of datasets, and the location of access. At the location of access, access via hosting is monitored faster when compared via localhost.

Detecting Hoaxes in Indonesian News Using TF/TDM and K Nearest Neighbor

The presence of the internet and the rapid growth of social media had given rise to the blossoming of hoax creation and distribution through it. A hoax can cause anxiety and reactivity to its readers and could harm a certain party. Thereby, it is important to detect and report hoaxes to stop its spreading as soon as possible. This research aims to utilize the K Nearest Neighbor (KNN) classification algorithm to detect whether a piece of news is a hoax or not. Experiments were done by using 74 hoaxes compiled from Indonesian hoax-debunking community websites and were being compared against 74 real news from various reputable news websites in Indonesia. The result showed that the model could give detection/classification accuracy up to 83.6% and that the model is prone to false positives detections. The characteristics of the resulted model and further research directions are then discussed.

Hoax classification and sentiment analysis of Indonesian news using Naive Bayes optimization

Currently, the spread of hoax news has increased significantly, especially on social media networks. Hoax news is very dangerous and can provoke readers. So, this requires special handling. This research proposed a hoax news detection system using searching, snippet and cosine similarity methods to classify hoax news. This method is proposed because the searching method does not require training data, so it is practical to use and always up to date. In addition, one of the drawbacks of the existing approaches is they are not equipped with a sentiment analysis feature. In our system, sentiment analysis is carried out after hoax news is detected. The goal is to extract the true hidden sentiment inside hoax whether positive sentiment or negative sentiment. In the process of sentiment analysis, the Naive Bayes (NB) method was used which was optimized using the Particle Swarm Optimization (PSO) method. Based on the results of experiment on 30 hoax news samples that are widely spread on social media networks, the average of hoax news detection reaches 77% of accuracy, where each news is correctly identified as a hoax in the range between 66% and 91% of accuracy. In addition, the proposed sentiment analysis method proved to has a better performance than the previous analysis sentiment method.

Fact Searching in Investigative Reporting in Indonesia: A Case Study of Tempo and Gatra Magazines

Investigative journalism is a new development in the Indonesian press and the study is at infancy. However, news magazines have investigated quite a number of scandals and problems in Indonesia during the Reformation era, post-New Order authoritarian regimes. This research examines fact searching activities at two main news magazines in Indonesia namely Tempo and Gatra in the process of producing investigative stories. This study deploys case study and an in-depth interview in its exploration of identifying specific processes involved in the process of producing investigative stories in these two news organizations. The interview was conducted with six informants including journalists at Tempo and Gatra news magazine, book authors and journalism analysts. The results of the study indicate that news magazines in Indonesia reveal various issues and cases, which are deliberately hidden to the community. The findings also demonstrate that the facts searching process encompasses several styles that effectively complement each other in news narration, especially with the particular concept, attitude, direction and effort to report its best to the society. This suggests that investigative stories in Indonesia news magazines emphasize the originality of news with in-depth coverage based on independent position, free from the influence of other societal entities, government’s pressure, as well as from other interests. Keywords: Investigative journalism, reformation era, news magazine, Tempo, Gatra.

CLASSIFICATION OF NEWS TYPES BY IMPLEMENTING ENHANCED CONFIX STRIPPING STEMMER

News has become a community need in the world. Managing a lot of news articles is not easy and takes a long time. Indonesia has various types of media platforms that display news, one of which is an online news portal. Automation systems that are capable of managing and grouping Indonesian language news articles are needed. This study designed and built a web-based application to classify types of Indonesian language news articles by implementing the Enhanced Confix Stripping Stemmer algorithm. The categories used in the system are entertainment, lifestyle, sports, technology, and economics. The data used is secondary data quoted from 2 online news portals in Indonesia. The system development method used is Rapid Application Development. The data used for testing amounts to 30 news. The average results obtained from the system accuracy test are 63%. This shows that the system performance for the classification of news types is good. The number of words in a news article is very influential during the classification process.