Social media is an online that functions as a platform for users to participate, share, create, and exchange information through various forums social networks. The rapid increase in activity causes the number of comments on media. This prone triggering debate due easy formation open discussions between users. However, often triggers emergence negative things, causing great fights use containing toxic words argue corner party or group. study conducted experiment detect sentences Indonesia...

Cryptocurrencies are digital assets that widely used for trading and investing. One of the characteristics traders take advantage profit is high volatility price. Its volatile rapidly changing prices have made cryptocurrency price predictions a challenging highly sought-after research topic. Cryptocurrency usually only use historical on dataset, while movements also influenced by other aspects such as sentiment contained in social media. This study proposes new machine learning method to...

Network security has become crucial in an era where information and data are valuable assets.An effective Intrusion Detection System (NIDS) is required to protect sensitive from cyberattacks.Numerous studies have created NIDS using machine learning algorithms network datasets that do not accurately reflect actual flows.Increasing hardware capabilities the ability process big made deep preferred method for developing NIDS.This study develops a model two algorithms: Convolutional Neural (CNN)...

Hospital is one of the most complex organization with highly intensive interaction between stakeholders (patients, nurses, doctors, staff, etc.). In operation a hospital, use Information technology has been proven to improve effectiveness and efficiency. However, in majority cases, processes achieve Strategic Objectives through implementation Technology are full challenges. Based on case study Dharmais Cancer Hospital, there many symptoms that identified by this lead 4 issues, namely: lack...

The Incorporation of Question Answering system holds immense potential for addressing Indonesia’s educational disparities between the abundance high school students and limited number teachers in Indonesia. These studies aim to enhance System model tailored Indonesian language dataset through enhancements IndoBART model. Improvement was done by incorporating Longformer’s sliding windows attention mechanism into model, it would increase proficiency managing extended sequence tasks such as...

Closed-circuit television (CCTV) systems have generated a vast amount of visual data crucial for security and surveillance purposes. Effectively categorizing level types is vital maintaining asset effectively. This study proposes practical approach classifying CCTV screenshot images using geometry group (VGG-19) transfer learning, convolutional neural network (CNN) classification model that works really well in image classification. The task compromise screenshots into two classes: “humans...

Currently, social media are commonly used by business organizations to get many advantages. Through media, gets feedback from their customer giving them a room for sharing opinion or comment. This kind of engagement became tool marketing and promotion. research proposes investigating the analysis comments on (Facebook Twitter) in transport online. study investigates mining fan page Facebook tweet Twitter three transports online Indonesia; Gojek, Grab, Uber using API service which is provided...

The permutation flow shop scheduling problem (PFSP) is a renowned in the research community. It an NP-hard combinatorial optimization that has useful real-world applications. In this problem, finding algorithm to handle massive amounts of jobs required retrieve actionable order reasonable amount time important. recently developed crow search (CSA) novel swarm-based metaheuristic originally proposed solve mathematical problems. paper, hybrid CSA (HCSA) minimize makespans PFSPs. First, make...

The number of news produced every day is as much 3 million per day, making readers have many choices in choosing according to each reader's topic and category preferences. recommendation system can make it easier for users choose the read. method that be used providing recommendations from same user collaborative filtering. Neural filtering usually being systems by combining with neural networks. However, this has disadvantage recommending similarity content such titles users. This research...

The objective of this study is to develop learning media by using the Computer Based Instruction (CBI) model for GOST algorithm cryptographic material (Gosudarstvennyi Standard). Besides, serves present computer media, especially in Informatics Engineering program. method used uses Research and Development (R & D) method. results development research are products form tutorial CDs with contents material: Cryptographic Theory, Encryption Process Decryption process alphabet text data. on...

This study will classify the text based on rating of provider application Google Play Store. research is classification user comments using Word2vec and deep learning algorithm in this case Long Short Term Memory (LSTM) given with a scale 1-5 detailed 1 lowest 5 highest data 1-3 rating, as negative combination ratings 2, 2 neutral 3, 3 positive 4 to get sentiment from users SMOTE oversampling handle imbalance data. The used are 16369 training testing be taken MyTelkomsel’s play.google.com...

Purpose The purpose of this study is to provide the location natural disasters that are poured into maps by extracting Twitter data. text extracted using named entity recognition (NER) with six classes hierarchy in Indonesia. Moreover, tweet then classified eight support vector machine (SVM). Overall, system able classify and mapping position content tweet. Design/methodology/approach This research builds a model map geolocation data NER. uses NER which based on region SVM. Findings...

Spam refers to unsolicited messages containing harmful content such as malware, viruses, phishing, and data theft.The web form of a government ministry's website is frequently targeted by spammers, causing disruptions, database overload, hindering communication with the public, security risks.While numerous studies have focused on spam detection, none addressed detection submissions multilingual specifically in English Indonesian.This study developed model address growing challenge received...

Extractive text summarization has been a popular research area for many years. The goal of this task is to generate compact and coherent summary given document, preserving the most important information. However, current extractive methods still face several challenges such as semantic drift, repetition, redundancy, lack coherence. A novel approach presented in paper improve performance an model based on bidirectional encoder representations from transformers (BERT) by incorporating topic...

This article discusses the analysis of customer loyalty using three data mining methods: C4.5,Naive Bayes, and Nearest Neighbor Algorithms real-world empirical data. The contain ten attributes related to are obtained from a national multimedia company in Indonesia. dataset contains 2269 records. study also evaluates effects size training accuracy classification. results suggest that C4.5 algorithm produces highest classification at order 81% followed by methods Naive Bayes 76% 55%. In...

Abstract Content based recommendation system tries to recommend items similar those a given user has likely in the past, whereas systems designed according collaborative paradigm identify users whose preferences are of and they have liked. The proposed is discussing about Data finitihotel recommendations for place across U.S. selected process calculating rating hotel categories on city. By from several city by combining two features namely taken hotelcategory as well. result will be...

In the process of developing a business, aspect-based sentiment analysis (ABSA) could help extract customers' opinions on different aspects business from online reviews. Researchers have found great prospective in deep learning approaches to solving ABSA tasks. Furthermore, studies also explored implementation text augmentation, such as Easy Data Augmentation (EDA), improve models’ performance using only simple operations. However, when implementing EDA ABSA, there will be high chances that...

Covid-19 has grown rapidly in all parts of the world and is considered an international disaster because its wide-reaching impact. The impact spread to Indonesia, especially slowdown economic growth. This was influenced by implementation Community Activity Restrictions (PPKM) which limited community activities. study aims analyze mapping public sentiment towards PPKM policies Indonesia during pandemic based on Twitter data. Knowing regarding expected help stakeholders policy evaluation...

Communication Center (Puskom) as a management of information and communication systems the Ministry Foreign Affairs Indonesia Representative, places Information Communications Technology (ICT) most strategic sector to support implementation foreign policy. Many problems faced by relates IT governance such as, reliability, availability lack infrastructure which causes problem hackers, natural disasters, others. This research aims implement service for improving quality in Puskom Affairs. ITIL...

The problem with telecommunications companies today is that transactional data more extensive than existing source tables. This makes business reporting less efficient and overwhelms query processing results in warehouses so they do not meet requirements. fast complex evolution of the digital world must be scalable to warehouse process, authors implement it using massive parallel (MPP) Greenplum database, users can get reports faster optimally. case study explains how MPP system implements...

As the main public company in telecommunication and broadband business Indonesia, PT XYZ has always tried to ensure adequate wifi access service meet needs of consumers. This service, called wifi-id, uses an Authentication, Authorization, Accounting (AAA) server by collaborating with other network providers. study was based on need PT. present reports or data quickly when needed, especially reconcile total monthly customer usage partners order do billing revenue sharing partnership. The...

Electronic ticket (eticket) provider services are growing fast in Indonesia, making the competition between companies increasingly intense. Moreover, most of them have same service or feature for serving their customers. To get back feedback customers, many use social media (Facebook and Twitter) marketing activity communicating directly with The development current technology allows company to take data from media. Thus, analyses. This study proposed developing a warehouse analyze such as...

One of the lack eigenface for prediction face recogniton is not good accuracy. This paper uses naive Bayes classifying result feature extraction to predict face. The normalization z-score added sharping To see performance proposed method, 200 datasets are divided into data training and testing by using cross validation (k=10). results show that method can image up 70%. Moreover adding Z-Score, accuration raise 89.5% (in average).