cover
Contact Name
Mesran
Contact Email
mesran.skom.mkom@gmail.com
Phone
-
Journal Mail Official
jurnal.bits@gmail.com
Editorial Address
-
Location
Kota medan,
Sumatera utara
INDONESIA
Building of Informatics, Technology and Science
ISSN : 26848910     EISSN : 26853310     DOI : -
Core Subject : Science,
Building of Informatics, Technology and Science (BITS) is an open access media in publishing scientific articles that contain the results of research in information technology and computers. Paper that enters this journal will be checked for plagiarism and peer-rewiew first to maintain its quality. This journal is managed by Forum Kerjasama Pendidikan Tinggi (FKPT) published 2 times a year in Juni and Desember. The existence of this journal is expected to develop research and make a real contribution in improving research resources in the field of information technology and computers.
Arjuna Subject : -
Articles 889 Documents
Sentimen Analisis Social CRM Pada Media Sosial Instagram Menggunakan Machine Learning Untuk Mengukur Retensi Pelanggan F. Safiesza, Qhairani Frilla; Afdal, M; Novita, Rice; Mustakim, Mustakim
Building of Informatics, Technology and Science (BITS) Vol 6 No 2 (2024): September 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i2.5269

Abstract

To create and maintain a superior competitive advantage in a knowledge-based economy, businesses must be able to utilize data and manage customer relationships through the implementation of Customer Relationship Management (CRM), particularly Social CRM. Social CRM is a renewal of business strategy that is created to engage customers in a collaborative conversation and create mutually beneficial value in a trusted and transparent business environment. Seeing this development as one of the successful culinary companies in the Souvenir sector in Pekanbaru, the company must be able to process all the information obtained. Currently, the company has never analyzed comments on social media, especially the Instagram account. These comments are useful for evaluation material and can be a parameter of customer satisfaction and to see the potential for customer retention. To assess positive and negative comments on the Instagram account, sentiment analysis can be carried out using machine learning, namely 3 classification algorithms, namely Naive Bayes Classifier (NBC), Support Vector Machine (SVM) and Random Forest (RF). The sentiment results show that the SVM and NBC algorithms obtain the best accuracy of 74.26% compared to RF, and the results of the social CRM analysis show that customers are more satisfied with the company in terms of products, services, and actions taken by the company, so that the company is considered capable of retaining its customers.
Perbandingan Algoritma Support Vector Machine dan Decision Tree untuk Klasifikasi Performa Perusahaan Utomo, Mario; Prathivi, Rastri
Building of Informatics, Technology and Science (BITS) Vol 6 No 1 (2024): June 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i1.5278

Abstract

The number of stock exchange investors in Indonesia reached 5.34 million by the end of December 2023. This figure is dominated by millennial generation investors, indicating a growing confidence in the fundamentals and economic prospects of the Indonesian capital market. However, the lack of financial literacy among this generation often results in ineffective and high-risk investments. Many millennials choose stocks based on short-term trends or recommendations that lack analysis. To address this issue, a more structured approach to stock selection is required. One method that can be employed is the classification of a company's performance based on its performance using various financial indicators and ratios. As the performance of a company affects the movement of its stock value, this research will compare Support Vector Machine and Decision Tree with the One Against All approach in classifying company performance. The features used for the classification of company performance consist of three financial ratios: profitability (ROA), liquidity (CR), and leverage (DER). The labels or targets in the classification are divided into three categories: normal, good, and unfavorable. This research will consider evaluations such as accuracy, cross validation, and confusion matrix. The results of the Support Vector Machine (SVM) algorithm demonstrated an accuracy of 86.67%, while the Decision Tree (DT) algorithm exhibited an accuracy of 93.33%. Consequently, the DT algorithm produced more accurate results than the SVM algorithm in classification. The number of stock exchange investors in Indonesia reached 5.34 million by the end of December 2023. This figure is dominated by millennial generation investors, indicating a growing confidence in the fundamentals and economic prospects of the Indonesian capital market. However, the lack of financial literacy among this generation often results in ineffective and high-risk investments. Many millennials choose stocks based on short-term trends or recommendations that lack analysis. To address this issue, a more structured approach to stock selection is required. One method that can be employed is the classification of a company's performance based on its performance using various financial indicators and ratios. As the performance of a company affects the movement of its stock value, this research will compare Support Vector Machine and Decision Tree with the One Against All approach in classifying company performance. The features used for the classification of company performance consist of three financial ratios: profitability (ROA), liquidity (CR), and leverage (DER). The labels or targets in the classification are divided into three categories: normal, good, and unfavorable. This research will consider evaluations such as accuracy, cross validation, and confusion matrix. The results of the Support Vector Machine (SVM) algorithm demonstrated an accuracy of 86.67%, while the Decision Tree (DT) algorithm exhibited an accuracy of 93.33%. Consequently, the DT algorithm produced more accurate results than the SVM algorithm in classification.
Implementasi K-Means Untuk Pengelompokan Makanan Cepat Saji Bagi Penderita Penyakit Obesitas Pradvenanta, Yoannes Dion; Prathivi, Rastri
Building of Informatics, Technology and Science (BITS) Vol 6 No 1 (2024): June 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i1.5279

Abstract

One in eight people in the world lives with obesity, a statistic that is worrying as it shows a significant increase compared to 1990. Obesity in adults has more than doubled, and obesity among adolescents and children has quadrupled. One factor in obesity is poor food quality. A lot of people who don't pay much attention to the quality of their food one of them is eating fast food because fast food consumption can be said to be good if the meal frequency is 1 time a week, if more than that and excess is said not good. Thus, there is a need for a fast food grouping model that helps obese people choose fast foods. The K-means algorithm is one of the ideal models for grouping fast foods. The results of the analysis using the elbow method show k=5, then consider three evaluations against the k=5 value: Sum Square Error (SSE), Silhouette Score, and Davies Bouldin Index (DBI). The results were data segmented taking into account the negative and positive nutrient content for obese patients. The data segmentation results found a fairly healthy cluster on label_0 with 244 data and an unhealthy cluster in label_2 with 25 data. From the cluster label_0, 244 of the data could be a healthy fast food choice for obesity patients
Sentiment Analysis of the Palestine-Israel Crisis on Social Media using Convolutional Neural Network Delva, Dwina Sarah; Lhaksmana, Kemas Muslim
Building of Informatics, Technology and Science (BITS) Vol 6 No 1 (2024): June 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i1.5282

Abstract

The issue of Palestine and Israel is currently ongoing and is becoming increasingly heated. The struggle for territory and power is the reason for this conflict, thus attracting the world’s attention, especially that of the the Indonesians. People actively express various views in the form of opinions via social media platforms such as Twitter. Communities are competing to make posts and tweet as a form of support for either party. Various tweets appear, making it difficult to draw conclusions through manual analysis. Therefore, this study employs automatic sentiment analysis to enable mass data processing. The sentiment analysis process uses a Deep Learning algorithm, specifically the Convolutional Neural Network (CNN). Convolutional Neural Network (CNN) is a Neural Network algorithms designed for visual shape processing and developed for classification tasks. Based on the explanation provided, it is expected to provide high accuracy and achieve the designed goals. This sentiment analysis research needs to be conducted because to understand and classify various forms of public sentiment toward the issue of Palestine and Israel, thereby providing an overview of the fluctuations in public sentiment concerning this matter in Indonesia. Outcomes of this investigation found the highest performance was achieved by the Convolutional Neural Network (Oversampling) algorithm with accuracy of 93.85%, precision of 93.76%, recall of 93.95%, and F1-score of 93.86%.
Sentiment Analysis About Legislative Elections using Deep Learning with LSTM and CNN Models Angraini, Nadya Arda; Lhaksmana, Kemas Muslim
Building of Informatics, Technology and Science (BITS) Vol 6 No 1 (2024): June 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i1.5283

Abstract

The election of legislative members is a significant moment from the perspective of democracy, influencing the policies and direction of a country. In the digital era, sentiment analysis regarding the election of legislative members through social media has become increasingly important for analyzing public opinions and providing insights into how people respond to and feel about candidates, parties, or specific issues. The authors of this study employ deep learning methods, specifically Long Short-Term Memory (LSTM) and Convolutional Neural Network (CNN) models, for sentiment analysis related to legislative member elections. These models were developed and trained using preprocessed datasets. The aim of this research is to identify the highest accuracy values of the LSTM and CNN models and to analyze and classify public sentiment regarding the 2024 DPR member election.The results of this study indicate that deep learning methods can provide valuable insights into public sentiment during the 2024 legislative elections. Using a CNN model with a data ratio of 80:20, the proposed model can categorize and identify sentiments with the highest testing accuracy. It is clear that the data ratio, which provides an optimal balance between training and testing data, has a significant impact on model performance. As a result, the CNN model achieves the best results, with an accuracy of 93.27%, an F1 score of 93.19%, precision of 93.52%, and recall of 92.73. This research makes an important contribution by applying the CNN model, which succeeded in achieving the best results in categorizing sentiment, demonstrating the highest test accuracy in analyzing public sentiment towards the 2024 DPR member elections.
El Niño Sentiment Analysis Using Recurrent Neural Network and Convolutional Neural Network Use GloVe Putrisia, Denada; Lhaksmana, Kemas Muslim
Building of Informatics, Technology and Science (BITS) Vol 6 No 1 (2024): June 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i1.5284

Abstract

Sentiment analysis regarding the El Niño climate change is a crucial aspect in understanding public perception and response. This enables deeper identification and understanding of the sentiments evident in online conversations. Sentiment analysis through deep learning approaches using Recurrent Neural Network (RNN) and Convolutional Neural Network (CNN) methods is conducted. RNN is a type of artificial neural network designed to process sequential data such as text or time. On the other hand, CNN utilizes convolutional layers to scan text with filters to capture local features like phrases and keywords determining sentiment. Leveraging GloVe representation technique enables the representation of words in numerical vector form capturing semantic relationships among words, facilitating sentiment analysis related to El Niño on social media. The aim of this study is to evaluate the performance of RNN and CNN methods in classifying El Niño-related sentiment with and without GloVe word representation, and to develop a model that can provide accurate and reliable sentiment analysis results. The contribution of this research indicates that the accuracy of sentiment analysis has been improved and can be a significant reference for further research in the field of text analysis and natural language processing (NLP). This study also emphasizes the crucial role of word representation techniques like GloVe in enhancing the performance of deep learning models. The results of the study indicate that the RNN and CNN methods with the utilization of GloVe provide better sentiment classification related to the El Niño issue in social media data, showing that the use of RNN and CNN models with GloVe features perform better compared to not using GloVe features. The use of the RNN algorithm with 80:20 split ratio testing produced an accuracy score of 94.90%, recall of 94.90%, precision of 94.94%, and F1-Score of 94.85%. Meanwhile, the use of the CNN algorithm with 90:10 split ratio testing produced an accuracy score of 94.61%, recall of 93.61%, precision of 94.69%, and F1-Score of 94.58%. This results in the conclusion that sentiment analysis using RNN modeling with GloVe features has better performance than CNN modeling, with an average accuracy rate of 94.90%.
Sentiment Analysis of TikTok Shop Prohibition Using a Random Forest and Decision Tree Praja, Yudhistira Imam; Muslim L, Kemas
Building of Informatics, Technology and Science (BITS) Vol 6 No 1 (2024): June 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i1.5285

Abstract

This research explores the impact of the closure of TikTok Shop by the Indonesian government on various aspects of the economy, the e-commerce industry, consumer behavior, and social media dynamics. As an e-commerce platform within the TikTok social media application, TikTok Shop has become a significant business information system that collects, provides, and stores information related to electronic buying and selling activities. Understanding the public's reaction to the closure of TikTok Shop is essential because it can influence consumer confidence, market stability, and future regulatory decisions. Public sentiment provides valuable insights into the potential economic and social consequences, guiding policymakers and businesses in making informed decisions. The closure of this platform has elicited both positive and negative reactions from the public, which are widely expressed through social media, especially Twitter. To analyze public sentiment regarding this issue, two relevant machine learning methods were used: Random Forest and Decision Trees. Random Forest is known for its efficiency in data mining and its ability to handle data imbalance in large datasets. Decision Trees offer similar accuracy and can be applied in both serial and parallel modes, depending on the available data capacity and memory. The results of this study are expected to provide in-depth insights into the implications of the closure of the TikTok Shop and the effectiveness of using machine learning algorithms in social sentiment analysis. This research yielded effective results with a 75.24% accuracy, 80.18% precision, 67.06% recall, and 73.04% F1 score.
Tourism and Travel Content Analysis for Market Segmentation using Toxicity and Sentiment Classification in Communalytic Singgalen, Yerik Afrianto
Building of Informatics, Technology and Science (BITS) Vol 6 No 1 (2024): June 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i1.5294

Abstract

This research highlights the significant impact of digital content on shaping tourist perceptions and behaviors, particularly emphasizing the influence of travel vlogs. Utilizing the Tourism and Travel Content Analysis (TTCA) framework, the study analyzed 1,972 review posts out of 2,250, revealing critical insights into viewer engagement and sentiment. Toxicity score calculations indicated prevalent negative interactions, with scores ranging from 0.05542 to 0.86967 for Toxicity, 0.00536 to 0.50704 for Severe Toxicity, 0.01921 to 0.59834 for Identity Attack, 0.03305 to 0.76573 for Insult, 0.03737 to 0.78492 for Profanity, and 0.01075 to 0.48617 for Threat, underscoring the need for compelling content moderation. Sentiment analysis using VADER and TextBlob demonstrated a generally positive reception of travel vlogs, with VADER classifying 3.73% of posts as unfavorable, 19.83% as neutral, and 76.44% as positive. In comparison, TextBlob classified 2.71% of posts as unfavorable, 35.59% as neutral, and 61.69% as positive for English posts. Notably, VADER and TextBlob agreed on sentiment classification for 446 out of 587 posts (75.98%), with a Cohen’s kappa statistic of 0.471, indicating moderate agreement. These findings suggest that well-regulated and thoughtfully designed digital content significantly enhances user engagement and optimizes destination marketing strategies. Future research should incorporate advanced analytical tools and comprehensive data sets to refine these insights further, supporting the development of more targeted and effective marketing efforts in the tourism sector. This study thus contributes to a deeper understanding of digital media's impact on tourism marketing, offering practical recommendations for leveraging content to foster positive and engaging tourist experiences
Analisis Sentimen Masyarakat Terhadap Pinjaman Online di Twitter Menggunakan Algoritma Naïve Bayes Classifier dan K-Nearest Neighbor Afandi, Rival; Afdal, M; Novita, Rice; Mustakim, Mustakim
Building of Informatics, Technology and Science (BITS) Vol 6 No 2 (2024): September 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i2.5300

Abstract

The very rapid development of technology has had a big impact on humans. The influence of technological developments that we can feel is in the financial sector. One thing that is quite popular lately is online loans. Pinjol or online loan is a fast and easy online money lending service via an application or website, with fast approval and disbursement, but often has high interest and short tenors. On Twitter, review comments and information used are stored in text form. One of the processes for retrieving text mining information in the text category is Sentiment Analysis to see whether a sentiment or opinion tends to be Positive, Negative or Neutral in the reviews of Pinjol application user comments. In the data collection results there were 600 initial data, namely 122 Positive reviews, 432 Negative reviews and 43 Neutral reviews. Then the sentiment classification process using the Naive Bayes and K-NN algorithms produces accuracy, precision and recall of 68%; 83% and recall 74% on the Naive Bayes algorithm, while the results of accuracy, precision and recall on K-NN are 72%; 74% and recall 96% with experiments using 80% training data and 20% test data
Perbandingan Algoritma SVM dan Decision Tree Dalam Klasifikasi Kepuasan Pengguna Aplikasi Migo E-Bike di Playstore Al Azkiah, Dina Sakinah; Erizal, Erizal; Hikmah, Fitri Nur
Building of Informatics, Technology and Science (BITS) Vol 6 No 1 (2024): June 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i1.5315

Abstract

Currently, transportation has become an essential need in daily life, and the rapid development of digital technology has had a significant impact on the use of services and interaction with mobile applications, including in the transportation sector. The Migo E-Bike app is the first electric bike rental service application in Indonesia, offering environmentally friendly services to reduce air pollution. This research aims to assess the effectiveness of two data mining algorithms, SVM and Decision Tree, in classifying user satisfaction of the Migo E-Bike app based on reviews and ratings on the Playstore. The research findings indicate that the Decision Tree algorithm performs better than SVM. The Decision Tree achieved an accuracy of 76.39%, with balanced precision and recall for both satisfaction categories. In contrast, SVM exhibited significant imbalance with an overall accuracy of only 51.25%. Therefore, the Decision Tree algorithm is more effective in handling the user rating dataset for the Migo E-Bike app.