AGENTS: Journal of Artificial Intelligence and Data Science

ANALISIS SENTIMEN KOMENTAR PENGGUNA TERHADAP GAME MOBA LOKAPALA DI GOOGLE PLAY STORE MENGGUNAKAN ALGORITMA SUPPORT VECTOR MACHINE

akbar

Di era modern ini, game banyak dipengaruhi oleh kemajuan teknologi. Perkembangan game yang semakin kompleks dan menawan dapat dimainkan secara terhubung menggunakan koneksi internet oleh jutaan pemain di seluruh dunia. Industri game di Indonesia menunjukkan perkembangan signifikan dengan munculnya berbagai game dari pengembang lokal, salah satunya Lokapala, game Multiplayer Online Battle Arena (MOBA) yang mengusung keunikan budaya Indonesia. Meskipun demikian, game ini menerima berbagai tanggapan dari pengguna di Google Play Store. Penelitian ini bertujuan untuk menganalisis sentimen pengguna terhadap game Lokapala di Google Play Store menggunakan algoritma Support Vector Machine (SVM). Data ulasan pengguna dikumpulkan dan dipra-proses melalui tahap-tahap seperti pembersihan data, tokenisasi, penghapusan stopwords, dan stemming. Setelah itu, fitur-fitur diekstraksi menggunakan metode TF-IDF. Hasil analisis menunjukkan bahwa SVM dengan kernel Radial Basis Function (RBF) berhasil mengklasifikasikan sentimen pengguna dengan akurasi sebesar 90% dari total 300 ulasan yang dianalisis. Proses ini tidak hanya membantu dalam memahami persepsi pengguna secara keseluruhan, tetapi juga mengidentifikasi aspek-aspek tertentu dari game yang mendapatkan apresiasi atau kritik. Dengan demikian, pengembang game dapat memanfaatkan hasil analisis ini untuk meningkatkan kualitas dan kepuasan pengguna, serta memperkuat daya saing game di pasar lokal maupun global.

Decision Support System Recommendations for Providing Social Aid with Multi Objective Optimization By Ratio Analysis (MOORA)

Zulhisham Ramli — 2024-02-28

Cash Direct Assistance (BLT) is one of the conditional assistance programs from the government as a form of poverty alleviation program. The selection process of potential recipients of BLT in Lamatti Riaja Village, Sinjai Regency, it is not entirely accurate as it is still done manually, resulting in many recipients not meeting the criteria. Based on this, research is conducted to design a decision support system that will facilitate the automatic checking of data for eligible residents who are entitled to BLT funds for each disbursement. This aims to make the selection process more objective, time-efficient, and minimize potential errors in selecting BLT recipients. In this research, the Multi Objective Optimization By Ratio Analysis (MOORA) method is employed. The calculation process utilizes the MOORA algorithm, and the implementation of the system is in the form of a website using the System Development Life Cycle design method, providing good and accurate results. The testing method used is Black Box testing. This research produces a Decision Support System website with the implementation of a data management subsystem using MySQL. The simulation results of the BLT recipient data calculation using the MOORA algorithm minimize errors in the selection process for potential BLT recipients.

Optimization of Student Graduation Predictions on Time Using Binning And Synthetic Minority Oversampling Technique (SMOTE)

Faidhul Rahman — 2024-02-28

On-time student graduation is a situation where a student graduates from their educational program at the time planned or determined by the relevant educational institution. This research aims to optimize predictions of student graduation on time using the Binning method to group variables into discrete categories and Synthetic Minority Oversampling Technique (SMOTE) to overcome class imbalances in the dataset. Data containing several variables was analyzed using the Naïve Bayes, Decision Tree and Random Forest machine learning algorithms. Model evaluation is carried out using metrics such as precision, Recall, accuracy, and F1-score. The results confirm that the combination of Binning and SMOTE has a significant impact on increasing prediction accuracy. It is hoped that the results of this research can contribute to increasing the accuracy of predicting student graduation on time. By optimizing the use of Binning and SMOTE, it is hoped that the prediction model can overcome the problem of data imbalance and provide more accurate information to higher education institutions to take the necessary preventive actions to increase student graduation rates and become a reference for similar research in the future.

Public Sentiment Analysis Towards the Implementation of 2024 Election Preparations Using the Maximum Entropy Method

Indra Ismawan Indra Ismawan — 2024-02-28

The General Election Commission (KPU) and the Election Supervisory Agency (Bawaslu) are tasked with coordinating and overseeing elections throughout the territory of Indonesia. Nevertheless, elections are not without various disputes, including violations of ethical codes and administrative issues. General elections are an important way for the public to exercise their political participation rights. In the era of information technology, the public is increasingly active in expressing their opinions and sentiments through social media such as Twitter. Utilizing data from social media, the maximum entropy method is employed to classify opinions found in tweets based on entropy values. This research also aims to measure the accuracy level of the maximum entropy method in sentiment classification. From the results of this study, using approximately 1,400 data, an accuracy rate of 87.08% was obtained, which is quite good and highly feasible for further development to achieve better results

Sentiment Analysis Of Online Loan Application Reviews Using Support Vector Machine (SVM) Method

ST. Aminah Dinayati Ghani — 2024-02-28

Online loans, abbreviated as "Pinjol," refer to the practice of lending money online through applications or websites without involving traditional financial institutions such as banks or other traditional creditors. Examples of applications in this field include AkuLaku and Kredivo. These apps operate in the e-commerce and credit provision sectors in Southeast Asia, including Indonesia. Despite their operational strengths, these applications have both advantages and disadvantages, leading to positive and negative reviews on platforms like the Play Store. The research aims to identify positive and negative reviews within these online loan applications that can influence users' decisions when choosing a particular app. SVM classification technique is employed to analyze positive and negative sentiments from these reviews. The accuracy results obtained after sentiment analysis for Kredivo are 81%, while for AkuLaku, it is 75%. A higher accuracy value indicates a better ability of the model to predict sentiments correctly. Visualization of impactful words based on word frequency is presented in the form of a Word Cloud. Therefore, based on the sentiment analysis conducted using the SVM model, the author suggests choosing the Kredivo app when selecting an online loan application, as the analysis indicates that Kredivo has better quality compared to AkuLaku.