International Journal of artificial intelligence research (IJAIR)
Not a member yet
271 research outputs found
Sort by
Detection of SQL Injection Attacks on MariaDB Using Hybrid Long Short-Term Memory
Full text linkThis study discusses the development of a SQL Injection attack detection system using the Long Short-Term Memory (LSTM) deep learning model. SQL Injection is a serious security threat to web applications that exploits vulnerabilities in user input to manipulate databases. The LSTM model was chosen due to its ability to process sequential data, which is relevant for analyzing the patterns and structure of SQL queries that are susceptible to attacks. The process begins by collecting and combining datasets from various sources, performing preprocessing to handle duplicate data, missing values, and gibberish queries, as well as analyzing the distribution of query lengths. The textual query data is then converted into a numerical representation through tokenization and padding. The processed dataset is divided into training and testing data. The Bi-directional LSTM model architecture is built with embedding, LSTM, dropout, and dense layers. The model is trained using the training data and its performance is evaluated using the test data, producing metrics such as accuracy, precision, recall, and F1-score. Evaluation results on the test data show a model accuracy of 99.99%, with precision of 99.99%, recall of 99.99%, and F1-score of 99.99% in distinguishing between normal queries and SQL Injection queries. The trained model and the tokenizer used are then saved for further testing purposes. This research demonstrates that the LSTM-based approach is highly effective in detecting SQL Injection attacks with high accuracy. Thus, the model can be deployed at the production level or production server
Modeling the Driving Factors of Educational Technology Innovation in Indonesian Universities: A Hybrid ISM–ANP Approach
Full text linkThis study aims to model the critical enablers driving technological innovation in higher education institutions in Indonesia by integrating Interpretive Structural Modeling (ISM) and Analytic Network Process (ANP). The hybrid approach provides both structural and quantitative insights into the interrelationships among eight identified enablers: policies and regulations, digital infrastructure, faculty competence, technology incentives, industry collaboration, student literacy, innovation culture, and data security. The ISM results classify policies and regulations and digital infrastructure as driving factors that form the foundational layer of innovation ecosystems. Meanwhile, faculty competence, technology incentives, and industry collaboration serve as linkage factors that bridge strategic policies and operational implementation, whereas student literacy, innovation culture, and data security emerge as dependent factors representing the system’s outcomes. The ANP results reinforce the ISM structure, revealing that policies and regulations (0.215) and digital infrastructure (0.187) have the highest influence, followed by faculty competence (0.142) and industry collaboration (0.130). The combined ISM–ANP framework demonstrates that sustainable educational technology innovation requires a synergistic interaction between governance, human resources, and digital culture. The findings provide a comprehensive model that can guide universities and policymakers in formulating evidence-based digital transformation strategies within the Indonesian higher education contex
Loan Origination System Implementation Model and Credit Business Process Value Creation in Improving Business Performance
Full text linkThe banking industry as a credit provider has changed substantially over the last century, namely that when a bank will lend money with collateral as collateral, the process is carried out in stages involving many bank officers. Currently, almost all operating financial institutions have been digitized, especially the credit application process. More and more customers are choosing digital loans over traditional loans because of the benefits they provide. The aim of this research was to analyze the picture of Business Performance which includes Customer Requirements, Digital Leadership, Loan Origination System Implementation and Credit Business Process Value Creation which influence it. The research method uses quantitative research with descriptive and verification research types. The research population was 64 branch offices which were the analysis units using a saturated sampling technique. The research instrument uses a questionnaire and data analysis techniques to determine the correlative relationship in this research using Partial Least Square. The research results show that Business Performance with the dominant dimension, namely Financial Performance, is influenced by Loan Origination System Implementation with the dominant dimension, namely Document Management and Credit Business Process Value Creation with the dominant dimension, namely Responsiveness. Loan Origination System Implementation and Credit Business Process Value Creation are influenced by Customer Requirements with the dominant dimension, namely Customer Expectation and Digital Leadership with the dominant dimension, namely Customer Focused
Igniting Determinants Of Tourism Service Purchase Decisions
No full textThis research aims to analyze the influence of promotion and price on purchasing decisions for tourism services in Lawangsewu, Semarang. This research employed quantitative methods with regression analysis. The author used primary data and purposive sampling, resulting in a sample size of 95 respondents. Partial Least Squares (PLS) analysis was used for this research. The data analysis revealed that promotion and price significantly influence purchasing decisions for tourism services in Lawangsewu, Semarang. Future researchers investigating related issues can use this study's findings as a guide. This study can also be used as assessment data to help business managers focus more on promotions to facilitate marketing campaigns with a wider audience
Portrait of Friendship among Catholic Youth: A Preparatory Step toward Catholic Marriage at St. Fransiskus Asisi Laverna Gunungsitoli and Kristus Gembala Baik Parishes
No full textThis study analyzes friendship among Catholic Youth (Orang Muda Katolik—OMK in Bahasa Indonesia) as a crucial factor in forming emotional, spiritual, and social readiness for Catholic marriage. The qualitative study was conducted in two parishes—St. Fransiskus Asisi Laverna Gunungsitoli and Kristus Gembala Baik—using in-depth interviews, participant observation, and document analysis. The findings indicate that healthy, value-based friendships within the Catholic Youth community contribute significantly to fostering responsible relationships as a foundation for marriage. Key aspects supporting relational maturity include trust, open communication, mutual respect, and active participation in religious activities. These factors reinforce the psychosocial aspects necessary for building stable and meaningful relationships before entering into marriage. This study highlights the need for a more systematic approach by the Church in integrating friendship development into premarital preparation programs, particularly in pastoral care for Catholic youth. Thus, friendship is not merely a social experience but serves as a pathway toward a holy and committed family life. Further research is encouraged to explore how faith-based education can strengthen friendships as a foundation for mature relationships
AI-Powered Decision Support Systems for MSMEs Growth Strategies in Emerging Markets
No full textMicro, Small, and Medium Enterprises (MSMEs) in emerging markets face significant challenges in strategic decision-making due to limited resources and dynamic environments. This study aims to explore the role of AI-Powered Decision Support Systems (AI-DSS) in supporting growth strategies using the dynamic capabilities theory. The study involved 250 food processing business actors who have adopted or considered implementing AI-DSS, with data analysis conducted using Partial Least Squares Structural Equation Modeling (PLS-SEM). The results indicate that dynamic capabilities and dynamic environments significantly influence the adoption of AI-DSS, ultimately enhancing business performance. Furthermore, dynamic environments were found to moderate the relationship between dynamic capabilities and AI-DSS adoption, indicating that adaptability to external changes is crucial in maximizing the benefits of AI technology. This research contributes theoretically to the literature on dynamic capabilities and technology while offering practical implications for food processing entrepreneurs and policymakers in emerging markets. The study highlights the importance of developing adaptive capabilities and leveraging AI technology to improve competitiveness amidst ever-changing business dynamics
A Multi-Feature Fusion Framework for Sentiment Analysis Based on Textual and Affective Signals
Full text linkSentiment analysis of social media content, particularly on platforms like Twitter, presents significant challenges due to the informal, brief, and context-dependent nature of user-generated text. Traditional lexicon-based and shallow machine learning approaches often fail to capture nuanced sentiment expressions, especially in the presence of slang, abbreviations, sarcasm, and emotionally charged language. To address these limitations, this paper proposes a novel tri-stream feature fusion framework that integrates contextual semantics, sequential dependencies, and affective signals for robust sentiment classification. The framework employs RoBERTa to extract rich contextual embeddings, Bidirectional Long Short-Term Memory (BiLSTM) networks to capture word-order and temporal patterns, and lexicon-based emotion vectors to enhance emotional cue detection. These heterogeneous features are concatenated at the representation level to form a comprehensive feature space, which is subsequently used to predict sentiment polarity via a fully connected neural network classifier. Extensive experiments conducted on the Sentiment140 dataset, comprising 1.6 million labeled tweets, demonstrate that the proposed approach significantly outperforms conventional baselines and recent hybrid models, achieving an accuracy of 92.1%. Additionally, ablation studies and misclassification analyses reveal each feature stream’s complementary contributions and highlight challenges in detecting sarcasm and implicit sentiment. Future work will integrate sarcasm-aware components and external knowledge sources to further enhance model interpretability and robustness
Interactive Dashboard Development for Student Performance Monitoring: Integrating Academic and Socio-Demographic Data
Full text linkStrategic decision making in institutional settings is often constrained by the fragmentation and heterogeneity of data across multiple sources. This study addresses this critical gap by developing and validating an interactive web-based dashboard designed to consolidate and transform heterogeneous institutional data from seven distinct sources into actionable insights. A complex feature engineering pipeline was necessitated, involving comprehensive data integration and structural consistency checks. Techniques like Text Normalization and Feature Mapping were applied to clean over a lot of inconsistent entries, alongside Feature Binning and Extraction to generate analytically robust metrics. The system was implemented using Python for data processing and ReactJS for the dynamic interface, and its viability was validated via structured User Acceptance Testing (UAT). The subsequent descriptive analysis provided key insights into student demographics, geographical reach, and enrollment compliance across academic levels. Crucially, the comprehensive UAT resulted in an outstanding overall acceptance score of very worthy. However, feedback analysis indicated a dominant user focus on visual aspects, with noted complaints regarding the suboptimal color scheme and contrast impacting user experience. The findings confirm that complex feature engineering is a viable and effective strategy for transforming fragmented institutional data into an immediately deployable strategic resource. This system offers a validated blueprint for data consolidation in higher education. Future work is accordingly directed toward revising the color palette and contrast ratios to enhance visual clarity and user experience, alongside continuous optimization of data completeness to maintain the dashboard’s utilit
Determining Quality of Service (QoS) of End-User Internet Networks with Data Sniffing and Classification Algorithms
No full textThe development of telecommunications technology in this world has changed very rapidly. Changes are made to access technology using the transmission media, which uses fiber optic technology, which has the advantage of being free from interference, large and fast data delivery capacity. An Internet Service Provider (ISP) is a provider of construction services and management of network infrastructure that always meets customer needs. Customer satisfaction with the services provided by ISP is also important in the era of increasingly tight market competition. Quality of Service (QoS) testing in internet networks needs to be done so that customers get optimal service. This study analyzes the quality of internet networks with fiber optic media on the end user side with the data sniffing method using Wireshark software that records video data traffic on the YouTube platform. The results of the data recording are processed using the QoS method with Throughput, Packet Loss, Delay, and Jitter parameters. The QoS assessment index is divided into Excellent, Good, Fair, and Poor classes according to the TIPHON standard. Data from these parameters is classified using the Naive Bayes, KNN, and Decision Tree methods. The results of applying the algorithms show the highest Accuracy value in the Decision Tree algorithm of 97%, while the highest Precision and Recall are in the KNN algorithm with values of 94% and 85%
Application of Convolutional Neural Network Based on ResNet18 for Alzheimer Disease Classification
No full textAlzheimer's disease is a form of progressive dementia that significantly impacts the quality of life of patients and their families. Early detection based on Magnetic Resonance Imaging (MRI) can support faster and more accurate diagnosis, but manual classification requires high expertise and is subjective. This study aims to develop an Alzheimer's MRI image classification model using a Convolutional Neural Network (CNN) based on ResNet18 with transfer learning to classify data into four categories: Mild Demented, Moderate Demented, Non-Demented, and Very Mild Demented. The MRI dataset was processed through pre-processing involving 128×128 grayscale conversion, pixel intensity normalization, and class balancing using class weighting. The model was trained using the Adam optimizer (lr=0.0001) with Early Stopping (patience=7) over 50 epochs. Evaluation using the validation set showed that the model achieved high accuracy for the Non-Demented class. The result indicates that ResNet18 with transfer learning can achieve an accuracy of 94.4%, making this model an effective approach for medium-scale classification of Alzheimer's MRI images