International Journal of artificial intelligence research (IJAIR)
Not a member yet
271 research outputs found
Sort by
Analysis of Motivation, Leadership Style, and Communication on the Performance of Members of the Sulawesi Truck Driver Association Kotamobagu Branch through Job Satisfaction As an Intervening Variable
Get PDFThe background of this research is that the performance of members of the Sulawesi Truck Driver Association, Kotamobagu City branch, has not been maximized. There are still accidents on the road or when loading and unloading goods and complaints from transportation users who use the services of members of the Sulawesi Truck Driver Association, Kotamobagu City branch personally which makes losses in terms of material service users. Such performance results in the organization's name becoming bad. The purpose of this study was to analyze motivation, leadership style, and communication on the performance of members of the kotamobagu branch of the sulawesi truck driver union through job satisfaction as an intervening variable. The research stages carried out were field observations, then literature studies, data collection both primary and secondary. Followed by research, data analysis and discussion, after that draw conclusions and provide suggestions. The results showed that the variables of motivation, leadership style, and communication had no effect on the job satisfaction of truck drivers who were members of the Sulawesi truck driver union, Kotamobagu city branch. motivation, leadership style, communication, and job satisfaction have no effect on the performance of truck drivers who are members of the Sulawesi truck driver union, Kotamobagu city branch
Implementation of Tensor Flow in Air Quality Monitoring Based on Artificial Intelligence
No full textChemicals that cannot be controlled today can pollute resources and the environment. Common sources of pollutants are due to public transportation, cigarette smoke, volcanic activity that emits volcanic ash, factory smoke, forest fires, biogas, or carbon dioxide. The purpose of this paper is to monitor air quality, detect air and anticipate pollution levels. With the specified algorithms, three algorithms will be used to create a good and accurate model where four different gasses are predicted: carbon dioxide, sulfur dioxide, and nitrogen dioxide, in this paper, there are four algorithms used for the Air Qualification Index which are Support Vector Regression, Linear Regression, and Ensemble Gradient Boosted Decision Tree. This research also includes quantitative research which is hypothesized to be evaluated against Root Mean Squared Error, Mean Squared Error, and Mean Absolute error, depending on the performance of the measurements made by artificial intelligence, and the lower error value is selected. Based on the algorithm to be predicted in this air quality monitoring, there are 5 air pollutants like Carbon dioxide, Sulfur dioxide, and Nitrogen dioxide, and the sensors to be used are two sensors like PM2.5 and PM10 that can be predicted
Knowledge Graph Construction for Rice Pests and Diseases
Get PDFThe agricultural industry in Indonesia confronts the simultaneous task of augmenting food production to satisfy escalating demand while proficiently handling crop losses caused by pests and diseases. This study introduces a novel approach that leverages knowledge graphs to transform traditional, expert-based knowledge into a dynamic and interconnected system for addressing these agricultural challenges. The study delineates constructing a comprehensive knowledge graph, commencing with data extraction with SPARQL queries, and progressing to ontology design, object property and datatype property specification, and instance generation. The resultant knowledge graph not only serves as an organized archive for pest and disease information but also gives a systematic framework for the integration, analysis, and decision-making of data in agriculture. This knowledge graph adds to the broader junction of data science and agriculture by improving the diagnosis, prevention, and control of rice diseases
Classification of Cervical Cancer Images Using Deep Residual Network Architecture
Get PDFAccording to data from the World Health Organization (WHO), cervical cancer is ranked second, with a high mortality rate in women every year. Cervical cancer is caused by the presence of the Human Papilloma Virus (HPV), which directly attacks the cervix. Additionally, an unhealthy lifestyle can cause attacks of this disease. Several methods can be used to detect cervical cancer early, one of which is Visual Inspection with Acetic Acid (VIA). Through VIA, tests can determine whether patients are infected with the HPV virus. The results of the VIA test can be seen with the naked eye, but medical experts have different opinions about the diagnosis made using their vision. Therefore, to assist medical practitioners in diagnosing the results of VIA, an examination with a technological approach was carried out. Digital imagery was used for the analysis. A medical expert’s Android camera was used with .jpg image format to capture pictures of the VIA test results. In this study, cervical cancer image classification was carried out from the results of the VIA test examination that had been carried out at Hasan Sadikin Hospital, Bandung, with as many as 255 data points for Negative VIA and 65 data points for Positive VIA. In the image processing of the VIA test results, CLAHE images and Canny Edge Detection images are used. Deep learning was used with the ResNet-50 and ResNet-101 architectural models to classify images, and different hyperparameter configurations, such as optimizers, learning rates, batch sizes, and input sizes, were tested. In this study, the best results were obtained using Canny Edge Detection images with hyperparameter configurations using the SGD optimizer with a learning rate of 0.1, a batch size of 32, and an input size of 224 × 224
Selecting the Optimal Location for a New Facility: A PROMETHEE II Analyst
Get PDFThis paper presents a case study of location selection for a new facility using the multi-criteria decision making (MCDM) method, PROMETHEE II. The PROMETHEE II method is a widely used method for solving problems that involve multiple criteria and alternatives. The method allows for the ranking of alternatives based on their overall net flow, which is calculated by weighting and comparing the criteria values for each pair of alternatives. The case study evaluated five different locations based on criteria such as access to transportation, availability of skilled labor, and cost of living. The results of the analysis indicate that C1 was ranked as the most attractive location, with the highest scores for all criteria and the highest overall net flow among all alternatives. A sensitivity analysis was performed to ensure that the results were robust and not sensitive to small changes in the weight of the criteria. The results of this study demonstrate the utility and effectiveness of the PROMETHEE II method in practice and provide valuable insights for further research and practical action
Unveiling the Synergy: Exploring the Intersection of Artificial Intelligence, Digital Management Information Systems, and Marketing Management in a Qualitative Research Study
Get PDFThis study investigates the integration of Artificial Intelligence (AI), Digital Marketing Information Systems (DMIS), and Marketing Management to enhance decision-making processes in marketing. The research aims to explore the extent of augmentation in marketing decision-making, identify indications of objectiveness in AI-driven analytics, and propose solutions to ensure transparency and accountability. Methodologically, the study conducts a systematic literature review to synthesize existing research on the topic. Findings suggest that AI technologies offer advanced analytics capabilities, enabling marketers to gain deeper insights into consumer behavior and market trends. However, concerns regarding biases in AI-driven analytics and challenges in data integration and dissemination are identified. The study underscores the importance of interdisciplinary collaboration, transparency, and explainability in AI algorithms to mitigate biases and enhance objectiveness. Moreover, it highlights the need for robust data governance policies and talent development initiatives to foster a culture of data-driven decision-making. The research contributes to theoretical understanding by redefining marketing practices through AI integration and offers practical insights for organizations to leverage AI, DMIS, and Marketing Management effectively
Integrating Sentiment Analysis and Quality Function Deployment for Product Development
Get PDFThe development of technology and media has made online data reviews a promising data source. Through machine learning utilizing text processing, data analysis of Ventela Public Low product reviews can be carried out—sentiment analysis is used to find class groups from each data. The classification algorithm is Naïve Bayes and Support Vector Machine (SVM). A classification model with the best performance and accuracy values will be selected. Word association is then applied to obtain information from the required class. Quality Function Deployment (QFD) is a tool used to assist designers in developing products. The results of the integration of sentiment analysis into QFD show that sentiment analysis produces information by the provisions of the QFD method and can support the product development process in terms of the amount of data various data topics and reduces the subjectivity of designers at the stage of determining Voice of Customer (VOC) and performance values of products and competitor
Face Detection Analysis of Digital Photos Using Mean Filtering Method
Get PDFFace detection in digital photos aims to get the face area in the digital photo. Usually, a lot of noise occurred when detecting faces in digital photos. This study applies the mean filtering method to improve digital photos by reducing noise. The accuracy of the mean filtering method is calculated using a confusion matrix, while the ability of this method is measured using the parameters of Mean Square Error (MSE) and Peak Noise to Signal Ratio (PNSR). Viola-Jones method was used to detect faces in this research. This method was chosen because it is one of the face detection procedures with high accuracy and good computational ability. Testing the mean filtering method obtained the lowest MSE of 9.33, while the highest PNSR of 14.37. The accuracy obtained by the mean filtering method using confusion is 90%. Based on these results, it can be concluded that the mean filtering method is feasible to be used in the case of face detection in digital photos
The Effect of Corporate Governance and Corporate Social Responsibility on Tax Avoidance in Manufacturing Companies Listed on the IDX
Get PDFThis study was conducted with the aim of: (1) examine the effect of the proportion of independent commissioners on tax avoidance; (2) examine the effect of audit quality on tax avoidance; (3) examine the effect of institutional ownership on tax avoidance; (4) examine the effect of managerial ownership on tax avoidance; (5) examine the effect of CSR disclosure on tax avoidance. The type of data used in this study is secondary data taken from manufacturing company reports from 2015-2018. The data is obtained from the Indonesia Stock Exchange which can be accessed on the official website of the Indonesia Stock Exchange (www.idx.co.id) as well as the company website of the related company. The data collection technique used is the documentation method. The data analysis method used is multiple linear regression analysis. The results of this study indicate that: (1) ownership of the proportion of independent commissioners has a negative and significant effect on tax avoidance; (2) audit quality has a negative and significant effect on tax avoidance; (3) institutional ownership has a negative and significant effect on tax avoidance; (4) managerial ownership has a negative and significant effect on tax avoidance; (5) CSR has a negative and significant effect on tax avoidance
Association Rule Mining System in Analyzing The Use Pattern of Drugs by Using Apriori
Get PDFData mining is a process to support decision making in finding information patterns in the data. In this study, Association Rule Mining will be implemented as one of the data mining techniques to analyze data and assist data of scientists in compiling raw data, formulating it and recognizing various patterns through a priori algorithms. The method used in this study is the Cross Industry Standard Process for Data Mining (CRISP-DM) Method by analyzing drug use patterns in health centers. The results of the study shows that by using the apriori algorithm, it found patterns and rules of widely used drugs that will provide recommendations in supporting decision making by health centers to submit drug procurement so that they can improve the quality of health services and minimize the risk of shortages or excess drug supplies and help health centers in optimizing drug inventory management. The results of the analysis using the apriori algorithm on the combination pattern of 2 itemsets produced 2 association rules for drug use, they are "If using Amoxicillin caplets 500 mg, then you will use paracetamol" with a confidence value of 80% and "If using Dexamethasone tablets 0.5 mg, then you will use Ascorbic Acid (Vit C) tablets 50 mg" with a confidence value of 100%