Journal of Applied Intelligent System

Implementation of Feature Selection Chi-Square to Improve the Accuracy of the Classification Model Using the Random Forest Algorithm on Coronary Artery Disease

2024-04-02T07:04:43+00:00

Coronary heart disease is a disease in which the occurrence of blockages in the blood vessels in the heart. Coronary heart disease is a fatal disease, it is better to get as much information about this disease as possible. Data Mining can classify whether a person has heart disease or not based on symptoms. Data mining builds a model that can predict whether a person has heart disease or not. How well a model performs classification can be determined from its accuracy value, but this accuracy value can still be improved. Increasing the accuracy value can be done by performing Feature Selection. The research object used in this research is a dataset about coronary heart disease obtained from the Kaggle website. The classification method used in this modeling is the Random Forest algorithm to classify whether a person has coronary heart disease or not. The Random Forest Algorithm is a classification algorithm consisting of Decision Trees for classifying. The Random Forest algorithm is used because it has been proven to produce good accuracy in several previous studies. The Feature Selection method used in this modeling is the Chi-Square hypothesis test to determine whether there is an effect of each independent variable on the dependent variable. This research compared the value of modeling accuracy without using Feature Selection with modeling using Feature Selection. The result of this study is that the model without Chi-Square Feature Selection produced an accuracy value of 96,05% and the model with Chi-Square Feature Selection produced an accuracy value of 97,33%.

Utilization Of Principal Component Analysis To Improve Emotion Classification Performance In Text Using Artificial Neural Networks

2024-04-02T07:04:44+00:00

Emotions, being transient and variable, differ across locations, times, and individuals. Automatic emotion identification holds significant importance across various domains, such as education and business. In education, emotional analysis contributes to intelligent electronic learning environments, while in business, it aids in assessing customer satisfaction with products. This study advocates the application of Principal Component Analysis (PCA) to enhance the performance of text emotion classification using the Artificial Neural Network (ANN) method. PCA, a pattern identification method, reduces text dimensions, improving the classification process by determining word similarities. PCA offers the advantage of dimension reduction without compromising information integrity. The classification approach involves two stages: one after PCA dimension reduction and the other without PCA post TF-IDF stage. The study's conclusive findings, incorporating PCA in ANN classification, demonstrated a notable increase in recall for the happy class, reaching 0.92 compared to the pre-PCA score of 0.91. Furthermore, precision in the sadness class improved to 0.90, surpassing the pre-PCA precision of 0.80. This affirms the efficacy of integrating PCA in enhancing the accuracy and performance of emotion classification in text analysis.

Predicting Gold Price Movement Using Long Short-Term Memory Model

2024-04-02T07:04:44+00:00

Gold, as a valuable commodity, has been a primary focus in the global financial market. It is often utilized as an investment instrument due to the belief in its potential price appreciation. However, the unpredictable and complex movement of gold prices poses a significant challenge in investment decision-making. Therefore, this research aims to address this issue by proposing the use of the Long Short-Term Memory (LSTM) model in time series analysis. LSTM is a robust approach to understanding patterns and trends in gold price data over time. In the context of time series analysis, historical gold price data includes daily, weekly, and monthly datasets. Each model with its respective dataset is useful for identifying patterns in gold prices. The daily model achieves an MSE of 452.2284140627481 and an RMSE of 21.26566279387379. The weekly model achieves an MSE of 1346.1816584357384 and an RMSE of 36.69034830082345. The monthly model achieves an MSE of 11649.597907584808 and an RMSE of 107.93330305139747. With these RMSE results, the LSTM model can predict gold prices effectively. Based on the trained models, it can also be concluded that gold prices exhibit long-term temporal dependence.

Counselor Application Frontend with Personality- matching Using Android-Based K-Means Clustering Algorithm

2024-04-17T13:25:23+00:00

Education is one of the most important things for people to have. Many people are competing for education to increase their abilities. Technology plays a big role in developing access to education to make it easier with many online applications and online classes education becomes easier. However, there are still many unresolved problems in this field of education, namely the emergence of the phenomenon of incompatibility between educators and students so that student interest decreases dramatically because of this. And also, the lack of learning materials taught that are not school subjects such as programming. Therefore, the author and team designed an application where this application can find students a learning mentor outside of school so that they can increase their knowledge. The application also provides a matching feature based on the student's personality so that the student can find a suitable tutor.

Optimization of Yolov5 Hyperparameter Using Adam Optimizer in Vehicle Object Detection

2024-04-02T07:04:46+00:00

Utilization of computer vision can be applied in various aspects of daily life, reducing dependence on human labor. One of its implementations is in industry, such as in the production process of motorized vehicles, to sort or classify parts or goods. The computer vision process involves many stages, such as image capture, image processing, image analysis, image recognition, and decision-making. In the automotive industry, computer vision has been used in autonomous or driverless electric vehicles, as well as in creating intelligent transportation systems. To detect objects in real-time, one of the options that can be used is to use the YOLO algorithm, which can detect objects in one stage with predictions of bounding boxes and class probabilities simultaneously. However, although YOLO has good performance, the architecture has some drawbacks, such as complexity and complicated hyperparameter congurations. To remedy this, the Adam optimization algorithm was introduced, which combines the momentum and RMSprop algorithms to adjust the learning rate adaptively and provide faster convergence in model training. This is evidenced by an increase in the value of mAP on Yolov5. These results prove that the Yolov5 method with Adam`s optimization is better than the Yolov5 method without optimization.

Optimization Of Neural Network Method Using Chi-Square Feature Selection In Poverty Data Classification

2024-04-02T07:04:46+00:00

Poverty is a fundamental problem that has become the center of attention in several aspects, for example from the government. The government needs poverty data and analyzes it to determine which poverty alleviation programs should be delivered to the right target or the poor. The aim of this study is to determine the accuracy of the classification of poverty in Batang District using the Neural Network method using the chi square feature selection. The dataset used in this study uses poverty data sourced from the Batang district BPS based on the results of the Susenas survey (National Economic Survey) for the 2022 time period. The results of this study indicate that the accuracy obtained for poverty classification using a neural network is 96.38% , with a precision value of 100%, and a recall value of 89.38%. Whereas when using a neural network with feature selection chi square, it gets an accuracy value of 93.68%, with a precision value of 91.07%, and a recall value of 90.26%. The contribution of this research is to develop a neural network method using feature selection chi square to improve the results of the accuracy of the classification is not poor or poor.

Person Re-Identification Using CNN Method With Combination of SVM and Semantic Segmentation

2024-04-02T07:04:47+00:00

Abstract – Person re-identification is a mechanized procedure of video investigation which has been widely studied in contemporary years. Research problems that are often raised in the field of a person's re-identification research are characteristic representations that are easily affected by closure (abhorrent to other objects). Furthermore, after extracting local features by means of a boundary box, the background image still contains and does not focus on the human body parts. This study comes up with a method combination of CNN, SVM classification, and semantic segmentation. CMC (Cumulative Matching Characteristics) and mAP (mean Average Precision) are measurements of assessment that will be utilized to measure the operation of re-identification. The ResNet + SVM + SSP-ReID technique performed best in the Market dataset, with a CMC increase of 3-10% (rank-1 through rank-20). The Market and CUHK03 (D) datasets both showed improvements of 1-4.1% in mAP.

Keywords Person re-identification; Feature extraction; CNN; SVM; Semantic segmentation;

Classification Email Spam using Naive Bayes Algorithm and Chi-Squared Feature Selection

2024-04-29T09:06:54+00:00

Spam email is a problem that disturbs and harms the recipient. Machine learning is widely used in overcoming email spam because of its ability to classify emails into spam or non-spam. In this research, the Naïve Bayes algorithm is initiated with the Chi-Squared selection feature to classify spam emails. So that the implementation is able to increase accuracy for better performance in classification. The feature selection method is used to direct the model's attention to features that are related to the target variable. In this study, the chi squared feature uses a value of K = 2500, with an accuracy of 98.83% which shows an increase in model performance compared to previous research. So that the Naïve Bayes model with the Chi-Squared selection feature is proven to provide better performance.

Completing Sudoku Games Using the Depth First Search Algorithm

2024-04-02T07:04:48+00:00

Sudoku is a digital game that is included in the type of logic-based puzzle game where the goal is to fill in the puzzle with random numbers. Therefore, in this research it is proposed to use Artificial Intelligence which contains the Depth First Search Algorithm to track the number of possible solutions that lead to only one so that it becomes efficient. This game has different levels of difficulty such as easy, medium and difficult. The time and complexity of execution will vary depending on the difficulty so it is proposed to use Android Studio software. The experimental results prove that there is an increase in playing the Sudoku game quickly and accurately by applying the Depth First Search Algorithm method. This is proven by the ability to complete this game using the Depth First Search Algorithm using the Android Studio programming language. The average time at the easy level is 11:04 minutes, at the normal level is 10:52 minutes, at the hard level is 25:46 minutes, and at the extreme level is 38 minutes.

Analysis of Inter-Subject and Session Variability using Brain Topographic Map

2024-04-02T07:04:49+00:00

The study described investigates the application of Brain-Computer Interface (BCI) technology, focusing on Motor Imagery (MI) signals which enable individuals to control movements through mental visualization. A major challenge in this field is accurately distinguishing between different movements, particularly when dealing with data from multiple subjects and recording sessions, known as inter-subject and inter-session variability. To address this, the authors employ the Wavelet Packet Transform-Common Spatial Patterns (WPT-CSP) method to enhance the resolution of MI signals. They visualize the results using Brain Topographic Maps (Topomaps) to depict brain activity during MI tasks, facilitating the analysis of variability across subjects and sessions. Utilizing dataset 2a from the Brain-Computer Interface Competition (BCIC) IV, the study demonstrates the efficacy of this approach in identifying variability patterns. This research holds promise for improving BCI technology applications in various domains, and future work could explore refining signal processing techniques and validation on larger datasets. Topomap.

Data Mining Application Analyzing Customer Purchase Patterns Using The Apriori Algorithm

2024-04-02T07:04:50+00:00

The study aims to implement Data Mining with Apriori Algorithm and Association Methods (shop cart analysis) to analyze the sales pattern of Kaffa Beauty Shop stores as a case study. Sales information obtained from stores is used to find out the repeated buying habits of cosmetic products. This analysis provides store owners with valuable information to make more useful decisions about product inventory management, marketing strategies, and other aspects of their business. The Apriori Algorithm implementation follows steps including data preprocessing, subsetting, frequent dataset search, and strong association rules (strong Association Rules). The results of the analysis show that there are important purchasing patterns among some cosmetic products that can be the basis of a more effective sales strategy. The study helps understand how data mining and Apriori Algorithms can be applied in business contexts such as Kaffa Beauty Shop stores. Therefore, the results of this analysis are expected to contribute greatly to improving business efficiency and optimizing marketing strategies for store owners and stakeholders. The research is also expected to show the enormous potential of data analysis to support optimal business decision making.

Naive Bayes Sentiment Analysis Study On Street Boba And Gildak Kediri Consumer Reviews

2024-04-02T07:04:51+00:00

Streetboba & Gildak Kediri outlet is a restaurant that serves a variety of Korean food menus and various kinds of drinks with boba and jelly toppings that are sold at low prices that suit the student's budget. This restaurant is located in East Java province which is precisely on Jalan Yos Sudarso No.43, Kediri City. With technological advances that continue to grow to affect various aspects, especially in the business and industrial world. Sentiment analysis is a technology that extracts or manages text to be expressed using text that can also be classified into positive and negative polarity. Consumer reviews are a form of communication that occurs in the sales process, the stage where potential buyers receive an explanation of the product posted and buyers receive reviews that explain the advantages or disadvantages of purchasing the product. In this study, sentiment analysis was conducted based on consumer opinions regarding social media accounts. The study aimed to use social media data to assess the service, cleanliness and quality of products offered by categorizing companies as having positive and negative reviews. To classify sentiment, the Naive Bayes method is used, which combines survey data collection methods, questionnaires, and observation data.

Customer Segmentation Using K-Means Clustering with RFM Method (Case Study : PT. Dewangga Travindo Semarang)

2024-04-02T07:04:54+00:00

PT. Dewangga Travindo is a company that operates in the field of travel services which includes tours, travel, and Hajj and Umrah pilgrimages which is based in the city of Semarang and has received permission from the Ministry of Religion No. D/606 of 2013. Every year there is always an increase in sales of services. Hajj and Umrah. The higher transaction activity every day results in a large buildup of data in the database which will only become data waste. The ability to process data is increasingly sophisticated using data mining, which is an activity of looking for relationships between items to obtain patterns as information to assist in decision making. However, considering the large number of competitors offering the same services, it is necessary to increase competitiveness to overcome market segmentation at PT Dewangga Travindo. For this reason, this research was carried out which aims to overcome customer segmentation using the Clustering method with the K-Means algorithm which produces a visual cluster model with RStudio tools using RFM attributes applied to carry out segmentation. The data used in this research is data on Hajj and Umrah pilgrims in the 2018-2020 period.

Mobile-Based Interactive Learning Media Design Using Augmented Reality Concepts

2024-04-02T07:04:55+00:00

One of the natural disasters that currently occurs frequently in Indonesia is earthquakes. An earthquake is a situation where the earth shakes due to volcanic activity, or collisions due to the movement of the earth's plates. Earthquake activity causes many problems. One of them is liquefaction. Liquefaction is an event that shows a loss of soil shear strength caused by an increase in pore water pressure. This occurs because the earthquake load occurred so quickly and briefly. Liquefaction is a description of the effects of an earthquake so that the soil layer loses its strength. Not many people know the process of earth movement which is the precursor to earthquakes. Especially people who have hearing limitations. For this reason, education is needed to be conveyed to the public. Educational techniques in virtual form will be a special attraction for people, especially deaf people. Augmented Reality (AR) technology is a technology where virtual objects and real objects are combined. In this research, an application was produced in the form of learning media that utilizes Augmented Reality (AR) technology with the Marker Based Tracking method. To make learning media more interactive, the technology was developed using the MDLC (Multimedia Development Life Cycle) method. The results of the tests carried out concluded that the application could be used as an interactive learning medium to increase knowledge about the occurrence of earthquakes and the effects of liquefaction for the community

Web-Based Public Street Lighting Complaint Application with Realtime Whatsapp Notification Using Prototype Method in Pemalang Regency

2024-04-02T07:04:55+00:00

Public Street Lighting (PJU) plays an important role in transportation infrastructure, especially at night. Currently, complaints about PJU damage are only made using social media. This research designs and builds a web-based PJU complaint system with real-time notifications via WhatsApp in Pemalang Regency. Data was collected through interviews, observations, and questionnaires. This system is built with PHP and MySQL, with WhatsApp notification integration to ensure accurate and real-time complaint information. The system trial involved the community, showing the system's effectiveness in increasing reporting efficiency and officer response. The system provides easy online reporting and real-time notifications via the website and WhatsApp. This system is expected to improve community services and PJU management. The results can be a reference for the development of similar systems in other areas.