Search

The new life adaptation protocol implemented in Institut Teknologi Sumaterea (ITERA) during the COVID-19 pandemic requires visitor identity check to restrict access to several buildings and reduce the massive spread of the corona virus. The inspection is carried out by manually collecting visitor information, which can result in a queues buildup of vehicles that want to enter the ITERA environment, and also, the identity of incoming visitors is not adequately monitored. In addition, the smart camera system for visit recording is built with the ability to identify visitors' faces, which triggers automatic gate access and records visitor information. Visitor information is stored in the form of identity number, name, visitor status, time/date of visiting, and information about the location of visitors. The visitor data is stored in a database and can be accessed by ITERA on the visitor website page. The system has an ability to trigger automatic gate to give an order to open or close the gate based on the face recognition results. The system development is divided into three stages: face recognition development, website-based data visualization creation, and hardware development. At each stage, the development produces a system with face recognition capabilities, a visitor recorder website, and a series of automatic gate hardware consecutively. The face recognition system connects directly to the website via the Firebase database. The development of the system capability for face recognition is carried out using a learning model of the K-Nearest Neighbor (K-NN) algorithm method, which obtains the training results have an accuracy of 99% using a confusion matrix method. [ABSTRACT FROM AUTHOR]

4C Skills refer to four essential skills needed for education and development in the 21st century. These skills include critical thinking, communication, collaboration, and creativity. 4C skills can be developed through learning by applying learning models that can train these skills. The problem-based learning model is a practical approach to developing 4C skills. This study aims to review indexed articles explaining how the PBL model develops 4C skills in science learning. The data for this study were researched using the Scopus database and focused on problem-based learning models and 4C skills (critical thinking, communication, collaboration, and creativity) in science education. This search was an exceptional literature review using the four-step PRISMA approach to identification, screening, eligibility, and inclusion. Sixty-three relevant articles from 22 Scopus-indexed journals over the past six years (2018-2023) were analyzed using bibliometrics and Excel spreadsheets. Based on the results, the PBL model and 4C skills research were mainly carried out at the upper secondary level, and Indonesia is the country with the most publications. Physics is a widely studied field that uses a quantitative method. Based on the results of the review of articles, it is clear that the PBL model can be used to develop 4C skills, especially critical thinking. [ABSTRACT FROM AUTHOR]

In recent times, the interaction between human beings and computers has become an increasing field of research due to its wide application of computational systems. Determination of the current state of human psychological problems, one of the most relevant techniques is emotion recognition. The dynamic behavior of human is crucial for its interpretation. So, Facial emotion recognition (FER) is an important research area for detection the of different facial emotions. In this paper, we proposed an Unconstrained Minimum Average Correlation Energy (UMACE) filter as a classifier for facial emotion recognition. We are applying this proposed method in the CK+ database and it can give a better recognition rate. [ABSTRACT FROM AUTHOR]

Aim-The aim of the study is to perform innovative detection of lung cancer using the Decision Tree classifier and also comparing it with the K-Nearest Neighbor classifier. Materials and methods-The lung cancer database includes records 32 samples (patients) and is therefore compiled with 32 ∗ 6 numerical data. The G power for samples is calculated from clincalc which contains two different groups where group 1 is taken as (n1=16) and for group 2 (n2=16), alpha (0.05), power (80 %) and enrollment ratio. Results-Implementation based on the MATLAB toolbox is done on CT lung images and the classification of these images is performed. In the proposed method, D-Tree numbers of samples are selected for each iteration and 90 % phase accuracy is obtained. There is a significant difference in Accuracy rate (P=0.072). Conclusion-The results of the study suggest that the accuracy of the decision tree algorithm is significantly higher than the nearest k-neighboring algorithm. [ABSTRACT FROM AUTHOR]

In this study, we take a different approach to feature extraction by focusing on the diagonal rather than the horizontal in order to statistically and structurally utilise handwritten characters. Substances and Techniques: This paper employs the Computer Vision Laboratory's (CVL) publicly available database for its dataset (NTHU). Using a G intensity of 0.8 and alpha and beta quality values of 0.05., 0.2, and a 95% confidence interval, we determined that a sample size of 280 (group 1=140, group 2=140) handwritten characters would provide adequate statistical power. Novel sample number (N=10) diagonal feature extraction and sample number (N=10) horizontal feature extraction are used to classify statistical and structural aspects of handwritten characters, resulting in greater precision. The outcomes show that the accuracy of the diagonal feature extraction classifier is 95.883, while that of the horizontal feature extraction classifier is 92.736. This study has a significance level of p=0.0213. In conclusion, statistical and structural features of handwritten characters make diagonal feature extraction more accurate than horizontal feature extraction. [ABSTRACT FROM AUTHOR]

The K-Nearest Neighbor approach outperforms the Fuzzy Algorithm in terms of accuracy during quick data retrieval for an innovative car parking system. The Components and Techniques: K-Nearest Neighbor Algorithm enables 20-sample database vehicle parking system, which improves innovative vehicle parking system's data retrieval accuracy and speed. The implementation was accomplished by developing a Python web application with the aid of the anaconda navigator. As a result, a car parking database is constructed using the K-Nearest Neighbor Algorithm, which improves efficiency by 87 percent in terms of secure accessibility compared to the FuzzyApproach, which improves efficiency by only 75 percent. The length of time it takes to log in and access the database has been analysed using SPSS (Statistical Package for the Social Sciences) with the independent variables of time and size and a significance level of p 0.05. K-Nearest Neighbor Algorithm, which is more significant than a Fuzzy Approach in terms of Accurate efficiency, has been employed in a protected system to improve the car parking database. When data is a dependent variable and sample size is an independent variable, SPSS analysis may show how reliable the data is. [ABSTRACT FROM AUTHOR]

The goal of the research is detection of disease in fresh fruits. The sensitivity and accuracy was calculated and compared with the k-nearest neighbor (KNN) classifiers. Materials and Methods: The research's data set was obtained from the Kaggle open access dataset. The research includes two groups, in which CNN is group 1 and KNN is group 2. Kaggle database is used for collecting different fruit images. The sample size was calculated using G power (80%) and alpha (0.05) and it is found to be 20 samples for each group. Result : CNN has achieved 98 % of accuracy and 1.0477 standard deviation while KNN method has achieved 90 % of accuracy and 0.65071 standard deviation. The Novel CNN has significant accuracy of p<0.05 (2 tailed). Conclusion: It is concluded that the detection of fruit disease using Convolution Neural Network appears to be significantly better compared to k-Nearest Neighbour Classifiers. [ABSTRACT FROM AUTHOR]

The primary purpose of this research is to use the Support Vector Machine (SVM) classifier and the Naive Bayes (NB) classifier to make arrhythmia predictions from the MIT-BIH database. With an alpha of 0.05, 95% confidence interval (CI), 80% power, and an enrollment ratio of 1, the proposed research employed SVM and NB machine learning algorithms to predict arrhythmia using the MIT-BIH dataset of 65 normal and 65 abnormal ECG signals downloaded from IEEE dataport in.xlsx format. We used the data mining programme WEKA 3.8.5 to determine how to categorise people with and without arrhythmia. IBM SPSS version 21 was used for the statistical analysis. When comparing SVM and NB classifiers, a statistically significant difference (p0.010) was found. Using WEKA's 10-fold cross-validation for training and testing, the SVM classifier outperformed the NB classifier with an 88.50% accuracy rate in classification (80.39 percent). [ABSTRACT FROM AUTHOR]

The novel adaboost method for accurate eye recognition in face photographs is the focus of this research project, which aims to implement said approach. The purpose of this investigation is to evaluate how well the boosting algorithm performs in comparison to the newly developed Adaboost method. Images of human faces were extracted from the ORL Face Database and analysed for characteristics that may be used for eye recognition. In this study, we employ a total of forty people: twenty from each of two groups. In the simulations, the pretest power was established at 0.8. In order to evaluate how effectively the innovative Adaboost algorithm works, metrics such as accuracy and sensitivity are produced and analysed. The Adaboost Algorithm has an accuracy of 98.73 percent, while the Boosting Algorithm has an accuracy of 86.41 percent, and the Sensitivity of the Adaboost Algorithm is 96.15 percent, while the Sensitivity of the Boosting Algorithm is only 83.63 percent. These results come from simulations run with SPSS. In the case of this model, we discover that the level of significance is 0.000. (2-tailed, p0.05). In this study, it is shown that the unique Adaboost method is superior to the boosting algorithm in terms of accuracy and sensitivity. [ABSTRACT FROM AUTHOR]

The primary goal of this research is to use J48 and K-Nearest Neighbor (KNN) classifiers to predict arrhythmia using the MIT-BIH database. With an alpha of 0.05, 95% confidence interval (CI), 80% power, and an enrollment ratio of 1, the proposed study used the J48 and KNN machine learning algorithms to predict arrhythmia using data from the MIT-BIH dataset consisting of healthy (n=65) and arrhythmia (n=65) ECG signals collected from IEEE dataport in.XLSX format. WEKA 3.8.5, a data mining tool, was used to distinguish between those with arrhythmia and those without. IBM SPSS version 21 was used for the statistical analysis. There was no discernible difference (p=0.025) between the J48 and KNN classifiers. Using WEKA's 10-fold cross validation to train, test, and verify the classifiers, we find that the J48 classifier is more accurate at classifying data (89.80 percent) than the KNN classifier (87.64 percent). [ABSTRACT FROM AUTHOR]

This research endeavours to evaluate the efficacy of the support vector machine (SVM) method against that of the K-Nearest neighbour Network (KNN) approach in removing thick fog from images. We gathered the Realistic Single Image Dehazing (RESIDE) dataset from the kaggle.com database. For each group's sample size of 22, the suggested SVM method is compared against the current KNN techniques. There are 44 samples altogether. Our suggested process include measures to actually accomplish the elimination of noise in photos. Metrics of performance, such precision and sensitivity, are computed. The suggested support vector machine (svm) has an accuracy and sensitivity of 94.9 and 38.5%, respectively, whereas k-nearest neighbour had 93.7 and 37.5%. Using the G power instrument and setting the minimum power to 0.8, we can determine how many samples are needed for this study. When comparing the SVM and KNN algorithms, the significance level reached was (p 0.05). Support vector machines provide far superior performance over the standard k-nearest neighbour, as shown both by experimental findings and Independent sample T-test statistical analysis using SPSS software. [ABSTRACT FROM AUTHOR]

Aim: The main aim of this study is to assist the visually impaired to dole out money from the innovative ratified ATM compared with smart ATM. Materials and methods: The Raspberry Pi which is a microcontroller acts to convert analog input signals to the digital output signal. A Biometric system with IVRS at an ATM promotes the transaction to succeed. The significance between the two groups was calculated using SPSS. The parameters were maintained as threshold 0.05, enrollment ratio as 1, confidence level 95%, and G-power 80%. Results: Ratified ATM was found to perform better than smart ATM in aiding the visually impaired. There exists a statistically significant difference (p=0.878, p>0.05) between ratified and smart ATM groups. The execution of this method has been found promising on the account of its sensitivity to the recognition of the customer's fingerprint as contained in the database. Conclusion: This method not only helps the visually impaired but also illiterates, dyslexia, and people with visual disabilities to withdraw money safely. [ABSTRACT FROM AUTHOR]

In this investigation, we compare the performance of the state-of-the-art ANN algorithm for vehicle detection with the more conventional SVM algorithm. Twenty representatives were selected at random from the database of the website. The threshold was kept at 0.05, the confidence level at 95%, the enrollment ratio at 1, and the G power at 80%. Accuracy metrics are calculated using the SPSS software. Results from an independent t test showed that the ANN algorithm outperformed the SVM algorithm, with a mean accuracy of 95%. The difference between the two groups was not statistically significant (p=0.11, p>0.05). In this research, it appears that the ANN algorithms perform better than the SVM algorithms when it comes to detecting vehicles in the datasets used. [ABSTRACT FROM AUTHOR]

The aim of this research is to use the Decision Tree classifier to identify lung cancer in scanned pictures and compare its performance to that of the Support Vector Machine classifier. Methods and materials The lung cancer database contains information on 60 samples (patients). Clincalc's sample G power has two groups, with alpha (0.05), power (80%), and enrollment ratio (n1 = 30) for group 1 and (n2 = 30) for group 2. Decision Tree outperforms standard SVM classifiers in terms of accuracy, with a 95.13 percent success rate. According to the results of this study, the decision tree method is more accurate than the Support Vector Machine classifier. [ABSTRACT FROM AUTHOR]

It is known that phrases are considered a common linguistic unit consisting of several words and a syntactic whole that cannot be translated literally. Idioms have linguistic and cultural significance and were formed as a result of symbolic similes and comparisons due to various situations in the life of each nation during its history. The linguo-cultural approach to the text in translation allows identifying and overcoming the obstacles caused by the culturally defined idio-ethnic and cultural differences in artistic discourse. Therefore, to achieve efficiency in automatic translation, it is important to create a parallel database of phraseological units (PhU). This article analyzes the factors involved in identifying idioms as lexical units and the creation of an idioms database. [ABSTRACT FROM AUTHOR]

This article is devoted to the development of intelligent models of groundwater modification based on modern GIS technology as well as mathematical models. The article is based on the geographic information system or GIS (GIS), which collects, stores, processes, analyzes and reflects spatial information, allowing you to obtain the geographic area and the information and knowledge associated with them. on the study of non-technological spatial information. It is based on the creation of a database for storing, analyzing and collecting data and data using a variety of specialized GIS applications. Geography and digital cartography show the use of specialized GIS programs. Arcin's ArcInfo and its additional modules describe the use of applications such as ArcMap, ArcCatolog, and ArcToolbox, the execution of functional operations by ArcMap and many other modules, large-scale image processing, and geographic linkage methods. [ABSTRACT FROM AUTHOR]

In this article, an algorithm and program were developed for image restoration and recognition using neural networks. Several methods of image recognition using neural networks were proposed to solve these problems. The use of artificial intelligence methods for image restoration and recognition, as well as the scientific novelty of research results, is related to the use of artificial intelligence methods and the advancement of modern practical software tools created for them, which are based on the characteristics of digital images. To recognize an image as an object, the human face was taken as an object, meaning the task was to separate the facial area from the image and determine whether or not the person's information is present in the database, and to develop a program that can recognize the person. [ABSTRACT FROM AUTHOR]

This article discusses the process of searching for information in the bibliographic database. Here developed an algorithm for obtaining results according to certain characteristics of the information related to the sought record. It is also mentioned about the functional blocks of preparation, creation of pair records, comparison of fields with pair Records, decision-making, which constitute the algorithm. [ABSTRACT FROM AUTHOR]

The examined pollution sources in the current paper include combustion and industrial processes; agriculture; household heating; waste and waste water treatment; road, rail and aviation transport. The investigated group of contaminants released from each of the listed pollution sources for the period 2011-2020 in Bulgaria are the following: sulphur and nitrogen oxides, non-methane volatile organic compounds, methane, carbon and dinitrogen oxide, carbon dioxide and ammonia. A hierarchical organization of the objects from the mentioned sets is built. This studied information from time series stored in xlsx files is structured in a relational database. In this connection, the values of contaminants from a certain pollution source in the relevant file are presented in matrix form. The variation of the number of columns in the formed rectangular matrix depends on the number of the examined years. The data are regrouped to obtain the transposed matrix. The components from each column of this matrix are entered into a selected field of a certain table from the database. The presented approach integrates the processes of searching the indicated data, restructuring and grouping separate datasets, and entering them in the built database. Certain criteria are defined in order to form and extract groups of elements from one or more tables of the database, as well as to establish the relevant dependencies for the studied objects. Analysis of variance and Tukey's range test are used for statistical assessment of the data. The approach can also be applied to examine datasets on air pollutant emissions for other countries. [ABSTRACT FROM AUTHOR]

Knowing the history of research conducted by the researchers in the past will help in shaping current and future exploration in the field. This study aims to find out which research topics that were popular in 2009-2019 in well-respected international journals. Employing a descriptive method, this research collected the data from 3 major ranking journals about science/physics education, namely the International Journal of Science Education (IJSE), Research in Science Education (RISE), and Science Education (SE). The data collection was carried out from April to September 2020 by extracting journals from the Clarivate Analytics WoS online database system. The most popular topicsphysics researchers in IJSE, RISE and SE journals in the period of 2009-2019 are science learning: contexts, characteristics and interactions (13.50%), science learning: development of student understanding (9.35%), and curriculum and assessment (8.47%), while thetopics of religious beliefs, methodological issues, textbook analysis (2.76%) and pre-service science teacher education (3.89%) gained the least attention. In addition, USA was the country that contributed the most articles in IJSE, RISE, and SE journals is USA with the percentage of 35.30% or about five times higher than the UK and Australia which ranked second and third with the percentage of 7.21% and 6.46%, respectively. [ABSTRACT FROM AUTHOR]

The development taking place in our world today includes all aspects of life, especially multimedia, and made dealing with images and information daily and easy to use. The reliance on multimedia information has also increased, as it has become easier for each user to obtain images and information. Images can be manipulated through editing programs, and this can be done with ease and professionalism. The objective of this research is to verify digital images and detect forgery, Forgery images have many types of methods like splicing, copy move and retouching images. In this research, we suggest an effective way to detect image splicing forgery by using the error level analysis algorithm (ELA) and relying on deep learning using the convolution neural network. The proposed method was applied to two databases: (CASIAV0.1 & CASIAV 0.2) The proposed method outperformed similar studies in this area where it Applied with CASIA V0.1 database and achieved accuracy of 99% also applied with CASIA V0.2 database and achieved accuracy of up to 95%. [ABSTRACT FROM AUTHOR]

Adaptive control is used when the automatic control system has to modify its own characteristics depending on the operating conditions. Such control is embedded in many contemporary systems in industry. It is based on the integration of knowledge about the process, presented as a database, with real-time monitoring depending on the incoming information. This paper proposes a method for implementation of adaptive control on asynchronous drive of electric vehicles. The necessary components are specified – the proposed method uses PLC, sensors providing the required information and a database with pre-recorded characteristics. The requirements to the database are determined – it must include the calibration characteristics of the electrical drive. Verification of the method was carried out by a laboratory simulator. Calibration of the electrical drive is performed and the required characteristics are obtained. Power losses and some important factors (coefficients) are calculated, many graphs are plotted. Implementation of this method will lead to increased energy efficiency and comfort of passengers in vehicles and allows quick and accurate fault detection. [ABSTRACT FROM AUTHOR]

Data on the poverty rate in Indonesia is very important because it can be used as a parameter of the prosperity of the Indonesian people and can also be an evaluation for the government. BPS is in charge of presenting real data through survey and census results and the data presented by the Central Statistics Agency is often delayed. To overcome the delay, predictions are made using the simple moving average method to extract and identify potential and useful knowledge information stored in large databases. [ABSTRACT FROM AUTHOR]

Currently, there are still very many Kindergartens whose security in checking the identity of the introduction or pick-up is still manual, some even do not do identity checks. This is felt to be less able to overcome parental concerns about child crime which is increasingly happening. We need a computer-based tool to increase child safety when in delivery and pick-up by knowing who the delivery person and pick-up are. If the child is delivered or picked up, then the delivery person or pick-up only needs to scan the Student Card by waiting for the successful display on the LCD and also the web display so that the delivery or pick-up data is stored in the database, then the child will be called automatically at the time of pick-up and the parents receive an SMS notification when the child is picked up. Based on the results of the overall design of a student security system using RFID that has been carried out, the results of the design that has been tested overall have been running and active as expected by the author with a success percentage of 94%. [ABSTRACT FROM AUTHOR]

Metabolism and therapeutic effect of nucleobase and nucleoside analog (NNA), in particular acyclovir (ACV) and ganciclovir (GCV) as antivirus and anticancer, were affected by NUDT15 enzyme. NUDT15 is able to hydrolyze thiophosphate metabolites of ACV and GCV, thus the patients with NUDT15 deficiency have better infection control. This study aimed to identify the best interaction between alternative potential ligands from the Herbal Indonesia Database with NUDT15 through basic pharmacophore virtual screening to aid patients with high NUDT15, to increase the treatment's effectiveness. Software used in the study were Marvin Sketch, LigandScout 4.4.5, AutoDock ver 4.2.6, LigPlus, and Notepad++. The databases were gathered from Protein Data Bank, PubChem, ChemSpider, SwissADME, admetstar, and Herbal Indonesia Database. Based on the bioinformatics analysis using software and database, twenty ligands were analyzed from the Herbal Indonesia Database. The top 3 ligands found were Torvanol A, DIMBOA glucoside, and 2-Phenylethylglucosinolate, which had the best ligand-protein docking results of analyzed parameters (lowest binding energy, lowest inhibition constant, following the Lipinski's Rules criteria, and safe toxicity profile). Torvanol A showed the best analysis results of parameters among analyzed ligands. The basic pharmacophore virtual screening helps us to have an understanding of protein and ligand interaction so that we are able to find the compound candidate which expected to have good interaction based on bioinformatic analysis. Torvanol A, DIMBOA glucoside, and 2-Phenylethylglucosinolate were potential ligands with good docking profile to NUDT15, thus expected to be candidates for further testing (in vitro and in vivo) to improve treatment of patients with high NUDT15. [ABSTRACT FROM AUTHOR]

The electronics laboratory is a place for practicum for electronics students. In this laboratory there are electronic components that must be protected from temperature and humidity. If the temperature is too cold and the humidity is too high, it can cause corrosion of electronic components, which can cause damage. Arduino is an open source electronic kit specially designed to make it easier for everyone to develop electronic devices that can interact with various sensors and controllers. Laboratory room temperature detection system using software and hardware. The software used is Arduino and the MySql database, while the hardware is power supply, D1 Mini microcontroller, DHT11 sensor and LCD. The temperature data obtained from the DHT11 sensor will be displayed on the LCD. In data processing using fuzzy method. [ABSTRACT FROM AUTHOR]

The objective of this research is to provide a literature review to discover and understand the development of astronomical and religious sciences in determining the new moon. The research method is descriptive with a qualitative approach. This research uses articles from the Google Scholar database, with the help of the Publish or Perish software, limited from 2012-2022 using the keyword "Astronomy Hilal" and found 105 relevant articles. After filtering, 25 articles were obtained based on inclusion and exclusion criteria that could be sampled for analysis. After conducting the research, it was found that the development of lunar determination in astronomy and religious sciences over the past 10 years can be grouped into six discussions, namely: 1) The concept of Matlak, 2) Definition of Hilal, 3) the issue of agreement on the criteria used 4) adoption of technology 5) feasibility of the Moon Observation Center (POB) and 6) the role of astronomy clubs. [ABSTRACT FROM AUTHOR]

Writing a scientific paper is one of the essential skills for a Researcher. We investigate the research trend of Writing Scientific Manuscript Research from the Scopus Database. The study was conducted on October 1, 2022, using bibliometric analysis. Based on the data collected from Scopus Database, we found 1,197 documents using "Writing Scientific Manuscript" keywords from 1960 to 2022. Data were analyzed using VOSviewer software to show the keywords map and correlations between the keywords. Generally, based on the data, there are 5 cluster keywords, namely: "scientific manuscript" (red cluster), "editing" (blue cluster), "report" (purple cluster), "evidence" (yellow cluster), and "history" (green cluster). The scientific manuscript correlates directly with guideline, preparation, submission, reviewer, revision, skill, art, and responsibility. Another interesting finding is the existence of "evidence" cluster. The research must base on the evidence or phenomena. Furthermore, the Research paper must reflect the history of the research development to find the state the art. In conclusion, the research about Writing Scientific Manuscript needs to be explored deeply since this research is still in good trend until the recent year. [ABSTRACT FROM AUTHOR]

Various limitation on the direct use of essential oil has stimulate encapsulation technology studies in the recent year. The study aimed to analyze the scope of essential oil encapsulation through a bibliometric analysis. This paper used Scopus database to collecting data about the subjects. VOSviewer was used for network visualization and map constructing. Started from 1988, a total of 1336 publications were collected for analysis. Over the last ten years, the number of articles published has increased significantly. Iran, China, and Brazil are most productive countries, while Indonesia published 26 articles in this topics. Based on VOSviewer visualization, the most influential topics are related to essential oil encapsulation method. However, overlay map showed new emerged author keywords related to the application of essential oil encapsulation for food packaging and microemulsion novel technique by pickering emulsion. [ABSTRACT FROM AUTHOR]

Learning based on the Science, Technology, Engineering, And Mathematics (STEM) approach is increasingly popular with technological developments in this era. This study aims to see the influence of the Science, Technology, Engineering, and Mathematics (STEM) Approach in physics learning, as well as the feasibility of using STEM at the educational level. The method used in this study is quantitative descriptive with meta-analysis techniques. The population in this study was taken from the Web of Science (WOS) data base, DOAJ Journal, and SCOPUS journal. Until 196 articles were obtained related to the STEM approach in physics learning, by combining keywords from the SCOPUS database of 123 articles, the DOAJ database of 68 articles, and in the WoS database (Web of Science) as many as 5 research articles. The research was conducted by utilizing JASP as a research tool. Overall results are obtained with an estimated effect size of 0.891 (large effect category). This shows that the STEM approach can have a positive influence on physics learning as a whole. [ABSTRACT FROM AUTHOR]