JOINCS (Journal of Informatics, Network, and Computer Science)
Not a member yet
80 research outputs found
Sort by
Facial Human Emotion Recognition by Using YOLO Faces Detection Algorithm: Pengenalan Emosi Wajah Manusia dengan Menggunakan Algoritma Deteksi Wajah YOLO
No full textDeep emotions have gained importance recently because they constitute a form of interpersonal nonverbal communication that has been demonstrated and used in a variety of real-world contexts, including human-machine interactions, safety, and health. The best elements of a human face must be extracted in order to forecast the proper emotion expression, making this method extremely difficult. In this work, we provide a brand-new structural model to forecast human emotion on the face. The human face is found using the YOLO faces detection technique, and its attributes are extracted. These features then help to classify the face image into one of the seven emotions: natural, happy, sad, angry, surprised, fear, or disgust. The experiment demonstrated the robustness and speed of the suggested structure. This paper made use of the FER2013 dataset. The experimental findings demonstrated that the proposed system's accuracy was 94%
Sarcasm Detection in News Headline Dataset with Ensemble Deep Learning Method: Deteksi Sarkasme Pada Dataset News Headline Dengan Metode Ensemble Deep Learning
Full text linkSarcasm, a prevalent linguistic device, is frequently used in public discourse, often causing offence and distress to the listener. The complexity inherent in detecting sarcasm is a significant and ongoing challenge in the field of sentiment analysis research. The widespread use of this phenomenon in diverse conversational contexts further complicates its identification in data sets full of human interactions. Deficiencies in methodologies for distinguishing such statements adversely affect the performance of sentiment analysis, especially in distinguishing negative, positive or neutral sentiments. Inaccuracies in sarcasm detection can affect the classification results of sentiment analysis. Therefore, sentiment analysis seeks to categorise sarcastic sentences that, despite appearing positive, actually contain negative meanings. This research aims to build a deep learning ensemble stack model. The basic deep learning methods used are Bidirectional Gated Recurrent Unit (BiGRU) and Convolutional Neural Network (CNN). LightGBM is used to perform stack ensemble of deep learning methods. The dataset used comes from the Kaggle website and consists of English headlines. The findings show that the stack ensemble method outperforms BiGRU and CNN, evidenced by an accuracy rate of 91.2% and an F1 score of 90.2%. Therefore, from the above discussion, it can be concluded that the LightGBM method emerges as the optimal solution for sarcasm detectio
Information System for Processing Police Report Data at the North Gorontalo Police Women and Children Protection Unit: Sistem Informasi Pengolahan Data Laporan Polisi Pada Unit Perlindungan Perempuan Dan Anak Polres Gorontalo Utara
Full text linkDesigning an Information System for Processing Police Report Data at the Women and Children Protection Unit of the North Gorontalo Police, so that it can assist officers in collecting large numbers of reports, and can present reports quickly. The method used is the Research and Development (R&D) method. The system is made with PHP and HTML programming. The database used is the MySql language. For modeling using UML (Unified Modeling Language). For testing this system using white box and black box testing where with this test the correct graph flow is obtained as a sample the researcher tests the Input Data flowchart Report on the admin system obtained is the value of region (R) = 3, Independent path = 3, and Cyclomatic Complexity (CC) = 3
Design Of An Expert System Using Bayes' Theorem Method For Web-Based Diagnosis Of Diseases In Caftage Animals
No full textKemajuan yang cepat pada teknologi informasi menjadikan teknologi sebagai kekuatan dalam berbagai bidang di era modern. Teknologi informasi khususnya pada bidang kecerdasan buatan telah membuat perangkat lunak sistem pakar. Sistem pakar adalah salah satu cabang ilmu dari kecerdasan buatan yang mengadopsi pengetahuan, fakta dan teknik penalaran pakar yang digunakan untuk memecahkan permasalahan yang biasanya hanya dapat dipecahkan oleh pakar dalam bidang tersebut, seperti pada peternak sapi di Balai Karantina Pertanian Kelas 1 Kupang, Wilayah Kerja Waingapu, terkadang sulit menemukan tenaga medis seperti dokter hewan ketika menemukan ternak sapi yang sakit. Salah satu bagian yang paling penting dalam penanganan kesehatan ternak adalah melakukan pengamatan terhadap ternak diduga sakit merupakan suatu proses untuk menentukan dan mengamati perubahan yang terjadi pada ternak melalui gejala-gejala yang diderita oleh sapi. Peternak sapi sering kali mengalami kendala dalam mengetahui penyakit sapi karena terbatasnya pengetahuan. Pada kondisi tersebut dibutuhkan peran seorang pakar dalam mengatasai penyakit pada sapi dengan menerapkan teorema bayes yang digunakan dalam statistika untuk menghitung peluang untuk suatu hipotesis. Metode yang digunakan yaitu metode waterfall dengan tahap analisis, desain, implementasi, dan pengujian. Metode pengumpulan data dilakukan dengan cara wawancara, observasi dan studi literatur
Testing The Accuracy of Fingerprint Recognition using Levenshtein Distance and Hamming Distance Methods : Uji Ketepatan Pengenalan Sidik Jari dengan Metode Levenshtein Distance dan Hamming Distance
Full text linkThe presence or evidence of attendance is crucial in monitoring the presence of every individual working in a particular field. Developing an employee attendance system using fingerprints can expedite the processing of data of employees who have or have not attended. One brand of machine used as a fingerprint attendance tool is Fingerspot Flexcode. The data obtained from the machine comes in the form of bitmap images that are converted into strings using encoding. Although the resulting string sequences are different, there is a possibility of similarity in fingerprint data among employees because the system cannot distinguish data precisely. Therefore, the comparison between the Levenshtein Distance and Hamming Distance methods is used to determine which method has the highest accuracy in processing the system's calculation. The method with the highest accuracy will determine the level of compatibility of the method with the tested tool. For example, 6 fingerprint data are taken from each of the 7 different employees, resulting in a total of 42 data as test data. The calculation results show that the accuracy of the Levenshtein Distance method is 80,76 % with a precision of 46,43 %, while the Hamming Distance method is 78,34 % with a precision of 30,50 % in processing string similarity in fingerprint data. Based on these results, it can be concluded that the Levenshtein Distance method is better in calculating similarity in fingerprint data compared to the Hamming Distance method because it has a higher level of accuracy and precision compared to the Hamming Distance method
Pocong Rush: Endless Runner Game Based On Finite State Machine: Pocong Rush: Endless Runner Game Berbasis Finite State Machine
Full text linkEndless runner Game is a Game where the player will move continuously indefinitely to get the highest point in the Game. Endless runner Game is synonymous with challenges in the form of obstacles and collective items. This study focuses on setting challenges in the non-playable character (NPC) behavior so that they can provide more dynamic challenges. Obstacle, player and NPC settings use the Finite state Machine method which is implemented in an html 5 based Game engine Construct. The implementation results are in the form of an html 5 Game which is then tested through 2 testing stages, namely blackox testing and beta testing. The test results using blackbox testing show that all functionality in the Game can run well, including the behavior of NPCs based on the designed FSM. Beta testing shows that the majority of respondents who have played endless runner Games stated that they like horror Games with local content nuances and the impression of humor and that the average Game difficulty level based on NPC responses to players is 63.75%
Development of Accounting Media to Help Financial Literacy for SMEs in Surabaya : Rancang Bangun Media Akuntansi untuk Membantu Literasi Keuangan bagi UKM di Surabaya
Full text linkImprovements in information technology and trade aspects during the COVID-19 pandemic provided a tremendous increase in financial inclusion, but unfortunately this was not matched by financial literacy, so that many MSMEs were disadvantaged by this. This activity aims to assist MSMEs in developing accounting media for MSMEs, so that they obtain real conditions related to more accurate bookkeeping. Data was collected by means of interviews and discussions. The results of discussions and observations with partners stated that a simple bookkeeping media was needed to help partners. Then from the results of this activity, they contribute in the form of simple accounting reporting media that help them manage finances better and provide reports that convey financial conditions quickly
Village Complaints Application System Based On Android Webview: Sistem Aplikasi Pengaduan Desa Berbasis Android Webview
Full text linkMany residents in the village of Entalsewu, Buduran District, Sidoarjo Regency, often complain about the situation in the village. Complainants do not know where to go to convey their complaints in the village environment. For example, roads that are damaged and have potholes due to frequent vehicles and extreme rainy weather along the roads in Entalsewu village and also the dead and dim public street lighting (PJU) have not been repaired so that road users experience accidents at night, many residents' houses flooded water that enters the house. Therefore, to overcome this problem, the author designed a village complaint application system based on android webview with a case study of village development. Which aims to make it easier for the public to make complaints in the village of Entalsewu, Buduran District, Sidoarjo Regency. This android application can be used by the public to make complaints online. This application can also make it easier for the village government to respond to community complaints. In this study, the author uses the Waterfall method as a process or flow of research results
Decision Support System for Determining the Best Santri Using the SAW Method: Sistem Pendukung Keputusan Penentuan Santri Terbaik Dengan Metode SAW
Full text linkThe I'dadiyah Az-zainiyah institution is an institution devoted to new students of the Nurul Jadid Islamic Boarding School Paiton Probolinggo at the junior and senior high school levels. This institution is devoted to new students who do not have a foundation in religion. To give appreciation to the students of the I'dadiyah Az-zainiyah Institute so that Santi is more enthusiastic to study religious knowledge every semester, the best students will be selected to be awarded in the form of scholarships. However, from the large number of participants, the management of the I'dadiyah Az-zainiyah Institution has difficulty determining the best participants because there are several variables that must be compared between one santri and another, so there is often a delay in the announcement of the best participants which results in delays in prospective students who get scholarships. from several schools under the auspices of the Nurul Jadid Islamic Boarding School. Seeing this, it is necessary to have a method that can make it easier for the administrators of the I'dadiyah Az-zainiyah Islamic Boarding School Nurul Jadid to determine the best new students or new students. The method taken by the researcher is the Simple Additive Weighting (SAW) decision support system method
Design and Build a Web-Based Guestbook Case Study Gresik Regency Education Office: Rancang Bangun Buku Tamu Berbasis Web Studi Kasus Dinas Pendidikan Kabupaten Gresik
No full textThis research was conducted at the Gresik Regency Education Office, namely how to design a Web-Based Guestbook and Guest Complaints Information System at the Gresik Regency Education Office. The management of guest data and guest complaints is still very conventional, so delays often occur in visiting services and in conducting guest complaint consultations. The purpose of this research is to make it easier for employees to process guest data and consult guest complaints and to make it easier for guests to register guest data and guest complaints wherever they are. This study uses the Waterfall method. The stages of the waterfall are analysis, system, coding design and testing. Methods of data collection through interviews and observation, literature study. The results of this study are to make it easier to register guests and submit/consult guest complaints and manage guest data, guest complaints