cover
Contact Name
-
Contact Email
-
Phone
-
Journal Mail Official
-
Editorial Address
-
Location
Kota semarang,
Jawa tengah
INDONESIA
JAIS (Journal of Applied Intelligent System)
ISSN : 25020493     EISSN : 25029401     DOI : -
Core Subject :
Journal of Applied Intelligent System (JAIS) is published by LPPM Universitas Dian Nuswantoro Semarang in collaboration with CORIS and IndoCEISS, that focuses on research in Intelligent System. Topics of interest include, but are not limited to: Biometric, image processing, computer vision, knowledge discovery in database, information retrieval, computational intelligence, fuzzy logic, signal processing, speech recognition, speech synthesis, natural language processing, data mining, adaptive game AI.
Arjuna Subject : -
Articles 10 Documents
Search results for , issue "Vol 7, No 3 (2022): Journal of Applied Intelligent System" : 10 Documents clear
IMAGE CLASSIFICATION OF LOCAL ROBUSTA AND ARABICA COFFEE SEEDS IN MALANG REGENCY USING GRAY LEVEL CO-OCCURRENCE MATRIX AND K-NEAREST METHODS Devita Widiawati; Muhammad Rijalun Shodaqu; Gilang Priambodo; Maulana Fajar Anas; Titien Suhartini Sukamto; Aris Nurhindarto
Journal of Applied Intelligent System Vol 7, No 3 (2022): Journal of Applied Intelligent System
Publisher : Universitas Dian Nuswantoro and IndoCEISS

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33633/jais.v7i3.7214

Abstract

Coffee is one type results current plantation  this favored by some among. Indonesia is in the order to four Becomes Robusta coffee export and producer in the world. Appearance communities coffee lovers make coffee as provider field profession for part big resident. In Indonesia, especially in the Regency of Trunk, a lot very Public around who has coffee plantations including namely Robusta coffee and Arabica coffee (coffea arabica) local. For some new people Do you know and love coffee yet? can differentiate type of coffee visually. In the era of increasingly digitalization, advanced like this. There are several method for differentiate something object among them that is processing digital image. Frequent problems occur that is many less consumers in determine Robusta and Arabica coffee types. From trouble that, then researcher designing a system classification on robusta and Arabica coffee beans could obtained with implementation algorithm K-Nearest Lightweight Classification (K-NN). [1] combined with extraction feature Gray Level Co-Occurrence Matrix (GLCM). Digital image dataset used that is a total of 194 pictures where inside it there is type image coffee beans. Image dataset Robusta and Arabica coffee beans each local number of 97 images. Image dataset shared into 20 test data and 174 training data. Testing conducted using Matlab software produce score accuracy highest at distance pixels=1 and the value of K=1 with respect to angle of 45° by 95%.
Expert System of Facial Skin Type Diagnosis and Skincare Recommendation Based on Certainty Factor Dadan Saepul Ramdan; Castaka Agus Sugianto; Rizqy Dimas Monica
Journal of Applied Intelligent System Vol 7, No 3 (2022): Journal of Applied Intelligent System
Publisher : Universitas Dian Nuswantoro and IndoCEISS

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33633/jais.v7i3.7150

Abstract

Facial treatment is an important need for everyone because the first sight of meeting someone is to see their face. Generally, facial skin type is just normal skin. However, several factors such as the environment, air, food, facial hygiene, and so on can affect the type of human facial skin. In this experiment, there were 5 types of facial skin, namely normal skin, dry skin, oily skin, combination skin, and sensitive skin. With the existence of various skin types, it makes some people confused in determining the type of facial skin. This also affects the selection of skincare or facial care according to the indications of each facial skin. Therefore an expert system was created to diagnose facial skin types. An expert system is a man-made system that is used to solve problems like an expert with knowledge from human to computer, although it does not give 100% absolute results, but expert systems are still helpful.
Testing the Budhara Digital Book Application (Borobudur Dalam Cerita) Sri Mulatsih; Raden Arief Nugroho; Valentina Widya Suryaningtyas; Aloysius Soerjowardhana; Ali Muqoddas; Erika Devi Udayanti
Journal of Applied Intelligent System Vol 7, No 3 (2022): Journal of Applied Intelligent System
Publisher : Universitas Dian Nuswantoro and IndoCEISS

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33633/jais.v7i3.7286

Abstract

Borobudur Temple is one of the icons of Indonesian tourism in the world. Unfortunately, tourists, domestic and foreign, do not have knowledge of the importance of the existence of Borobudur Temple as a cultural heritage. Evidence of the lack of knowledge of domestic tourists about the conservation of Borobudur Temple is reflected in the vandalism of thousands of people who left stains on the Borobudur stone. This has the potential to damage the rock layers of the temple. Foreign tourists themselves are also seen as lacking comprehensive knowledge about Borobudur Temple. They know a lot about the history of Borobudur, but do not understand the history of the villages surrounding the supporters of Borobudur Temple. The history of the villages around Borobudur is an important element in supporting Borobudur as a cultural heritage. In addition, efforts are needed to maintain tourist interest in visiting Borobudur Temple and the surrounding villages during the limited number of visitors. For this reason, a way is needed to educate the general public about the history and conservation efforts of the Borobudur Temple and the surrounding villages which have historical and geographical links. Based on these problems, the researchers developed a digital book called Buddhara (Borobudur Dalam Cerita) and has been testing using questionnaire.
Malware Detection Using Decision Tree Algorithm Based on Memory Features Engineering Adhitya Nugraha; Junta Zeniarja
Journal of Applied Intelligent System Vol 7, No 3 (2022): Journal of Applied Intelligent System
Publisher : Universitas Dian Nuswantoro and IndoCEISS

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33633/jais.v7i3.6735

Abstract

Malware is malicious software that can harm, manipulate, steal from victim's device system. Due to the diverse needs of using internet services, security threats are also increasingly difficult to detect. now attackers are starting to develop malware that can change their own signature which is referred to as polymorphism. Therefore, improvements in the traditional approach to detecting the presence of malware are needed to be improved. One of the malware detection approaches, memory-based analysis technique has proven to be a powerful and effective analytical technique in studying malware behavior. In this study, the implementation of a Decision Tree-based classification algorithm was carried out to analyze the data set. Classifier model was created for the purpose of classifying malware based on memory features engineering. The result shows that the Decision Tree machine learning algorithm has been well performed with accuracy to 99.982 %, a false positive rate equal to 0.1% and precision equal to 99.977%
Data Mining Algorithm Testing For SAND Metaverse Forecasting Indri Tri Julianto; Dede Kurniadi; Muhammad Rikza Nashrulloh; Asri Mulyani
Journal of Applied Intelligent System Vol 7, No 3 (2022): Journal of Applied Intelligent System
Publisher : Universitas Dian Nuswantoro and IndoCEISS

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33633/jais.v7i3.7155

Abstract

Metaverse is a technology that allows us to buy virtual land. In the future life in the real world can be duplicated into the Metaverse to increase efficiency, effectiveness, and a world without being limited by space and time. To buy land in the Metaverse, one can be done by using SAND. SAND is a crypto asset from a game called The Sandbox which functions as a transaction tool where in that game we can buy land and build it for various purposes just like we can store our Non-Fungible Tokens there. Metaverse is a digital business that will promise in the future because it offers easy and fast transactions. This study aims to compare the exact algorithm for making predictions about the SAND cryptocurrency used to buy Metaverse land. 7 algorithms are being compared, namely Deep Learning, Linear Regression, Neural Networks, Support Vector Machines, Generalized Linear Models, Gaussian Process, and K-Nearest Neighbors. The research method used is Knowledge Discovery in Databases. The research results show that the Support Vector Machines Algorithm has the most optimal Root Means Square Error value, root_mean_squared_error: 0.022 +/- 0.062 (micro average: 0.062 +/- 0.000). Based on this comparison, the Support Vector Machines Algorithm is suitable for predicting SAND Metaverse prices.
Manhattan, Euclidean And Chebyshev Methods In K-Means Algorithm For Village Status Grouping In Aceh Province Amali Amali; Gatot Tri Pranoto
Journal of Applied Intelligent System Vol 7, No 3 (2022): Journal of Applied Intelligent System
Publisher : Universitas Dian Nuswantoro and IndoCEISS

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33633/jais.v7i3.7037

Abstract

The Ministry of Villages, Development of Disadvantaged Regions and Transmigration (Ministry of Villages PDTT) is a ministry within the Government of Indonesia in charge of developing villages and rural areas, empowering rural communities, accelerating the development of disadvantaged areas, and transmigration. The 2014 Village Potential Data (Podes 2014) is data released by the Central Statistics Agency in collaboration with the Ministry of Villages PDTT in unsupervised form and consists of 6474 villages in the province of Aceh. Podes 2014 data is based on the level of village development (village specific) in Indonesia by using the village as the unit of analysis. Data mining is a method that can be used to group objects in a data into classes that have the same criteria (clustering). One of the algorithms that can be used for the clustering process is the k-means algorithm. Grouping data using k-means is done by calculating the shortest distance from a data point to a centroid point. In this study, a comparison of the distance calculation method on k-means between Manhattan, Euclidean and Chebyshev will be carried out. Tests will be performed using the execution time and the davies boulder index. From the tests that have been carried out, it is found that the number of villages in each cluster is 2,639 developing villages, 1,188 independent villages, 1,182 very underdeveloped villages, 1,266 developed villages and 199 disadvantaged clusters. The Chebyshev distance calculation method has the most efficient accumulation of time compared to Manhattan and Euclidean, while the Euclidean method has the most optimal Davies Index.
Fuzzy Logic for Determination of Community Assistance Using the Tsukamoto Method for Residents of Kasreman Village, Rembang Dahlan Dahlan; Dini Rohmayani; Rachmat Iskandar
Journal of Applied Intelligent System Vol 7, No 3 (2022): Journal of Applied Intelligent System
Publisher : Universitas Dian Nuswantoro and IndoCEISS

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33633/jais.v7i3.7162

Abstract

The obstacle to regional progress and the main cause of social problems is due to the large number of poor people, so there must be a poverty management program by the government, one of which is citizen assistance. The selection process by the local village apparatus is very much needed in the process of determining the recipients of citizen assistance, because the quota for the recipients of citizen assistance is less than that of registrants for citizen assistance. The distribution of aid does not fall to the right party resulting in injustice to other underprivileged families so that it creates several problems, where the method that will be used is Tsukamoto's Fuzzy Logic. In this study, the data used are land area, income of residents, number of dependents of the family. The evaluation method carried out in this study is using a confusion matrix, for one test the level of accuracy produced is 92.74%. Based on the experiment, it can be concluded that the Tsukamoto algorithm is quite accurate in determining citizen assistance to the residents of Kasreman Village, Rembang.
Decision Support System Recommendation Housing Using AHP And Saw Method Palangka Raya City Gatot Tri Pranoto; Ismasari Nawangsih; Edy Widodo
Journal of Applied Intelligent System Vol 7, No 3 (2022): Journal of Applied Intelligent System
Publisher : Universitas Dian Nuswantoro and IndoCEISS

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33633/jais.v7i3.7038

Abstract

Palangka Raya City, as one of the provincial capitals in Indonesia, which has an area of around 2,400,000 km2, is a strategic city as a service and distribution hub for the industrial, trade, government and education sectors. Regional Policy of the city government with the existence of a development plan in the City of Palangka Raya as an implementation of the city space with all the disadvantages of its designation resulting in a distribution pattern of urban land types which in fact is not evenly distributed throughout the city. This research was conducted based on the results of observations made in several Marketing Agents in the Palangkaraya Region, which included 5 districts where the survey results obtained several marketing agents for KPR housing with the aim of facilitating the purchase of KPR housing. The purpose of this study is to design a decision support system that is used to support the decision to purchase housing loans in the Palangkaraya area. Based on the research that has been done, it is expected that the results of the purchase decision support system for the KPR recommendation with the best value can be a recommendation for the purchase. This system is designed with the AHP and SAW methods to help prospective residents to determine the house based on the desired criteria.
Classification and Regression Trees (CART) Algorithm for Employee Selection Aulia Rahmawati; Rizal Muhammad Affandi; Dea Debora Aprillia; Daffa Maulana; Zudha Pratama; Moch. Sjamsul Hidajat
Journal of Applied Intelligent System Vol 7, No 3 (2022): Journal of Applied Intelligent System
Publisher : Universitas Dian Nuswantoro and IndoCEISS

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33633/jais.v7i3.7201

Abstract

Recruitment is the main key in an effort to improve the quality of human resources in a company. Good or bad employees greatly affect the quality of the company. Therefore, it is necessary to be thorough and take a long time in screening applicants in order to get competent, professional and as expected prospective employees. The absence of professional staff to conduct employee selection is the background of this research. So the researcher uses the CART algorithm for the classification of employee recruitment, so it is hoped that it can help companies in conducting employee selection. The dataset was obtained from the selection of freelance daily workers at the Pati Regency Civil Service Police Unit in 2018, totaling 290 prospective employees. Based on calculations on 5-fold cross validation, the resulting accuracy is 98.27%, precision is 99.13% and recall is 96.88%.
Naive Bayes Performance in Analysis of Public Opinion Sentiment Against COVID-19 Ayu Hendrati Rahayu; Ari Sudrajat
Journal of Applied Intelligent System Vol 7, No 3 (2022): Journal of Applied Intelligent System
Publisher : Universitas Dian Nuswantoro and IndoCEISS

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33633/jais.v7i3.7134

Abstract

The huge impact caused by the COVID-19 pandemic has made many people express their opinions on Twitter social media. There are various responses given by the community that are negative and positive. The dataset comes from kaggle with more than 750 tweets of data. Classification designed by the Naive Bayes method. Implementation through preprocessing, case folding, tokenizing, stopword removal, TF-IDF, and cross validation has been able to produce quite high accuracy. After classification, validation will be carried out with Cross Fold Validation. The best value is on cv5 where accuracy = 0.847, precision = 0.855, recall = 0.83, and f1 score = 0.842.

Page 1 of 1 | Total Record : 10