Desti Fitriati, Desti
Unknown Affiliation

Published : 14 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 5 Documents
Search
Journal : Jurnal Riset Informatika

Sales Analysis Using Apriori Algorithm Roja' Putri Cintani; Fitriati, Desti
Jurnal Riset Informatika Vol. 7 No. 4 (2025): September 2025
Publisher : Kresnamedia Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34288/jri.v7i4.351

Abstract

PT JR Pangan Semesta is a company that produces fast food in the form of Donuts and Sweet Bread under the Deroti brand. The sales and promotion methods that have been carried out have weaknesses because the company has difficulty ensuring the right amount of bread production, so there is often excess or lack of stock. In addition, the promotional strategy used has not included the concept of bundling, so the maximum promotional potential has not been fully explored. To overcome these problems, the use of data mining methods is proposed, one of which is the Apriori Association Rule algorithm. Apriori algorithm is used to find consistent sales patterns and find strong product relationships by analyzing sales transaction data. In this study, sales patterns were analyzed at PT JR Pangan Semesta with a minimum support value of 16% and a minimum confidence value of 60%. The analysis results show that there are three products that are often purchased together by consumers, namely Fried Bread, Deroti Donuts, and Eco Donuts. The three products form one valid association rule, so that the rule can be used as a reference for developing efficient production methods for bread and donuts and implementing sales strategies in the form of bundling products to maximize profits.
Comparison of Breast Cancer Classification Using Decision Tree ID3 and K-Nearest Neighbors Algorithm to Predict the Best Performance of Algorithm Daldiri, Zyhan Faradilla; Fitriati, Desti
Jurnal Riset Informatika Vol. 5 No. 2 (2023): March 2023
Publisher : Kresnamedia Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34288/jri.v5i2.206

Abstract

One of the leading causes of death is cancer. The most common cancer in women is breast cancer. Breast cancer (Carcinoma mammae) is a malignant neoplasm originating from the parenchyma. Breast cancer ranks first in terms of the highest number of cancers in Indonesia and is among the first contributors to cancer deaths. Globocan data in 2020 shows that the number of new breast cancer cases reached 68,858 (16.6%) of the total 396,914 new cancer cases in Indonesia. Meanwhile, deaths reached more than 22 thousand cases (Romkom, 2022). This death rate is increasing due to insufficient information about breast cancer’s early symptoms and dangers. Of this lack of information, a system is needed that can provide information about breast cancer, such as early diagnosis. Several parameters and classification data mining techniques can predict which patients will develop breast cancer and which do not. In this study, a comparison of the classification of breast cancer using the Decision Tree ID3 algorithm and the K-Nearest Neighbors algorithm will be carried out. Attribute data consists of Menopause, Tumor-Size, Node-Caps, Deg-Malig, Breast-Squad, and Irradiant. The main objective of this study is to improve classification performance in breast cancer diagnosis by applying feature selection to several classification algorithms. The Decision Tree ID3 algorithm has an accuracy rate of 93.333%, and the K-Nearest Neighbors algorithm has an accuracy rate of 76.6667%.
IMPLEMENTATION OF GENETIC ALGORITHM IN THE CURRENT SCHEDULING SYSTEM : Ulum , Pateh; Fitriati, Desti
Jurnal Riset Informatika Vol. 3 No. 2 (2021): March 2021 Edition
Publisher : Kresnamedia Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34288/jri.v3i2.57

Abstract

Scheduling courses is a routine job in academic activities at a college. In its implementation, the scheduling process is not easy to do because many factors need to be considered, several factors that are considered, such as the willingness of lecturers to teach, the availability of classrooms. Besides that, it is also necessary to pay attention to the number of classes in each subject. Course scheduling is a combination of courses, days, time, lecture space, and consideration of lecturers' willingness to teach. To solve the course scheduling problem, a system that can handle the scheduling process is needed. The method that can be used to solve this problem is to use the Genetic Algorithm approach. The genetic algorithm is a scheduling algorithm that can combine lecture time and space automatically by applying a natural or gene selection system. Based on the research that has been done, the genetic algorithm can solve scheduling problems quickly, which only takes 15 seconds for 78 classes and uses as many as 16 chromosomes. Also, the fitness value of all chromosomes is 0, this means that the scheduling results obtained are optimal.
SMART SYSTEM FOR AUTOMATIC CROP AND RECOGNITION PLAT NUMBER Fitriati, Desti; Pasha, Nira Ravika; Hariyanto, Bambang; Murtako, Amir; Nursari, Sri Rezeki Candra
Jurnal Riset Informatika Vol. 3 No. 2 (2021): March 2021 Edition
Publisher : Kresnamedia Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34288/jri.v3i2.60

Abstract

Based on data from the Central Statistics Agency in 2018, it was written that the number of motorbikes for the Indonesian region was 120.10 million or 82% and for cars 26.75 million or around 18% of the total population. With the increasing population of motorized vehicle users, it will result in an increase in problems that occur in traffic violations and also the technology security system in the parking system. Most of the existing parking systems still require parking attendants. In addition, the existing system only discusses the opening and closing of bars and providing information on parking lots. Although the existing system already uses artificial intelligence to read plate numbers, the officers are still matching it. Of course this is not effective and efficient because the use of artificial intelligence is not purely done by the system. To overcome this, the solution given in this study is to create a parking system that can read plate numbers automatically and store vehicle entry data directly into the database. The system created can also open and close the door latch automatically. The template matching image processing technique was chosen to solve this problem. Based on the experimental results, the system can recognize plate numbers with an accuracy of 83%. For further research, it is necessary to introduce vehicle ownership and provide parking information so that the parking system becomes more perfect.
Implementation of Data Mining To Determine the Association Between Body Category Factors Based on Body Mass Index Fitriati, Desti; Amiga, Bima Putra
Jurnal Riset Informatika Vol. 2 No. 4 (2020): Period September 2020
Publisher : Kresnamedia Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34288/jri.v2i4.128

Abstract

The development of the increasing flow of globalization in the field of science and technology as well as increased income has resulted in reduced physical activity of the community which results in diverging diet and physical activity which makes a person not pay attention to his body shape. This method of calculating the Body Mass Index can be used to determine a person's body shape. Several factors can affect the value of the Body Mass Index, including individual factors, consumption patterns, and lack of physical activity which leads to a sedentary lifestyle. These factors are made into 69 itemsets which will be used as the basis for questions in the questionnaire to collect a dataset that will later be processed using the FP Growth algorithm and looking for association rules that have the highest support x confidence value. From the 490 calculation data, the results are categorized into 10, each of which is Men with a Very Thin BMI with the highest support x confidence value of 39.56%, Men with a Thin BMI of 55.90%, Men with a Normal BMI of 70%, men with a fat BMI of 49.23%, men with an obese BMI of 41.34%, women with a very thin BMI of 41.37%, women with a thin BMI of 37.21%, Normal BMI is 68.83%, women with obese BMI are 41.65%, and women with obese BMI are 42.91%.