IJDASEA (International Journal of Data Science, Engineering, and Analytics)

Model Selection for Forecasting Rainfall Dataset

Amri Muhaimin — 2024-11-19

The objective of this research is to obtain the best method for forecast- ing rainfall in the Wonorejo reservoir in Surabaya. Time series and causal ap- proaches using statistical methods and machine learning will be compared to forecast rainfall. Time series regression (TSR), autoregressive integrated moving average (ARIMA), linear regression (LR), and transfer function (TF) are used as a statistical method. Feedforward neural network (FFNN) and deep feed-for- ward neural network (DFFNN) is used as a machine learning method. Statistical methods are used to capture linear patterns, whereas the machine learning method is used to capture nonlinear patterns. Data about hourly rainfall in the Wonorejo reservoir is used as a case study. The data has a seasonal pattern, i.e. monthly seasonality. Based on the cross-validation and information criteria, the results showed that DFFNN using the time series approach has a more accurate forecast than other methods. In general, machine learning methods have better accuracy than statistical methods. Furthermore, additional information is ob- tained, through this research the parameter that best to make a neural network model is known. Moreover, these results are also not in line with the results of M3 and M4 competition, i.e. more complex methods do not necessarily produce better forecasts than simpler methods.

Geometric Brownian Motion and Value at Risk For Anal-ysis Stock Price Of Bumi Serpong Damai Ltd

Di Asih I Maruddani — 2024-11-19

Investment is one of the activities that last actually attractive to the people of Indonesia. One of the most widely traded financial assets in the capital market is stocks. Stock prices frequently experience challenges to predict changes, so they can increase or decrease at any time. One method that can be applied to predict stock prices is GBM. Then, the risk can be measured using the VaR risk measure. The GBM model is determined to be accurate in predicting the stock price of BSDE.JK, with a MAPE value of 5.17%. By using VaR-HS and VaR CFE, the prediction of risk of loss at the 95% confidence level for the period 06/07/21 is -0.0597 and -0.0623.

Diagnosis of Diabetes Using Naïve Bayes Classifier Method

Tasya Ardhian Nisaa — 2024-11-19

Not a few people suffer from diabetes, diabetes is usually caused by genetic inheritance from parents and grandparents. Not only from heredity but many criteria or characteristics can determine a person has diabetes. This research was conducted by looking for a dataset on Kaggle that contains criteria for someone diagnosed or undiagnosed with diabetes such as age, gender, weakness, polyuria, polydipsia, and others. Furthermore, from these criteria, predictions are calculated using the Naive Bayes classification method where this method is one of the data mining techniques. This prediction calculation uses the Python programming language. From these criteria, each criterion is grouped with similarities and the results of the program that have been made can diagnose someone with diabetes. The prediction calculations that have been carried out have resulted in 90% accuracy, 93% precision, 89% recall, 92% specificity, and 91% F1-Score.

Implementation of Data Mining in Shopping Cart Analysis using the Apriori Algorithm

Susy Rahmawati — 2024-11-19

Market basket analysis is one of the techniques of knowledge mining used in a broad dataset or database to find a collection of items that are interwoven. Generally used in a sale, the most relevant shopping cart data is used. This methodology has been widely applied in different multinational or foreign industries and is very useful in consumer buying preferences. Technology advances change business trends dramatically, shifting customer demands require increased surgical accuracy of business. In this research, the writer wants to analyze the shopping cart using a priori algorithm, with a dataset from the Kaggle web. Using anaconda software features with the Python programming language which is expected to create knowledge overwriting consumer buying patterns. In conclusion, this pattern can be used to support an industry in managing its company activities.

Selection of Notification Based on Priority Scale with Fuzzy Algorithm

Mohammad Faisal Riftiarrasyid — 2024-11-19

Notification is one method that works as a marker that there is information waiting to be read. But along with the times, notifications are increasingly filled with information that is considered less important for device users. So there needs to be a breakthrough to overcome this. This study aims to design a system that can help users to sort out notifications that are considered important and not. It is proven that the system can sort notifications based on the given metrics.